Search is not available for this dataset
article stringlengths 4.36k 149k | summary stringlengths 32 3.35k | section_headings listlengths 1 91 | keywords listlengths 0 141 | year stringclasses 13
values | title stringlengths 20 281 |
|---|---|---|---|---|---|
Cancer metabolism has received renewed interest as a potential target for cancer therapy . In this study , we use a multi-scale modeling approach to interrogate the implications of three metabolic scenarios of potential clinical relevance: the Warburg effect , the reverse Warburg effect and glutamine addiction . At the intracellular level , we construct a network of central metabolism and perform flux balance analysis ( FBA ) to estimate metabolic fluxes; at the cellular level , we exploit this metabolic network to calculate parameters for a coarse-grained description of cellular growth kinetics; and at the multicellular level , we incorporate these kinetic schemes into the cellular automata of an agent-based model ( ABM ) , iDynoMiCS . This ABM evaluates the reaction-diffusion of the metabolites , cellular division and motion over a simulation domain . Our multi-scale simulations suggest that the Warburg effect provides a growth advantage to the tumor cells under resource limitation . However , we identify a non-monotonic dependence of growth rate on the strength of glycolytic pathway . On the other hand , the reverse Warburg scenario provides an initial growth advantage in tumors that originate deeper in the tissue . The metabolic profile of stromal cells considered in this scenario allows more oxygen to reach the tumor cells in the deeper tissue and thus promotes tumor growth at earlier stages . Lastly , we suggest that glutamine addiction does not confer a selective advantage to tumor growth with glutamine acting as a carbon source in the tricarboxylic acid ( TCA ) cycle , any advantage of glutamine uptake must come through other pathways not included in our model ( e . g . , as a nitrogen donor ) . Our analysis illustrates the importance of accounting explicitly for spatial and temporal evolution of tumor microenvironment in the interpretation of metabolic scenarios and hence provides a basis for further studies , including evaluation of specific therapeutic strategies that target metabolism .
Cancer remains one of the leading causes of death worldwide . A central challenge in understanding and treating cancer comes from its multi-scale nature , with interacting defects at the molecular , cellular and tissue scales . Specifically , the molecular profile at the intracellular level , behavior at the single-cell level and the interactions between tumor cells and the surrounding tissues all influence tumor progression and complicate extrapolation from molecular and cellular properties to tumor behavior [1–3] . Understanding the multi-scale responses of cancer to microenvironmental stress could provide important new insights into tumor progression and aid the development of new therapeutic strategies [2] . Therefore , cancer must be studied and treated as a cellular ecology made up of individual cells and their microenvironment . This ecological view should account for the competition and cooperation of different molecular and cellular players , and for both the physical and biological characteristics of the environment in which tumor evolves . Such perspectives complement studies of the genetic drivers of tumor and potentially provide new bases for treating this disease [4] . Central to an ecological perspective of tumors is metabolism , the biochemical process by which cells derive energy and biomass from the nutrients available in their environment while excreting products of metabolism back to the environment . This exchange of metabolites impacts the distribution of resource in the environment and sets constraints on the availability of resources to individual cells [5] . Therefore , metabolism couples the behavior of individual cells to the characteristics–spatial-temporal organization and phenotypic make-up–of the full population . Recently , cancer metabolism has drawn renewed attention in the field of cancer biology [4 , 6] . Following the early observations of the unique tissue-scale metabolic profile of tumors made by Otto Warburg in the 1920s , discoveries of oncogenes and molecular cues in tumor-associated metabolic alterations have renewed the hope for therapeutic routes that target cancer metabolism [7] . In his seminal work , Warburg noted the distinct metabolic profile of tumor cells with high glycolytic rate and lactate production in the presence of oxygen . This so-called Warburg effect or aerobic glycolysis has been widely observed in different types of tumor cells ( Fig 1A , ① ) [8] . This original observation by Warburg led him to hypothesize that aerobic glycolysis is caused by impaired respiration; in turn , this defect results in cancer [9] . It is now well accepted that this hypothesis is incorrect as most tumor cells retain functional mitochondria [10 , 11]; we still lack a full understanding of the origin and consequences of the Warburg effect . More recently , other hypotheses have been proposed in the field of cancer metabolism such as the reverse Warburg effect ( Fig 1A , ② ) and glutamine addiction ( Fig 1A , ③ ) [3 , 12–16] . Despite support of these three hypotheses from various experimental studies , significant uncertainty remains with respect to their definitions , their origin , and their impact on tumor progression and therapeutic interventions . Unraveling these fundamental questions could open a clearer path to targeting cancer metabolism as a therapeutic strategy . In the past few years , studies of cancer metabolism have begun to elucidate how the metabolic alterations in tumor cells can influence tumor progression [9 , 17–21] . A definitive characteristic of tumor cells is uncontrolled proliferation . Compared to healthy cells that remain quiescent in most of their life cycle , tumor cells proliferate rapidly , accompanied by high rates of metabolic uptake . This metabolic profile of tumor cells leads to significant depletion of metabolites in the local microenvironment , resulting in resource limitations . Additionally , byproducts and waste products produced by the metabolism of tumor cells can potentially hinder the growth of neighboring cells or act as sources of alternative metabolic substrates [16 , 22 , 23] . Although studies have made efforts to capture these experimental observations mathematically [20 , 24–29] , we are unaware of computational studies that test the implications of these hypotheses with respect to metabolic behaviors at the individual cell level , intercellular interactions mediated by shared metabolic environment , and the collective behavior that together define fitness and growth potential of the tumor . Recent computational work has made progress toward capturing the multi-scale complexity of cancer . These studies investigated the effect of tumor microenvironmental factors , specifically molecular cues and metabolites , on tumor population dynamics and provided insights into the cooperative behaviors of tumor subpopulations [30–34] . Similar intraspecies competition or cooperation are often observed in microbial organisms and heavily studied from a population ecology perspective [35–37] . Theories and modeling tools are better developed in the microbial field due to the relatively convenient validation from experiments [38–40] . In this study , we take a multi-scale modeling approach to describe the intracellular , cellular , and multicellular behaviors of cells within a tumor ( Fig 1 ) . With this framework , we investigate the following hypotheses: Warburg Effect/Aerobic glycolysis ( ① ) , Reverse Warburg ( ② ) , and Glutamine Addiction ( ③ ) . We begin by translating hypotheses from experimental studies into constraints and objectives within the FBA ( Fig 1A ) . We proceed to use FBA to obtain the yield coefficients ( Y = maximum growth rate/flux of metabolite ) for use in Monod-like kinetics of cellular growth at the individual cell level ( Fig 1B ) . Finally , we simulate the growth dynamics of these cells at the multicellular scale to elucidate the implications of these metabolic scenarios ( Fig 1C ) . We address the impact of the metabolic phenotypes implied by current hypotheses on the growth dynamics of tumor cells in the resource-limited microenvironments that emerge after tumor initiation . This modeling framework opens a route to explore tissue-scale tumor dynamics with explicit account taken for these metabolic scenarios in an efficient manner .
Fig 1 illustrates , schematically , the multi-scale approaches we use . At the intracellular scale , we use Flux Balance Analysis ( FBA ) to construct a network that captures the central metabolism of mammalian cells ( Fig 1A ) . In Fig 1A , the arrows represent fluxes of species within a reduced representation of cell metabolism and cell growth; the detailed network used in FBA is presented in S1 Fig . Key steps associated with three hypotheses are labeled: Warburg effect ( ① ) is distinguished by high glycolytic flux and lactate production; reverse Warburg ( ② ) is distinguished by the uptake of lactate; and glutamine addiction ( ③ ) is distinguished by uptake of glutamine as a carbon source to feed TCA cycle . We build the biomass template reaction ( S1 Fig ) based on major precursors for biomass synthesis by reducing Shlomi and coworker’s genome scale biomass template [20] . We impose a cellular maintenance reaction with a baseline rate to define the required minimum metabolism of cells ( see Methods ) . We modify constraints and objective functions within the FBA network to define the characteristics of the different hypotheses ( labeled in Fig 1A ) . We estimate parameters based on literature ( see S1 Table ) . We acknowledge that the altered metabolic phenotype of tumor cells may be due to prior genetic events that occurred in the cell , such as loss of tumor suppressors ( e . g . , p53 ) [41] . However , we only consider the metabolic phenotypes of the cells at fixed genetic profiles here since we focus on impact of metabolic profiles on tumor growth over time scales ( days ) that are short relative to those required for the emergence and accumulation of genetic alterations in the cells ( months or years ) . At the cellular scale ( Fig 1B ) , we use the imposed maximum growth rates ( μm , n [hr-1] ) and the metabolic uptake and production rates of the metabolites ( qi/n [g/g-DW-hr] ) obtained from FBA to determine yield coefficients ( ( Yi/n [g-DW/g] ) for each metabolite ( i ) and corresponding metabolic phenotype ( n ) : Yi/n=−μm , nqi/n ( 1 ) These yield coefficients link our intracellular treatment of metabolism by FBA and our cellular and multicellular treatment of resource utilization and growth . We model biomass ( Xm [g] ) growth of each cell type as a Monod-like process parameterized by maximum growth rate for each metabolic phenotype , μm , n [hr-1] and a Monod function of metabolite concentrations , fn ( {Cj} ) Monod: dXmdt=Xm⋅∑nμm , n⋅fn ( {Cj} ) Monod ( 2 ) We provide detailed discussions of the Monod functions in the next Section . We used the same value of maximum growth rate for each phenotype of each cell at both the FBA ( Fig 1A ) and cell-scale ( Fig 1B ) . We report parameter values in S1 Table . Additionally , we use a threshold in cell diameter to define the doubling of the cell by linking biomass growth to the volume ( Vm [L] ) expansion of the cell at a fixed dry mass density ( ρ [g-DW/L] ) : dVmdt=1ρdXmdt ( 3 ) To bridge the treatment of metabolic processes at the cellular and multicellular scales , we solve steady state species balances for each explicit metabolite at each time step within iDynoMiCS [39]: ∂Ci∂t=Di∇2Ci+∑mρ∑nqi/n⋅fn ( {Cj} ) Monod ( 4 ) where Ci [g/L] is the concentration of ith metabolite , Di [m2/day] is the diffusion coefficient of ith metabolite . Here , the species balances can be safely treated as being at steady state because the time step in our simulation ( 1 hour ) is selected to resolve cell growth and is long compared to typical transients in metabolism [39] . We integrate Eqs 1–4 into iDynoMiCS to track the growth of individual cells within a continuum matrix occupied by other cells in which metabolites diffuse ( term 1 in Eq 4 ) . The concentration of metabolites at the multicellular scale governs the cellular biomass growth ( Eq 2 ) and the biomass kinetics in turn influences the concentration profile of metabolites ( term 2 in Eq 4 ) , and subsequently the growth kinetics of the surrounding cells . Cells are treated as hard spheres [38] . This spatial-temporal interaction between the cells and the microenvironment is a dynamic process that changes at each time step within iDynoMiCS . We simulate growth in both radial and axial geometries in iDynoMiCS ( Fig 1C ) : 1 ) Radial , two-dimensional growth ( Fig 1C i ) ) –tumor cells ( red ) grow radially out from an initial cluster of cells with metabolites supplied at the edge of the cell mass such that radial gradients of concentration emerge ( color map ) . As the tumor grows , concentration gradients of metabolites become significant , making the tumor growth a diffusion-limited process that can result in different growth dynamics as well as distinct spatial distribution of cell subpopulations . 2 ) Axial , one-dimensional growth ( Fig 1C ii ) ) –layers of tumor cells ( red ) and stromal cells ( blue ) are initiated near a blood vessel that supplies metabolites ( from the top ) , such as glucose and oxygen in the blood stream . Growth pushes cells deeper into the tissue , away from the vessel , such that strong gradients of metabolite can again occur . The radial simulations ( Fig 1C i ) ) provide a qualitative understanding of the growth dynamics in different metabolic scenarios; axial simulations ( Fig 1C ii ) ) allow us to further quantify the observed dynamics . In both cases , we initiate tumor cell clones ( same Monod parameters ) surrounded by a varying number of layers of stromal cells ( defined by distinct metabolic and growth parameters–see Fig 2 and Table 1 ) . These arrangements capture tumor growth with initiation occurring at different distances from local vascular structure and thus at different levels of metabolic stress . We proceed to track growth as a function of depth of initiation and metabolic phenotype . We perform 11 replicates , with randomly seeded initial positions of tumor and stromal cells within their compartments; all other parameters in the simulation were kept the same across these replicates for each metabolic scenario . Additionally , we kept these random initial seeding positions of cells the same across simulations of the three metabolic scenarios to eliminate any effect that comes from the difference in the initial seeding when comparing the scenarios . As we are interested in initial stages of avascular growth , we do not account for later stage processes such as angiogenesis . Further , we do not account for cell death explicitly in our simulations; tumor cells in zones with severely depleted metabolites remain quiescent based on the Monod-like growth kinetics . When evaluating total tumor size , this assumption is equivalent to counting dead cell mass within the necrotic core as part of the tumor; this definition is consistent with that of previous studies [42–46] . With the aim of providing intuition on the outcomes of simulations and characteristic physical parameters , we also calculate the Krogh length , shown schematically in Fig 1C iii ) . Here , we define the Krogh length of a metabolite as the length at which the concentration of metabolite becomes zero given the uptake of the metabolite with zeroth order growth kinetics for the cell phenotype in the region ( see Methods ) . While this is an extremely simple model that couples zeroth order kinetics with a continuum description of reaction and diffusion in the tissue , we will show that it provides insights into the characteristics by which reaction and diffusion govern the growth of tumors . With this multi-scale computational framework , we study the tumor population dynamics in a spatial-temporal manner and investigate the consequences of different hypotheses in cancer metabolism from a population ecology perspective . This perspective examines the impacts of phenotypic composition , spatial structure and reaction-diffusion on tumor growth . Before we further specify the hypotheses depicted in Fig 1 individually , we define the metabolic phenotypes of the cell types implicated in these hypotheses based on observations in the literature . We integrate our interpretations of these metabolic mechanisms into FBA to obtain the uptake and production rates of metabolites ( see Table 1 ) : In our approach , we assume each cell type ( e . g . , healthy stromal cell ) can adopt more than one metabolic phenotype ( e . g . , aerobic under normoxic conditions and anaerobic under hypoxic conditions ) . These different metabolic phenotypes are implemented as objective functions and constraints in FBA and in turn , result in different flux distributions ( Fig 2 , coded by color ) . We then obtain yield coefficients ( Yi/n ) for the ith metabolite in the nth metabolic phenotype of cells by linking maximum growth rate ( μm , n ) of the mth cell type to the uptake and production rates ( qi/n ) ( Eq 1 ) ; the Yi/n serve as measures of the efficiency with which the metabolites generate biomass: the bigger the value of Yi/n is , the more efficiently the nth metabolic phenotype utilizes the ith metabolite to grow . Fig 2 summarizes predictions from FBA for the metabolic profiles of these cell types under distinct metabolic phenotypes . The metabolic switch from normoxia to hypoxia or to hypoglycemia leads to drastic changes in metabolic fluxes; the values in box represent fluxes of the specific metabolites when they display different metabolic profiles , coded by color ( see caption ) . These flux distributions in turn lead to different uses of metabolites as reflected in yield coefficients ( presented in Table 1 ) . We present detailed description of each phenotype of the cells in the following subsections .
To gain a qualitative understanding of the impact of the various metabolic scenarios on tumor growth in a diffusion-limited microenvironment , we first ran simulations in an unconstrained 2-D domain , as shown in Fig 1C i ) ; metabolites were delivered through a diffusive boundary layer of fixed thickness that surrounds the growing tissue ( see Methods ) . Fig 3 presents the form of the tumors at initiation ( t = 0 ) and after 100 days of growth for Warburg tumor cells ( WN = 2 ) with healthy stromal cells ( top row ) , Reverse Warburg cells with hijacked stromal cells ( middle row ) , and glutamine-addicted tumor cells with healthy stromal cells ( bottom row ) . The three columns are for initial seeding of tumor cells beneath 1 , 3 , and 5 layers of stromal cells , as indicated in the images of the initial configuration of the cells ( t = 0 ) . As the number of layers of stromal cells increases , the growth of tumor cells becomes compromised due to the reduced access to the metabolites . By hindering diffusion and consuming oxygen and glucose , the stromal cells decrease the accessibility of these metabolites to the tumor cells . In all cases , we note that the proliferation of the tumor cells led to their breaking through the layers of stromal cells; for the cases with significant growth , the stromal cells became engulfed within the tumor , as is frequently observed in actual tumors [61] . We also note the emergence of irregular front of the tumor in the scenario of Reverse Warburg effect . We suspect that this irregularity arises from growth instability due to the moderate availability of metabolites at the growth front [35] . We note qualitatively different effects of the addition of layers of stromal cells on tumor growth for the different scenarios: with 5 layers of stromal cells , the growths of both Warburg and Glutamine-addicted tumor cells were strongly delayed , whereas the impact on the growth of Reverse Warburg cells was modest . These observations motivate a deeper investigation of the mechanisms that control response to metabolic stress in these scenarios . We proceeded to dissect the metabolic scenarios further with simulations in a confined geometry in which solute ( i . e . , metabolite ) diffusion and tissue expansion were constrained along a single direction , as shown in Fig 1C ii ) . This axial scenario approximates the local environment adjacent to a blood vessel ( upper boundary ) . Fig 4 presents an overview of the growth behavior in this geometry . For this overview , we simulated the Warburg scenario , with Warburg tumor cells ( WN = 2 ) and healthy stromal cells ( also see Fig 2 ) . Fig 4A shows the snapshots of tumor growth and the corresponding concentration fields of oxygen and glucose at various time-points for tumors initiated beneath 1 , 3 , and 5 layers of stromal cells . In the colormaps of the concentration fields , we see that when the tumor initiated closer to the source ( top row with 1 layer of stromal cells ) , the Warburg tumor cells had access to ample oxygen and glucose to fuel their growth at early time ( t = 0 , empty circle ) ; at late time ( t = 25 days , filled circle ) , significant depletion of both oxygen and glucose occurred , but the uppermost layer of tumor cells still benefited from high metabolite concentrations to grow . However , when the tumor initiated farther away from the source ( middle and bottom rows ) , the diffusion limitations and consumption by the stromal cells limited the metabolites available to the tumor cells , even at early times ( empty diamond , empty square ) . This limitation persisted until the tumor cells broke through the stromal layer and gained access to higher concentrations of metabolites ( filled diamond , filled square ) . Fig 4B presents the trajectories of tumor growth from 11 simulation runs in each case shown in Fig 4A . We first note that for all initial conditions , the growth appears to proceed through two phases , starting with slower growth that then transitions to faster growth; these two regimes are most evident for 3 and 5 layers of stromal cells . By observing the cellular configurations in the simulations ( see S1–S3 Movies ) , we identify that the transition occurs when the tumor cells break through the layers of stromal cells and gain access to high concentrations of metabolites . When the tumor cells started to grow , the reaction-diffusion in the intact layers of stromal cells limited the supply of metabolites to the tumor cells . Under such conditions , the growth of tumor cells was significantly compromised due to the lack of oxygen ( note the more severe depletion of oxygen relative to glucose in Fig 4A , also see Eq 7 in Model ) ; the microtumor was nearly quiescent . Once this slow growth led to the penetration of one or more tumor cells through the layers of stromal cells , those tumor cells transitioned toward their aerobic growth regime ( term 1 in Eq 7 ) and quickly overwhelmed the stroma . Interestingly , the growth rates after breakthrough were constant ( the growth curves are linear in time ) and independent of initial conditions ( all late-time slopes are the same in Fig 4B ) . This constant growth rate is distinct from the exponential growth that one would expect resulting from saturating Monod-like growth kinetics ( Eq 5–10 in Model ) . This observation illustrates an important consequence of a diffusion limited microenvironment . We will comment further on the origin of this constant rate below . In the case of 1 layer of stromal cells ( black curves ) , the growth transitions rapidly ( within the first days ) to a high , constant rate . Furthermore , the trajectories of all the random initial seeding conditions are very similar . For 3 and 5 layers of stromal cells ( blue and red curves ) , the first , slow phase lasts longer because the tumor cells experienced more severe limitations in their initial configurations . Additionally , in these cases , the trajectories of different initial conditions diverge strongly from one another due to the differences in the moment of transition from slow to fast growth . This observation reflects the fact that the time for tumor cells to break through the stroma is sensitive to small differences in the initial configuration of cells . We now proceed to use axial simulations like those in Figs 1C ii ) and 4 to investigate the growth dynamics in each of the three metabolic scenarios .
Within our scope of study of the Warburg effect through the multi-scale modeling approach ( Figs 4 and 5 ) , we confirmed a common hypothesis that Warburg effect impacts tumor cell fitness in metabolically limited microenvironments [64] . Interestingly , our predictions suggest that there may exist an optimal level of Warburg effect ( reflected by the ratio of pyruvate fluxes to lactate and to the TCA cycle; the Warburg Number ) for tumor cells to adopt depending on the details of the metabolic microenvironment in which the tumor cells initiate . This observation may help explain the experimentally observed phenotypic heterogeneity in cancer metabolism [65 , 66] . Such adaptation could occur via modification of the fluxes of pyruvate , for example with changes in enzymatic rates along either the TCA cycle or glycolytic pathways . From an ecological perspective , our predictions indicate that Warburg effect may provide a basis for adaptation of tumor cells to different environmental metabolic stresses [67] . For the reverse Warburg effect scenario ( Fig 6 ) , we provide the first mathematical description of the multi-cellular metabolic interactions proposed by Sotgia et al . [55] . We used our framework to explore the intracellular and multicellular consequences of reverse Warburg effect due to the interaction between glycolytic stromal cells ( hijacked stromal cells ) and lactate-consuming tumor cells ( reverse Warburg tumor cells ) . We predict that the hijacked stromal cells have higher yields on oxygen than healthy stromal cells ( Fig 2B i ) and Table 1 ) . This information confirmed the intuitive proposal of Sotgia et al . that the hijacked stromal cells assist in the growth of tumors that initiate deep within the stroma by allowing more oxygen to penetrate into the tumor compartment ( Fig 6C–6E ) . We further note that , due to the adaptive character of reverse Warburg tumor cells , they are not sensitive to local lactate concentration in aerobic growth regimes ( term 1 and 2 on the right-hand side of Eq 9 can be combined ) ; this characteristic means their aerobic growth remains limited by oxygen and glucose only . Additionally , due to the utilization of lactate as carbon source in energy production in these tumor cells , their yields on oxygen is lower compared to tumor cells in the scenario of Warburg effect ( requiring more oxygen for the same mole of carbon consumed ) . Therefore , the reverse Warburg effect leads to slower growth in favorable metabolic microenvironment ( i . e . , abundant source of metabolites available ) . However , when tumors initiate in microenvironments where resources are significantly reduced , the host-parasite relationship implied by the reverse Warburg effect ( via cooperative utilization of oxygen between hijacked stromal cells and tumor cells ) can provide growth advantage to tumors . Given that such growth advantage depends on the detailed structure of the metabolic microenvironment , we suggest that one must use a multi-scale framework like the one presented here to investigate the implications of these metabolic scenarios . In the exploration of glutamine addiction ( Fig 7 ) , we defined the metabolic phenotype by hypothesizing that glutamine addiction coexists with Warburg effect . This hypothesis led us to propose a coupled contribution to biomass synthesis of tumor cells from glucose and glutamine as joint carbon sources . Specifically , we aimed to explore the role of glutamine in anaplerosis ( as a carbon source to replenish the TCA cycle ) . We demonstrated with FBA that under our interpretation , glutamine addiction led to an increase uptake of oxygen ( i . e . , lower yield on oxygen ) in glutamine-addicted tumor cells to maintain their redox balance and to meet the energy demand; this lower yield on oxygen represents a cost of using glutamine in the TCA cycle . We see the impact of this lower yield on oxygen in the reduced growth rate of glutamine-addicted tumor cells relative to Warburg tumor cells . We thereby conclude that glutamine addiction via the process of anaplerosis does not confer an advantage to the overall tumor growth primarily due to the strong dependence on oxygen . We argue that glutamine is not an effective alternative carbon source because tumor cells remain limited by glucose and oxygen . Our study constrains future considerations of the roles of glutamine addiction in tumor growth by clearly demonstrating that the anaplerotic pathway cannot , alone , provide a growth advantage to tumors . With our focus on the anaplerotic role of glutamine using a simplified metabolic network , we did not account for other roles of glutamine in cellular demand explicitly [68 , 69] . For example , glutamine is known to be an important nitrogen source in nucleic acids and amino acids synthesis [57 , 70 , 71] . Additionally , glutamine contributes to the pool of metabolites that maintains NADPH/NADP+ balance [69 , 72] and to produce glutathione as an antioxidant to help the cell resist oxidative stress during rapid metabolism [70 , 72] . We conclude that a more detailed investigation that accounts for the multi-scale implications of these additional pathways is needed in the future . With our approach , the growth curves captured in our spatially resolved model ( a slow growth regime followed by a fast unidirectional linear growth ) are compatible with the experimentally observed growth of avascular solid tumors [63 , 73] . Previous studies attributed the linear growth regime observed at late-time tumor growth to available space for growth and cell diffusion at the edge of the tumors [73 , 74] . Here , our simulations and analysis indicate that this effect can be entirely explained by diffusion limitations of metabolites . In our exploration of Warburg and reverse Warburg effect , our approach provided a basis for exploring the heterogeneity in metabolic phenotypes that has been suggested by recent experiments [65 , 66] . For example , the crossover of growth rates that we observed from early to late times ( Figs 5C and 6C ) suggests that adaptation of metabolic phenotypes ( e . g . , from high to intermediate WN or from RW to Warburg ) could improve overall growth potential of tumors . In parallel with experimental approach , computational tools allow for high throughput investigation of hypotheses that are emerging rapidly in the field of cancer study [24 , 29 , 33 , 34 , 68 , 75 , 76] . Particularly , a multi-scale modeling framework such as the one presented here can provide a basis for predicting cell-level to tissue-scale response to therapeutic interventions . For example , the action of inhibitors of key regulators of cellular metabolism such as PI3K [77] can be accounted for in FBA as flux constraints ( e . g . , a reduced upper bound on glycolytic flux ) ; the obtained uptake rates of metabolites could then be propagated through to the tissue-scale ABM in our framework in order to examine the effect on tumor growth at the population scale . We finish by emphasizing that our interpretations of the three metabolic scenarios studied here are not unique either with respect to the choices of constraints and objectives imposed for FBA or the details of the cellular configurations within our simulations . Our modeling framework can accommodate a large diversity of hypotheses and should serve as a powerful tool with which to evaluate emerging ideas and experimental observations from the rapidly evolving field of cancer metabolism .
To capture the intracellular details of different metabolic phenotypes of cells , we adopt the well-established framework of FBA . In our study , the central carbon metabolism of human was constructed with 140 reactions and 92 metabolites ( S1 Fig ) . Of those 140 reactions , 34 consist of boundary exchange of metabolites such as uptake and secretion , 26 consist of mitochondrial exchange of metabolites with the cytosol , 1 is the biomass template reaction , 1 reaction for maintenance , and the 78 remaining reactions are transformations of metabolites that occur in the cytosol and mitochondrion . The biomass template reaction for growth in the human model was adapted from the Shlomi et al . ’s genome-scale model [20] . Shlomi et al . ’s biomass template reaction consists of 30 biomass compounds including amino acids ( 0 . 78 g/g-DW ) , nucleotides ( 0 . 06 g/g-DW ) , and lipids ( 0 . 16 g/g-DW ) . These biomass requirements were combined and reduced to their upstream precursors for simplification in our biomass template reaction . For example , stoichiometric equivalence of ribose-5-phosphate , the precursor of nucleic acids , was used in place of nucleotides in their final form . For the maintenance rate , we first sampled a range of values from 0 to 10 mmol ATP/g-DW-hr [78–80] and concluded that the overall qualitative trend of our FBA results was not affected by this choice . Therefore , for simplicity , a maintenance rate of 5 mmol/g-DW-hr is used consistently for all cell types . This maintenance rate represents 73% of the total energy expenditure , comparable to what was previously reported for mammalian cells , which is 65% [80] . Using our reduced biomass function , our glucose yields ( YG/n ) matched closely with that of Shlomi and coworkers [20] . For example , within the metabolic phenotype of WN = 0 , at the same growth rate range and maintenance rate of 0 , the yield coefficient ( specific growth rate per glucose ) of our reduced order model ( 0 . 0984 g-DW/mmol ) was within 4% of that found with Shlomi et al . 's genome scale model ( 0 . 094 g-DW/mmol ) . Under hypoxic conditions ( CO << KO ) , we assumed a quiescent phenotype for all cell types . To capture the hypoxic condition , we minimized the oxygen uptake rate while maintaining a growth rate of 1 ×10−6 hr-1 to represent the quiescent state . For tumor cells in the metabolic scenarios of reverse Warburg effect and glutamine addiction , we used a quiescent phenotype for tumor cells under hypoglycemic conditions ( CG <<KG ) . We achieve this condition in FBA by minimizing glucose uptake while allowing uptake of lactate or glutamine respectively and constraining growth rate to be 1×10−6 hr-1 . iDynoMiCS is an individual-based modeling platform originally built for the study of microbial biofilms [39] . It allows computation of diffusion-reaction kinetics at individual cell level and has multiple built-in kinetic mechanisms , including Monod forms as in Eqs 5–10 . Additionally , iDynoMiCS treats the cell movement through two mechanisms: displacements due to pressure-induced convection at the global scale based on Darcy’s law , and sterically induced displacements that avoid overlapping during the expansion and division of neighboring cells at a local scale . During a simulation , the pressure that is directly proportional to the rate of biomass generation or degradation is computed first to induce global convection , followed by the computation of “shoving” ( random displacement ) at local scale; these displacements are selected by a relaxation algorithm to avoid steric overlap . The shoving mechanism is propagated through all cells until the number of cells that are still moving is negligible , and leads to local random displacements of cells [39] . In our case , since we are explicitly interested in studying how diffusion-reaction kinetics impact the tumor growth under various hypotheses on cancer metabolism with no specific consideration of molecular guidance for cell movements , the random , local cell motion provided by iDynoMiCS serves as a reasonable approximation of cell dynamics within the tissue [81] . The 2-D simulation domain is discretized into a square grid on which the reaction-diffusion equation is solved by finite difference at each time step ( Eqs 2 and 4 ) . The domain is also divided into two compartments: the “tank” and the “biofilm” . The tank serves as the source of metabolites; we interpret this compartment to be the blood stream with which the tissue exchanges nutrients . The “biofilm” defines the tissue where the metabolites undergo diffusion and reaction; the local reaction rate for each metabolite is set by the density and metabolic character of the cells in the grid element . A boundary layer defines the resistant to diffusive mass transfer between the blood stream ( “tank” ) and the cells ( “biofilm” ) . In our axial simulations , we allowed the exchange of metabolites only at the top of the domain by having zero-flux boundary condition at the bottom of the domain and periodic boundary conditions on the sides and in the 3rd dimension ( S3 Fig ) . We set the concentrations of metabolites in the “tank” at their physiological concentrations in human blood stream ( S1 Table ) . We selected a grid size for solving reaction-diffusion process to match individual mammalian cell size ( ~10 μm , [82] ) and a boundary layer thickness , h , to represent the thickness of the vascular endothelium ( S1 Table ) . The size of the cell was used to determine the density of the cell based on dry cell mass ( S1 Table ) . With the density of the cell fixed , we calculated the spherical volume of the cell from biomass growth by conservation of mass . This volume was then used to calculate the diameter of cells at each time step . The calculated diameter at each time step was then used to compare to a threshold value to determine the division of the cell . Once the computational domain was defined , we then specified the reactions that govern the cell growth . In each reaction , we chose parameters such as half saturation constant ( S1 Table ) . Together with parameters such as diffusion coefficients and physiological concentrations of metabolites obtained from the literature , we checked that the calculated value of the Krogh length ( e . g . , ~40μm for oxygen ) was in the right range for mammalian tissue . In the calculation of Krogh length , we treat the tissue as a continuum and represent consumption of oxygen and glucose as being zeroth order within the steady state reaction-diffusion equation . We calculated the Krogh lengths to determine the limiting metabolite in tumor cell growth in different metabolic scenarios ( i . e . , different WNs , Fig 5F ) . The Krogh lengths represent the typical depth of penetration of metabolites into the tumor compartment . We omitted the consumption contributed by anaerobic growth of the cells by assuming the metabolites get completely depleted before the cells switch to anaerobic growth regime . The calculation of Krogh lengths is illustrated in S2 Fig . The metabolite with shorter Krogh length will play a more significant role in determining the growth dynamics of tumor cells . In Figs 5–7 , we evaluated early-time growth rates as the initial slope of the growth curves by taking the difference of the averaged number of tumor cells for the first two outputs of simulation and dividing by the time interval . The time intervals are 10 days , 20 days and 50 days for the cases of 1 , 3 and 5 layers of stromal cells for all three metabolic scenarios . Late-time growth rates were obtained in a similar fashion but evaluated at different time intervals due to the difference in breakthrough times in different cases . A growth over 30 days between the time points 30 and 60 days was used in the case of 1 layer of stromal cells . A growth over 80 days between the time points of 120 and 200 days was used for calculation of late-time growth rates in the case of 3 layers of stromal cells . A growth over 200 days between the time points of 400 and 600 days was applied to the calculation of late-time growth rates in the case of 5 layers of stromal cells . These choices of time ranges were applied consistently in all three metabolic scenarios . | Cancer metabolism is an emerging hallmark of cancer . In the past decade , a renewed focus on cancer metabolism has led to several distinct hypotheses describing the role of metabolism in cancer . To complement experimental efforts in this field , a scale-bridging computational framework is needed to allow rapid evaluation of emerging hypotheses in cancer metabolism . In this study , we present a multi-scale modeling platform and demonstrate the distinct outcomes in population-scale growth dynamics under different metabolic scenarios: the Warburg effect , the reverse Warburg effect and glutamine addiction . Within this modeling framework , we confirmed population-scale growth advantage enabled by the Warburg effect , provided insights into the symbiosis between stromal cells and tumor cells in the reverse Warburg effect and argued that the anaplerotic role of glutamine is not exploited by tumor cells to gain growth advantage under resource limitations . We point to the opportunity for this framework to help understand tissue-scale response to therapeutic strategies that target cancer metabolism while accounting for the tumor complexity at multiple scales . | [
"Abstract",
"Introduction",
"Model",
"Results",
"Discussion",
"Methods"
] | [
"cell",
"physiology",
"carbohydrate",
"metabolism",
"medicine",
"and",
"health",
"sciences",
"chemical",
"compounds",
"oxygen",
"carbohydrates",
"cell",
"metabolism",
"organic",
"compounds",
"glucose",
"metabolism",
"glucose",
"physiological",
"processes",
"oxygen",
"meta... | 2018 | Multi-scale computational study of the Warburg effect, reverse Warburg effect and glutamine addiction in solid tumors |
There are a number of vaccine candidates under development against a small number of the most common outbreak filoviruses all employing the virus glycoprotein ( GP ) as the vaccine immunogen . However , antibodies induced by such GP vaccines are typically autologous and limited to the other members of the same species . In contrast , T-cell vaccines offer a possibility to design a single pan-filovirus vaccine protecting against all known and even likely existing , but as yet unencountered members of the family . Here , we used a cross-filovirus immunogen based on conserved regions of the filovirus nucleoprotein , matrix and polymerase to construct simian adenovirus- and poxvirus MVA-vectored vaccines , and in a proof-of-concept study demonstrated a protection of the BALB/c and C57BL/6J mice against high , lethal challenges with Ebola and Marburg viruses , two distant members of the family , by vaccine-elicited T cells in the absence of GP antibodies .
The family Filoviridae includes 5 distinct viruses in the Ebolavirus genus: Zaire Ebola virus ( EBOV ) , Sudan virus ( SUDV ) , Reston virus ( RESTV ) , Tai Forest virus ( TAFV ) , and Bundibugyo virus ( BDBV ) ; 2 viruses in the Marburg-virus genus: Marburg virus ( MARV ) and Ravn virus ( RAVV ) ; and 1 virus in the Cuevavirus genus: Lloviu virus ( LLOV ) . The first identified filovirus disease was caused by MARV and occurred in Europe in 1967 . Since then , there have been over 50 recorded zoonotic outbreaks causing hemorrhagic fevers in humans and non-human primates with 90% fatality rates [1 , 2] . There is no vaccine or drug licensed against any member of the filovirus family . Thus , development of an effective vaccine is of great importance for public health in Africa , where outbreaks occur periodically , as well as for the rest of the world . At least seven vaccine platforms vectored by human and simian ( chimpanzee ) adenoviruses HAdV-5 , HAdV-26 , ChAdV-3 , vesicular stomatitis virus ( VSV ) , human cytomegalovirus , modified vaccinia virus Ankara ( MVA ) , plasmid DNA , subunit proteins and virus-like particles have been tested in nonhuman primates ( NHPs ) and encouraging results were obtained with two candidates , replicating VSV-ZEBOV ( EBOV ) and non-replicating ChAd3-ZEBOV , showing a single dose efficacy against EBOV challenge [3 , 4] . However , before the 2013 epidemic , only one vaccine reached phase 1 trial in humans and was abandoned . Facing the 2013 epidemic , the most promising vaccines were moved to clinical trials [5–10] and one , rVSV-ZEBOV reported efficacy in a human phase 3 trial [6] . During the 2018 Ebola outbreak in the Democratic Republic of Congo , death toll was reduced to 29 due to a number of factors; the rVSV-ZEBOV vaccine was experimentally deployed , but no data indicated its contribution to the reduced outbreak . Most of the above efforts focus on EBOV , because this virus is historically the most frequent cause of filovirus outbreaks , and all employ the virus glycoprotein ( GP ) . While there is a high degree of conservation in the GP within one species , so that , for example , antibody responses to EBOV vaccine would likely cross-react with other EBOV outbreak variants , protection against other filoviruses by the current vaccines will be very low [11] . Indeed , rVSV-ZEBOV induced 50% cross-protection for SUDV [12] and protection against other more distant viruses of the filovirus family would likely be much lower and require a multi-species vaccine [13] . An ideal vaccine should be effective not only against the currently prioritized outbreak species , but across all variants of the 8 distinctive filovirus members and provide a degree of protection even against the likely existing , but as yet unencountered species . Induction of CD8+ T-cells provides such an opportunity . The FILOcep1&2 vaccines constructed here aim to induce protective T-cell responses against viruses across the filovirus family . While the four most conserved regions of the filovirus family were identified and the theoretical corresponding epigraph regions were computed previously [11] , in the present work , we describe construction of the candidate pan-filovirus T-cell four-component vaccine vectored by simian adenovirus and poxvirus MVA , demonstrate their broad immunogenicity in the BALB/c and C57BL/6J strains of mice and report a solid protection of mice by vaccination from highly lethal EBOV and MARV experimental challenges . This protection was mediated solely by T-cell responses in the absence of GP-specific antibodies . The possible role of this vaccine in the preparedness for the future filovirus outbreaks as well as its use for treating residual infection are discussed .
The FILOcep1&2 vaccines aim to induce protective T-cell responses against viruses across the filovirus family . This is achieved by targeting the most similar , structurally and functionally conserved regions among the virus proteomes , and maximizing the match of the vaccine to all potential 9-mer T-cell epitopes ( PTE ) within these regions by computing bi-valent Epigraph sequences [11] . Epigraphs are the next generation of the pluri-valent mosaic design [14] aiming to maximize the coverage of a diverse , variable population of pathogens by bioinformatics-assisted computed amino acid sequences . The main improvement over mosaic is that epigraphs “walks” through the protein sequence and ensures that all PTE sequences used occur in the natural isolates present in the starting database . The rationale of the immunogen design is as follows . The best match is ensured with the EBOV species , which historically seeded the most outbreaks . The immunogens still have an excellent match to the other common outbreak species SUDV and MARV , and within the conserved regions maintain a good match to all other known filovirus PTEs [11] . Overall , there is a minimum of 8/9-amino acid match within a PTE to 80% filovirus isolates . Each of the four regions of epigraph 1 and epigraph 2 differ in about 10% amino acids , include a span of minimum 100 amino acids , together total 827 amino acids , and ensure broad representation of human leukocyte antigens ( HLAs ) for the restricted epitopes . To decrease potential induction of strong irrelevant CD8+ T cells recognizing new and , therefore , irrelevant non-viral epitopes generated by joining two adjacent regions together , the four conserved filovirus regions are assembled in two unique orders: 1-2-3-4 in FILOcep1 and 4-3-2-1 in FILOcep2 ( Fig 1 ) . The FILOcep1&2 immunogens were delivered to the cells of the immune system employing non-replicating engineered chimpanzee adenovirus ChAdOx1 and non-replicating poxvirus MVA as vaccine vectors [15] . The combination of these heterologous vectors has been shown to induce robust CD8+ T-cell responses in human volunteers for other indications [5 , 16 , 17] . Here , synthetic open-reading frames coding for FILOcep1 and FILOcep2 were inserted into the vector genomes to be administered in a four-component vaccine regimen , whereby the ChAdOx1 . FILOcep1 + ChAdOx1 . FILOcep2 vaccines were used together as a prime and MVA . FILOcep1 + MVA . FILOcep2 were used together as a boost ( Fig 1 ) . We optimized and characterized the vaccine-elicited T-cell responses in the BALB/c mice ( H-2d ) . For each vaccine component individually and two epigraphs together , four escalating doses were administered intramuscularly . The frequencies of FILOcep1&2-specific T cells were determined in an IFN-γ ELISPOT assay employing 12 pools of variant peptide pairs derived from the two FILOcep1 and FILOcep2 epigraphs . Thus , doses ranging from 106 to 5x108 infectious units ( IU ) were assessed for ChAdOx1 . FILOcep1 ( C1 ) and ChAdOx1 . FILOcep2 ( C2 ) individually and for half-doses together as C1C2 , and the dose of 1x108 IU was chosen for further vaccinations ( Fig 2A ) . For MVA . FILOcep1 ( M1 ) , MVA . FILOcep2 ( M2 ) and two half-doses of M1M2 , a range from 1x105 to 1x107 plaque-forming units ( PFU ) was tested and 107 PFU was chosen for further experiments ( Fig 2B ) . Broadly specific responses against 8 pools with higher that 50 SFU/106 splenocytes and dominant pools P3 and P12 were induced , which summed across all 12 pools for the combined C1C2 and M1M2 deliveries to median of 4207 and 1109 SFU/106 splenocytes , respectively . Next , we determined that C1C2 was synergistically boosted with M1M2 totalling median 12495 SFU/106 splenocytes ( Fig 3A ) . The most potent was a combination of C1C2 delivered into one site and M1M2 into another site over the mixed C1M1 and C2M2 administration ( Fig 3B ) . Administration of all four vaccine components at the same time was much less potent than heterologous C1C2 prime and M1M2 boost separated by 3 weeks ( Fig 3C ) . In the BALB/c mice , we mapped highly stimulatory 15-mer peptides ( S1 Fig and Fig 3D ) and used their pairs , one from each epigraph , to demonstrate induction of plurifunctional IFN-γ , TNF-α , IL-2 and CD107a responses . CD8+ T cells produced mainly IFN-γ , TNF-α and degranulated ( CD107a ) concurring with their cytolytic capacity , while CD4+ T cells produced IFN-γ , and IL-2 ( Fig 3E ) . Between 27% to 61% of CD8+ T cells produced 3 functions in parallel , while CD4+ T cells were mainly monofunctional . We narrowed down the two most immunodominant CD8+ T-cell responses in peptides 105 and 336 to ASFKQALSNL ( AL10 ) and GYLEGTRTLLAS ( GS12 ) , respectively ( Fig 3F ) . The optimal length of these epitopes present in the two vaccine epigraphs were compared to the sequences across the filovirus family . Epitope variants N/ASFKQALSNL in FILOcep1 and FILOcep2 matched EBOV and MARV , respectively , and differed for several other filoviruses with the strongest ASFKQALSNL ( MARV and RAVV ) yielding 1000 SFU/106 splenocytes and SSFKAALGSL ( SUDV ) and LAFKSALEAL ( LLOV ) not recognized at all . In contrast , GS12 was conserved across the entire family and strongly recognized at 1300 SFU/106 splenocytes ( Fig 3G ) . Next , we set out to assess the protective efficacy of the vaccine-elicited T cells , in our case in the absence of any GP-specific antibody , against two distant filoviruses , EBOV and MARV . Using the best regimen of the 4 vaccine components identified above , groups of 20 BALB/c mice received either the FILOcep1 and FILOcep2 vaccines or control eGFP vaccines , the latter expressing enhanced green fluorescent protein ( eGFP ) as an irrelevant protein with no homology to the filovirus family ( Table 1 ) . Four animals in each group were sacrificed 1 week after they received rMVA and , employing two different commercial IFN-γ ELISPOT kits , high frequency T cells specific for the FILOcep1&2 immunogens were detected in animals receiving the test vaccines , while no FILOcep1&2-specific responses were induced by the control eGFP vaccines ( Fig 4A ) . This confirmed compatible immunopotency between the Oxford and Winnipeg laboratories . Of the remaining 16 animals in each group , 8 were exposed to a lethal challenge with 1000 LD50 of mouse-adapted EBOV ( Mayinga ) [18] and 8 with 1000 LD50 of mouse-adapted MARV ( Angola ) [19] 4 weeks post vaccination and their body mass was recorded daily . While all the animals in the control group started losing mass precipitously and either died or had to be euthanized between days 4 and 6 post challenge , all the FILOcep1&2 vaccine recipients maintained normal body mass and survived till the end of the scheduled protocol on day 29 post challenge ( Fig 4B ) . In the repeat experiment , mice were sacrificed 3 and 5 days after challenge and the EBOV and MARV genomes were quantified in the blood , spleen , liver , kidneys and lungs . For EBOV , between 4 and 6 log10 fewer genomes per mg of tissue were found in FILOcep1&2 vaccinated mice , while for MARV , virus was only detected in FILOcep1&2 vaccine recipients in the blood at 1000 genomes/mg on day 5 after challenge ( Fig 4C ) . We conclude that the T-cell responses induced by the ChAdOx1 . FILOcep1 + ChAdOx1 . FILOcep2 prime-MVA . FILOcep1 + MVA . FILOcep2 boost regimen protected the BALB/c mice from both the EBOV and MARV lethal challenges and did so in the absence of glycoprotein antibodies . We also assessed the breadth of T-cell responses induced in the C57BL/6J strain of mice ( H-2b ) . Groups of mice were immunized with either FILOcep1 , FILOcep2 or combined half-doses of both epigraphs and a pattern on immunodominance was observed distinct from that in the BALB/c mice with the strongest peptide pools P3 , P4 , P5 and P7 ( Fig 5A ) . The challenge experiment followed the design in Table 1 . Four mice were sacrificed on day 28 of the schedule to confirm induction of FILOcep1&2-specific T-cell responses in the vaccine recipients using the four most dominant peptide pools and some variability among animals in the relative frequencies of T-cells was noticed ( Fig 5B ) . Following experimental challenge with Ebola and Marburg viruses of 8 animals per group , control animals started to lose their body mass and all died or were euthanized by day 7 post challenge with the exception of one MARV-challenged control mice , which regained mass and was still alive on day 28 . In contrast , all mice which received the FILOcep1&2 vaccines kept gaining body mass and stayed alive till the end of the protocol ( Fig 5C ) . Thus , the FILOcep1&2 vaccines protected against Ebola and Marburg viruses in both the BALB/c and C57BL/6J strains of mice carrying different H-2 molecules and presenting different peptides , and the vaccine-elicited T cells did so in the absence of challenge virus-specific antibodies ( Fig 6 ) .
In the present work , the ChAdOx1-MVA/FILOcep1&2 vaccines induced broadly specific , plurifunctional T-cell responses in mice and proved the concept that a pan-filovirus T-cell vaccine alone , in the absence of GP antibodies , can confer a 100% protection against experimental 1000 LD50 lethal challenges with filoviruses of two different genera and do so in the BALB/c and C57BL/6J strains of mice . In our experience with chimeric T-cell immunogens similar to the FILOcep1&2 proteins delivered by DNA , recombinant simian adenoviruses and MVA , and administered to mice , NHPs and humans , induction of transgene product-specific antibodies was extremely rare [20–22] . Because the intracellularly expressed proteins in the absence of any naturally evolved folding are unstable and there is no surface GP included , no readily detectable filovirus-specific antibodies were induced . Therefore , we consider it highly unlikely that anti-FILOcep1&2 antibodies contributed the observed protection . The strongest 15-mer peptides in the BALB/c mice were mapped in pools P3 , P11 and P12 . For a few epigraph variant pairs , the responses were similar , for other pairs , one variant was poorly or not recognized at all . This may reflect the differences that the bi-valent epigraph has to cover even for some of the most conserved protein regions; the coverage for variable regions must be worse . Two most immunodominant epitopes recognized by CD8+ T cells were narrowed down to the optimal length . The strongest epitope of the two , GS12 , was conserved across the eight filoviruses . The other epitope AL10 happened to have a perfect match in FILOcep1 and FILOcep2 to the two challenge viruses EBOV and MARV , respectively , even though the EBOV variant yielded 5-fold lower specific T-cell frequencies and a great depth of recognition resulting from the bi-valent vaccine immunization was not achieved for this epitope . Thus , the coverage of epitopes in the filovirus species by the bi-valent vaccine will likely differ for each individual epitope and the protective effect against viruses will depend on the number of different epitopes recognized by the vaccine-elicited effector cells . Sometimes one protective invariant epitope may suffice , while recognition of multiple epitopes provides a better chance for virus control . Previously , a protective role of T cells in immunity against EBOV in mice was suggested by studies in genetically modified mice [23] and by passive transfer of lymphocytes [24 , 25] although in the one of the studies , a role for humoral immunity was also implicated [24] . A protection by T cells against several heterologous EBOV species was also reported by Hensley and colleagues [12] . There are a very few known HLA-restricted epitopes derived from filoviruses . The FILOcep1&2 regions span amino acids 131–420 in the nucleoprotein , 71–193 in matrix , and 540–854 and 952–1060 in the RNA polymerase , and of these the nucleoprotein has been the most studied . Searching the Immune Epitope Database ( IEDB; https://www . iedb . org/ ) for known T-cell epitopes in filoviruses currently yields 10 well defined human CD8 T-cell epitopes , of which 7 are contained in the vaccine ( Table 2 ) . The number of HLA-restricted PTEs covered by the FILOcep1&2 vaccines can be estimated by shifting an 8- , 9- and 10-amino acid-long window across the 827-amino acid proteins , which gives 820 8-mer , 819 9-mer and 818 10-mer PTEs times two for the two FILOcep1&2 immunogens . Each of these PTEs then needs to be predicted for binding to major HLA alleles . This analysis would almost certainly yield more than sufficient number of human epitopes for all major HLAs to induce a broad response in every individual . Immunodominance will always be established narrowing down the response specificities . The protective potential of the FILOcep1&2 vaccine-elicited responses in humans can be only established by exposure of vaccinated individuals . Our next step is to determine whether or not the efficacy of T cells alone induced by the ChAdOx1-MVA/FILOcep1&2 vaccines translates to NHPs . If the mouse protection is replicated in NHP , identification of the correlates of protection in NHPs might greatly encourage testing the immunopotency of these vaccines in human volunteers . In the past , CD8+ T cells induced by HAdV-vectored vaccine conferred protection of NHPs against EBOV infection [26 , 27] . The likely absence of an opportunity to demonstrate a human phase 3 efficacy may allow an alternative licensure pathway . A successful pan-filovirus vaccine would have multiple uses such as generation of vaccine stockpiles for containment of future outbreaks , elimination of the 2013 and 2018 outbreak remnants , elimination of virus reservoirs in survivors , provision of long-term protection in high risk populations including health workers and may even help saving highly endangered western gorillas .
Two DNA fragments carrying the two FILOcep1 and FILOcep2 ORFs were synthesized ( Life Technologies ) using humanized codons and were preceded by the consensus Kozak sequence to -5 nucleotides to maximize protein expression . The parental non-replicating MVA originates directly from Professor Anton Mayr , passage 575 dated 14 December , 1983 . The FILOcep1 , FILOcep2 or eGFP ORFs were cloned into transfer plasmid p856MVA-GFP-mH5 under control of the modified H5 promoter . Through homologous recombination , the expression cassettes were directed into the thymidine kinase locus on the MVA genome . Recombinant MVAs were made as described elsewhere . Briefly , chicken embryo fibroblast ( CEF ) cells grown in Dulbeco’s Modified Eagle’s Medium supplemented with 10% FBS , penicillin/streptomycin and glutamine ( DMEM 10 ) were infected with parental MVA at MOI 1 and transfected using Superfectin ( Qiagen ) with 3 μg of p856MVA-GFP-TD-mH5 . FILOcep1 or p856MVA-GFP-TD-mH5 . FILOcep2 DNA . The cell lysate from this recombination was harvested and used to infect CEF . These cells were MoFlo-single cell sorted into 96-well plates and these were used to culture recombinant virus upon addition of fresh CEF . Those wells containing suitably infected cells were harvested and screened by PCR to confirm identity and test purity . Plaque picking was performed until the culture was free of parental virus , as determined by PCR . The virus was then bulk-prepared and purified on a 36% sucrose cushion , titred and stored at –80 oC until use . The ChAdOx1 vaccine vector is derived from ChAdV isolate Y25 of group E adenoviruses , and pre-existing antibodies to group E are rare in human populations . Its genome modifications include removal of the E1 , E3 and a substitution of simian region E4 with the HAdV-5 E4 orf4 and orf6/7 genes . For the generation of recombinant ChAdOx1s , the FILOcep1 and FILOcep2 ORFs were subcloned under the control of the human cytomegalovirus immediate early promoter into plasmid pENTR4_Mono and inserted at the E1 locus of the ChAdOx1 genome by GalK recombineering . Recombinant ChAdOx1 vaccines were rescued by transfection of HEK293A T-Rex cells ( Invitrogen/ThermoFisher Scientific ) using linearized plasmid . The presence of the transgene and absence of contaminating empty parental adenovirus were confirmed by PCR . The virus was titred to determine infectious units ( IU ) per ml , assayed by spectrophotometry to quantify the number of virus particles per ml and stored at –80 oC until use . Six-week-old female BALB/c or C57BL/6 mice were purchased from Envigo ( UK ) and housed at the Functional Genomics Facility , University of Oxford . Mice were immunized intramuscularly under general anesthesia either with varying amounts of rChAdOx1s and rMVAs . On the day of sacrifice , spleens were collected and cells isolated by pressing organs individually through a 70-μm nylon mesh of a sterile cell strainer ( Fisher Scientific ) using a 5-ml syringe rubber plunger . Following the removal of red blood cells ( RBC ) with RBC Lysing Buffer Hybri-Max ( Sigma ) , splenocytes were washed and resuspended in R10 ( RPMI 1640 supplemented with 10% FCS , penicillin/streptomycin and β-mercaptoethanol ) for ELISPOT and intracellular cytokine staining ( ICS ) assays . All peptides were at least 90% pure by mass spectrometry ( Ana Spec , San Jose , CA , USA and Synpeptide Co Ltd , Shanghai , China ) , were dissolved in DMSO ( Sigma-Aldrich ) to yield a stock of 10 mg/ml , and stored at –80°C . Three hundred and ninety FILOcep1&2-derived peptides 15-mer overlapping by 11 amino acids were divided into 12 pools P1-P12 of 34 to 47 individual peptides in a way that variant peptides were always present in the same pool for use in ICS and ELISPOT assays . 17 pairs of stimulatory ‘BALB/c’ peptides were employed as specified in each figure . The peptides were used at a final concentration of 1 . 5 μg/ml each . The ELISPOT assay was performed using the Mouse IFN-γ ELISpot kit ( Mabtech , Stockholm , Sweden ) or FluoroSpot kits ( Mabtech and Cellular Technology Limited , Cleveland , OH , USA ) according to the manufacturer’s instructions . For the former , immune splenocytes were collected and tested separately from individual mice . Peptides were used at 1 . 5 μg/ml each and splenocytes at 5 × 104 cells/well were added to 96-well high protein binding Immobilon-P membrane plates ( Millipore ) that had been precoated with 5 μg/ml anti-IFN-γ mAb AN18 ( Mabtech , ) . The plates were incubated at 37°C in 5% CO2 for 18 hours and washed with PBS before the addition of 1 μg/ml biotinylated anti-IFN-γ mAb ( Mabtech ) at room temperature for 2 hours . The plates were then washed with PBS , incubated with 1 μg/ml streptavidin-conjugated alkaline phosphatase ( Mabtech ) at room temperature for 1 hour , washed with PBS , and individual cytokine-producing units were detected as dark spots after a 10-minute reaction with 5-bromo-4-chloro-3-idolyl phosphate and nitro blue tetrazolium using an alkaline phosphatase-conjugate substrate ( Bio-Rad , Richmond , CA , USA ) . Spot-forming units were counted using the AID ELISpot Reader System ( Autoimmun Diagnostika ) . The frequencies of responding cells were expressed as a number of spot-forming units/106 splenocytes . Splenocytes or PBMCs isolated from whole blood were stimulated with peptide at 2 μg/ml , ionomycin and phorbol myristate acetate ( PMA ) at 2 . 0 μg/ml and 0 . 5 μg/ml , respectively , or tissue culture media with 1% DMSO as a negative control . The cultures were supplemented with anti-CD107a PE-conjugated mAb ( eBioscience ) . The cells were incubated at 37 oC , 5% CO2 for 2 hours prior to the addition of Brefeldin A and monensin ( BD Biosciences ) and then left in culture overnight . The cells were centrifuged briefly , washed in PBS plus 5% BSA ( Sigma-Aldrich ) and the pellet re-suspended in 40 μl of CD16/32 with LIVE/DEAD fixable aqua stain ( Molecular Probes , Invitrogen ) . Cells were washed , a mastermix of anti-membrane marker mAbs was prepared containing CD4 APC/Cy7 ( Biolegend ) , CD3 PerCP-eFluor710 and CD8a eFluor 450 ( both from eBioscience ) and 40 μl added to each tube . The cells were incubated at 4 oC for 30 min and then permeabilized using Fix/Perm solution ( Becton-Dickinson ) for 20 min at 4 oC . The cells were washed with Perm Wash buffer ( Becton Dickinson ) and a mastermix of anti-intracellular molecule mAbs was prepared containing IFN-γ PE-Cy7 , IL-2 APC and TNF-α FITC ( all from eBioscience ) . The cells were incubated at 4 oC for 30 min , washed and resuspended in Perm Wash buffer prior to running on an LSRII flow cytometer ( Becton-Dickinson ) . Groups of eight 6- to 7-week-old BALB/c or C57BL/6J female mice ( Charles River ) were vaccinated intramuscularly under general anesthesia with 1x108 IU total of rChAdOx1s followed by 1x107 PFU total of MVAs . At day 35 of the protocol or day 24 post-vaccination ( Table 1 ) , all the mice received a challenge dose of 1000x the 50% lethal dose ( LD50 ) of either mouse-adapted EBOV or mouse-adapted MARV in 200 μl of DMEM ( pH 7 . 4 ) by intraperitoneal injection . All animals were monitored daily for signs of disease , survival and body-mass change for 14 days followed by additional 14 days monitoring of survival . FILOcep1&2- and control eGFP-vaccinated mice were challenged with either mouse-adapted EBOV or MARV . Blood and tissues ( liver , spleen , kidney and lungs ) from 4 mice per vaccinated group were collected upon euthanasia at day 3 and 5 post-infection to determine viral RNA levels . RNA of mouse blood and tissues were extracted using QIAamp viral RNA minikit ( Qiagen ) and the RNeasy mini Kit ( Qiagen ) according to the manufacturer's instructions . Viral RNA levels were quantified by reverse transcription quantitative PCR ( RT-qPCR ) targeting viral polymerase gene and using the Light Cycler 480 thermal cycler ( Roche , Germany ) . The primers and probes are shown in Table 3 . Cycling conditions were as follows: 63°C for 3 min and 95°C for 30 sec , followed by 45 cycles of 95°C for 15 sec and 60°C for 30 sec . EBOV and MARV concentrates were prepared from virus-infected Vero cell culture supernatants by unlracentrifugation and inactivation by γ-irradiation . Viruses were lysed in a loading buffer and an equivalent of 8 μg of virus or PBS were separated on 15% SDS-polyacrylamide gel and transferred onto a nylon filter ( Amersham International ) , and the filters were blocked and incubated with mAbs 14E2 ( EBOV NP ) , 7A12 ( MARV NP ) or combined mouse sera from each animal group diluted 1:1000 . Bound antibodies were detected using horse radish peroxidase ( HRP ) -conjugated protein A ( Amersham International ) followed by enhanced chemiluminiscence ( ECL; Amersham International ) . Statistical analyses were performed using Graph Pad Prism version 7 . Responses were assumed to be non-Gaussian in distribution , thus non-parametric tests were used throughout and medians ( range ) are shown . Multiple comparisons were performed using the Kruskal-Wallis test . Groups with the same in vitro restimulations were compared using two-tailed Mann-Whitney U tests . Two-tailed P values were used and P values of less than 0 . 05 were considered statistically significant . Chicken embryo fibroblasts were prepared at Poultry Health Services Ltd , Huntingdon , UK and , in the United Kingdom , there is no need for Ethics permission for killing 7-day-old chicken embryos . All mouse procedures and care in Oxford were approved by the local Clinical Medicine Ethical Review Committee , University of Oxford and conformed strictly to the United Kingdom Home Office Guidelines under the Animals ( Scientific Procedures ) Act 1986 . Experiments were conducted under Project License 30/3387 held by T . H . All the animal challenge experiments were performed in the biological safety level 4 ( BSL-4 ) facility at the Canadian Science Centre for Human and Animal Health ( CSCHAH ) in Winnipeg , Canada . All mouse procedures and care at the Canadian Science Center for Human and Animal Health ( CSCHAH ) were approved by the local Animal Care Committee and conformed strictly to the Canadian Council on Animal Care ( CCAC ) . Experiments were conducted under Animal Use Document H17-007 held by X . Q . | Development of an effective vaccine against filovirus outbreaks is an important public health aim . Here , we demonstrate the principle that cellular responses can not only protect two strains of mice against a high lethal virus challenge of 1000 LD50 in the absence of glycoprotein antibodies , but a single epigraph T–cell vaccine can do so against distant members of the filovirus family , EBOV and MARV . This suggests a possibility that this candidate vaccine also protects against other known as well as yet unencountered viruses of the filovirus family; it is a pan-filovirus vaccine . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] | [
"blood",
"cells",
"viral",
"vaccines",
"medicine",
"and",
"health",
"sciences",
"immune",
"cells",
"enzyme-linked",
"immunoassays",
"pathology",
"and",
"laboratory",
"medicine",
"immune",
"physiology",
"pathogens",
"immunology",
"microbiology",
"vaccines",
"viruses",
"f... | 2019 | Complete protection of the BALB/c and C57BL/6J mice against Ebola and Marburg virus lethal challenges by pan-filovirus T-cell epigraph vaccine |
While most models of randomly connected neural networks assume single-neuron models with simple dynamics , neurons in the brain exhibit complex intrinsic dynamics over multiple timescales . We analyze how the dynamical properties of single neurons and recurrent connections interact to shape the effective dynamics in large randomly connected networks . A novel dynamical mean-field theory for strongly connected networks of multi-dimensional rate neurons shows that the power spectrum of the network activity in the chaotic phase emerges from a nonlinear sharpening of the frequency response function of single neurons . For the case of two-dimensional rate neurons with strong adaptation , we find that the network exhibits a state of “resonant chaos” , characterized by robust , narrow-band stochastic oscillations . The coherence of stochastic oscillations is maximal at the onset of chaos and their correlation time scales with the adaptation timescale of single units . Surprisingly , the resonance frequency can be predicted from the properties of isolated neurons , even in the presence of heterogeneity in the adaptation parameters . In the presence of these internally-generated chaotic fluctuations , the transmission of weak , low-frequency signals is strongly enhanced by adaptation , whereas signal transmission is not influenced by adaptation in the non-chaotic regime . Our theoretical framework can be applied to other mechanisms at the level of single neurons , such as synaptic filtering , refractoriness or spike synchronization . These results advance our understanding of the interaction between the dynamics of single units and recurrent connectivity , which is a fundamental step toward the description of biologically realistic neural networks .
The existence of a chaotic phase is a common property of large networks of neurons with random connectivity [1 , 2] . Chaotic dynamics has been proposed as a mechanism for internally-generated cortical variability [3–5] and the richness of the dynamics at the edge of chaos has been exploited to learn complex tasks involving generation of temporal patterns [6–12] . In these and other related approaches , the chaotic behavior of the network mainly arises from the random interactions , whereas the dynamics of single neurons are typically given by first-order differential equations . The simplicity of single neuron dynamics in these models allows to quantitatively determine the chaotic phase of synaptically coupled neurons using dynamical mean-field theory ( DMFT ) [1] , even in networks with more realistic connectivity structure [2 , 12–14] . A fascinating question is what kind of activity emerges in neural networks that are subject to additional biological constraints . Biological neurons exhibit rich multi-dimensional internal dynamics [15–18] that are inconsistent with first-order equations . However , a theoretical understanding of the emergent activity patterns in networks of more realistic multi-dimensional neuron models is largely lacking . Here , we develop a theoretical framework that extends DMFT to multi-dimensional rate neurons . Using this framework , we show that the power spectrum of the network activity in the nonlinear , strongly coupled regime , emerges from a sharpening of the single-neuron frequency response function due to strong recurrent connections . Our theory uses firing rate models with two or more variables per unit . While rate-based models [19 , 20] discard information on the exact spike-timing of single neurons , they have the advantage of being accessible to an analytical characterization of their dynamics . However , commonly-used one-dimensional rate models cannot fully capture the dynamics of the mean activity of a population of spiking neurons , such as the synchronization of neurons in response to a stimulus onset [21–23] , an effect that is readily observed in integrate-and-fire models [24–28] . To capture rapid synchronization after stimulus onset in rate models , it is necessary to consider at least two equations per rate neuron [29–31] . Multi-dimensional models also account for additional cellular mechanisms such as refractoriness [32] , spike-frequency adaptation ( SFA ) [28 , 33–35] , synaptic filtering [36 , 37] , subthreshold resonance [38] or for the effect of dendritic compartments [39 , 40] . To be specific , we focus on SFA , the decrease of a neuron’s firing rate in response to a sustained stimulus , but our theory can also be applied to other phenomena . SFA is present in neurons at all stages of sensory processing , and is believed to play a crucial role for efficient coding of external stimuli [15] . Moreover , SFA over multiple timescales represents an efficient solution for information transmission of sensory signals whose statistics change dynamically [16 , 18 , 41] . It is therefore of great interest to understand how adaptation and recurrent connections interact to shape network dynamics and signal transmission [42 , 43] . If connections and adaptation are weak , the network dynamics can be largely understood within linear response theory . In particular , in the presence of signals and noise , linear response theory predicts that adaptation shapes signal and noise in precisely the same manner [35] , canceling the noise-shaping effect of adaptation [42–44] . In contrast , in strongly coupled networks generating chaotic fluctuations [1] , linear response theory is not applicable and the effect of adaptation on the signal transmission in this case remains poorly understood . Here , we show that introducing adaptation into a strongly-coupled network of rate units shifts the network to a state of “resonant” chaos that is qualitatively different from the chaotic behavior of the network without adaptation . In this state , the network generates a stable rhythm corresponding to a narrow-band peak in the power spectrum which is robust against quenched disorder in adaptation parameters ( heterogeneity ) . We show that in this new regime the network has two interesting functional properties: first , the correlation time increases with the adaptation timescale; second , the low-frequency power of the chaotic activity is strongly decreased , enabling a better transmission of slow signals . In the Results section , we first present the microscopic model for the network of rate neurons with adaptation and describe its dynamical regimes . Then , we introduce the mean-field approximation that allows us to describe the resonant chaotic state and to study its functional consequences . Finally , we present the general multi-dimensional model that allows to introduce multiple mechanisms at the single-neuron level . Detailed derivations are provided in the Methods section , while in the Discussion we examine possible extensions and generalizations .
The dynamics of the 2N-dimensional dynamical system in Eqs 1 and 2 for large N is too high-dimensional to be studied at the microscopic level . In contrast , using dynamical mean-field theory [1] , we can find properties of the network dynamics that are independent of the specific connectivity realization . In what follows , we will assume that the external input Ii ( t ) to each unit is an independent realization of the same stationary Gaussian process with zero mean . Following [1] , we approximate the network input to a representative unit i with a Gaussian process η , an approximation valid in the large-N limit [45 , 46] . The mean-field equations read ( see Methods ) x ˙ ( t ) =- x ( t ) - a ( t ) + η ( t ) + I ( t ) ( 5 ) a ˙ ( t ) =- γ a ( t ) + γ β x ( t ) , ( 6 ) where η ( t ) is a Gaussian process with zero mean and whose autocorrelation must be computed self-consistently by imposing ( see Methods ) ⟨ η ( t ) η ( s ) ⟩ = g 2 ⟨ ϕ ( x ( t ) ) ϕ ( x ( s ) ) ⟩ . ( 7 ) Due to the linearity of the mean-field equations , x ( t ) is a zero-mean Gaussian process which is fully characterized by its second-order statistics , i . e . the autocorrelation function in time domain , or the power spectral density Sx ( power spectrum ) in frequency domain , defined as the Fourier transform of the autocorrelation S x ( f ) = ∫ - ∞ ∞ e - 2 π i f τ 〈 x ( t + τ ) x ( t ) 〉 d τ . By Fourier transforming Eqs ( 5 ) and ( 6 ) , we find that the power spectrum is the solution of S x ( f ) = G ˜ ( f ) ( g 2 S ϕ ( x ) ( f ) + S I ( f ) ) , ( 8 ) where Sϕ ( x ) ( f ) and SI ( f ) are the power spectra of ϕ ( x ) and I respectively , defined analogously to the one of x . Importantly , Sϕ ( x ) ( f ) is a functional of Sx ( f ) , which can be computed semi-analytically for simple nonlinearities such as the piecewise-linear function , Eq 4 , as detailed in Methods , section “Effect of nonlinearities on second-order statistics” ) . The factor G ˜ ( f ) is the square modulus of the linear response function of an uncoupled single unit , and is given by G ˜ ( f ) = γ 2 + ω 2 ω 4 + ( 1 + γ 2 - 2 β γ ) ω 2 + γ 2 ( 1 + β ) 2 , ( 9 ) with ω = 2πf . By solving iteratively the mean-field equation for the power spectrum ( Eq 8 ) in the absence of external input ( see Methods , section “Iterative procedure to solve the mean-field theory” for details ) , we find that if g < gc ( γ , β ) , the power spectrum converges to zero , Sx ( f ) → 0 , at all frequencies . Therefore the mean-field variable x is constantly equal to zero . This is consistent with the presence of a stable fixed point at zero and it indicates that , for large N , the fixed point solution is the only possible one . On the other hand , if g > gc ( γ , β ) , the mean-field network is characterized by a nonzero , continuous power spectral density ( Fig 2 ) . This is an indication that , at the microscopic level , the network is in a chaotic state [47] . However , we stress that a more rigorous proof of chaos would require the computation of the maximum Lyapunov exponent of the network , which we will not perform . In contrast to a network without adaptation [1] , we find that in the presence of adaptation the network can be in two qualitatively different chaotic regimes . For very weak and/or fast adaptation , the chaotic fluctuations are qualitatively the same as for the network without adaptation , i . e . the power spectrum is broad-band with maximum at f = 0 ( Fig 2b ) . We refer to this regime as to the non-resonant regime . On the other hand , for strong and/or slow adaptation , the mean-field network settles in a new regime , characterized by an autocorrelation that decays to zero via damped oscillations and , equivalently , by a power spectrum that exhibits a pronounced resonance band around a nonzero resonance frequency fp ( Fig 2a ) . The decaying autocorrelation function and the continuous power spectral density are an indication that the network is—also in this regime—in a state of microscopic chaos . This new dynamical state , that we refer to as resonant chaos , is qualitatively different from the one of the non-resonant regime and from the one of the non-adaptive network . Strikingly , whether the network settles in the resonant or in the non-resonant regime can be predicted purely based on the single-unit adaptation properties . More precisely , if β < βH ( γ ) , the function G ˜ ( f ) is monotonically decreasing with the frequency f , i . e . it exhibits a low-pass characteristic ( Fig 2b ) . This low-pass behavior of the single neuron is reflected by a power spectrum of the network that is also dominated by low frequencies , albeit less broad . The network power spectrum corresponds exactly to the non-resonant regime discussed above . In contrast , if β > βH ( γ ) , the single neuron response amplitude G ˜ ( f ) exhibits a maximum at a nonzero frequency f 0 = 1 2 π - γ 2 + β γ 2 ( β + 2 γ + 2 ) . Such a resonance peak is typical of a band-pass filter ( Fig 2a ) . The frequency f0 is identical to fm ( Eq 28 ) , which is derived from the imaginary part of the critical eigenvalue at the Hopf bifurcation ( see Methods , section “Fixed-point stability” ) . The single-neuron linear response characteristics are qualitatively preserved in the fluctuating activity of the recurrent network , which also exhibits a power spectral density dominated by a nonzero frequency fp . This regime corresponds to the resonant regime discussed above . Interestingly , we find numerically that fp = f0 , i . e . the resonance frequency is not affected by the introduction of recurrent connections ( Fig 2c , tested up to g = 5gc ( γ , β ) ) . We notice that the non-resonant and resonant regimes are consistent with the fixed point stability analysis of the network in the microscopic description . Indeed , the resonant and non-resonant regimes match the regions in which we observe Hopf or saddle-node bifurcations , respectively ( see Methods ) . Using simulations of the full microscopic network , we verify that the mean-field description is a good approximation of the system for large but finite N . In Fig 2e we show that the probability density of the activation variable x measured from the microscopic simulations matches the Gaussian distribution predicted by the mean-field theory , with relatively small finite-size effects that increase close to the criticality ( see Fig 2e , g = 1 . 5gc ( γ , β ) ) . Moreover , the mean-field solution provides a good description of the system for a wide range of adaptation parameters γ , β ( Fig 2c ) . To summarize , single-neuron properties determine whether the network settles in a resonant or non-resonant chaotic state through the factor G ˜ ( f ) . This is a general property that is also valid for more complex rate models . Indeed , for a single-neuron model with squared linear response function given by an arbitrary G ˜ ( f ) , we have that the network power spectrum Sx ( f ) is the result of a nonlinear sharpening of G ˜ ( f ) ( see Methods , section “Qualitative study of the iterative map” for details ) . As a result , the network activity exhibits the same frequency bands that are preferred by single neurons , albeit much narrower . We will discuss more general rate model in section “Network of multi-dimensional rate neurons” . While the resonance frequency in the resonant regime seems to depend solely on the single-neuron properties , the introduction of recurrent connections increases the coherence of the stochastic oscillations , i . e . decreases the width of the resonance band . The narrower the resonance band , the more coherent the oscillatory behavior will be . To quantify the increase of the oscillation coherence , we measure the quality factor ( Q-factor ) of the stochastic oscillations , defined as Q = f p Δ f HM , ( 11 ) where fp is the frequency of maximum height of the power spectrum Sx and ΔfHM is the frequency width of the power spectrum Sx ( f ) at the half-maximum . Intuitively , for a narrow-band oscillation , the quality factor quantifies the number of oscillation cycles during the characteristic decay time of the autocorrelation function . For a single neuron driven by white noise , the single-neuron power spectrum of x is proportional to G ˜ ( f ) . Compared to this reference shape , we find a higher Q-factor in the recurrent network ( Fig 3a ) , corresponding to a sharper resonance peak in the power spectrum ( see also Methods , section “Qualitative study of the iterative map” ) . When approaching the criticality from the chaotic phase , g → gc ( γ , β ) + , the quality factor diverges ( Fig 3a ) , i . e . the dynamics approach regular oscillations . While the Q-factor measures the decay time constant of the autocorrelation function relative to the mean oscillation period , it is also interesting to consider the absolute correlation time of the activity . As a measure of correlation time of a stochastic process we use the normalized first moment ( center of mass ) of the absolute value of the autocorrelation function ( e . g . [49] ) , t c = ∫ 0 ∞ τ | C x ( τ ) | d τ ∫ 0 ∞ | C x ( τ ) | d τ . ( 12 ) Since the Q-factor diverges when g → gc ( γ , β ) , in this limit the corresponding autocorrelation exhibits sustained oscillations with a diverging correlation time . Due to the increase of the Q-factor , the correlation time also diverges when g → gc ( γ , β ) ( Fig 3a ) . In the regime of slow adaptation , a single unit driven by white noise can have a larger correlation time than a recurrent network ( Fig 3b ) . This is due to the fact that in this regime the correlation time of the single unit driven by white noise is dominated by the long tail of the autocorrelation . The introduction of recurrent connections increases the oscillatory component , giving a larger “weight” to the short time lags , thus decreasing tc . Nevertheless , the correlation time increases with the timescale of adaptation τa for both the single unit driven by white noise and the recurrent network ( Fig 3b ) . Note that the Q-factor goes to zero for very large adaptation timescale ( γ → 0 ) , so that the dominant contribution to the correlation time in this regime is the non-oscillatory one . In order to go beyond the study of the spontaneous activity of the network , we consider its response to an external oscillatory signal . While signal transmission in linear systems is fully characterized by the frequency response function of the system and by the noise spectrum of the output , the situation is different in the nonlinear neural network that we study here . Similarly to previous approaches [4] , we provide oscillatory input to each unit in the microscopic network , randomizing the phase ( Fig 4a ) I i ( t ) = A I cos ( 2 π f I t + θ i ) , ( 13 ) where θi ∼ U ( 0 , 2π ) . The corresponding power spectral density of the input is given by S I ( f ) = ( A I 2 / 4 ) · ( δ ( f - f I ) + δ ( f + f I ) ) . Thanks to the phase randomization , the network still reaches a stationary state and the mean 〈x ( t ) 〉 remains at zero . Notice that even if in the case in which the input is a perfect sinusoidal and therefore non-Gaussian , the mean-field equation for the power spectrum ( Eq 8 ) is still valid . However , since x is also not Gaussian anymore , in order to find the mean-field solution we need to modify our iterative scheme by splitting the activation variable x into its Gaussian and its oscillatory part [4] . The presence of the input affects the dynamics of the mean-field network , quantified by the power spectral density ( Fig 4b ) . If the input is given while the network is in the chaotic regime ( g > gc ) , sharp peaks at the driving frequency fI and multiples thereof are elicited by the external input , standing out from a background power spectrum that is deformed compared to the case without the external input . For fI > fp , as in the example , the bumps of the background spectrum are slightly shifted toward larger values . The opposite happens if fI < fp . Notice that both this shift and the shaping of the chaotic activity are nonlinear effects due to the recurrent dynamics . As an additional nonlinear effect , the network activity also exhibits harmonics at the driving frequency of the external input . To characterize the response to the external stimulus , we split the power spectrum Sx ( f ) into an oscillatory component and a chaotic component that constitutes the background activity S x ( f ) = S bkg ( f ) + S osc ( f ) ≔ S bkg + ∑ k = 1 ∞ b k ( δ ( f - k f I ) + δ ( f + k f I ) ) , ( 14 ) where bk are positive coefficients and we included the multiples of the driving frequency in order to account for the harmonics . To solve the mean-field equations numerically , we have to consider a finite frequency bin Δf ( in our numerical results , Δf = 0 . 001 ) . As a consequence , the heights of the delta peaks in the power spectrum in Eq 14 are finite and depend on Δf . First , we will look at the transmission of the oscillatory signal near the driving frequency , i . e . how much of the peak in the power spectrum Sx ( f ) at f = fI is due to the oscillatory drive and how much is due to the background activity . At the driving frequency fI we write ( see Fig 4c ) S x ( f I ) = A bkg + A osc ≔ S x ( f I - Δ f ) + S x ( f+Δf ) 2 + b 1 Δ f , ( 15 ) i . e . we measure the contribution of chaotic activity to the power spectrum at the driving frequency by interpolating the power spectrum at neighboring frequencies . The signal-to-noise ratio ( SNR ) at the driving frequency fI is then given by SNR ( f I ) = A osc A bkg . ( 16 ) Notice the size of the frequency bin Δf scales the SNR , but since we are interested in the dependency of the SNR on fI and not in its numerical value , this scaling factor can be neglected . Finally , we have seen in the example in Fig 4b that the oscillatory input can suppress background activity at frequencies far from fI . In order to quantify this chaos-suppression effect , we split the total variance of x into two contributions ( Fig 4c ) Var ( x ) = ∫ - ∞ ∞ S bkg ( f ) d f + 2 ∑ k = 1 ∞ b k ≕ P bkg + P osc . ( 17 ) We have seen how adaptation , by changing the response function of singe neurons , shapes the chaotic dynamics of a recurrent network and consequently the signal-transmission properties of the network . In biology , several other mechanisms could contribute to the response properties of neurons , such as synaptic filtering , facilitation or the presence of dendritic compartments [28–40] . We account for multiple of such mechanisms by considering a general D-dimensional linear-nonlinear rate model . The first variable x i 1 is an activation variable that defines the output rate y via a nonlinear function ϕ , i . e . y i ( t ) = ϕ ( x i 1 ( t ) ) , as in the adaptation case . The remaining D − 1 variables are auxiliary variables . We assume that the rate ϕ ( x j 1 ( t ) ) is the only signal that unit j uses to communicate with other units . Conversely , the signals coming from other units only influence the variable x i 1 , i . e . the rate of unit j is directly coupled only to the first variable of unit i . The choice of having the same variable sending and receiving signals is dictated by simplicity and is not necessary for the development of the theory . Unit i receives input from all the other units , via a set of random connections Jij , sampled i . i . d . from a Gaussian distribution with mean zero and variance g2/N . The resulting network equations are x ˙ i α ( t ) = ∑ β = 1 D A α β x i β ( t ) + δ α 1 ( ∑ j = 1 N J i j ϕ ( x j 1 ( t ) ) + I i ( t ) ) J i j ∼ N ( 0 , g 2 / N ) ( 21 ) where δαβ is the Kronecker delta symbol . Subscripts ( in Latin letters ) indicate the index of the unit in the network and run from 1 to N , while superscripts ( in Greek letters ) indicate the index of the variable in the rate model and run from 1 to D . The matrix A is assumed to be non-singular and to have eigenvalues with negative real parts . By generalizing the mean-field theory to the case of the D-dimensional rate model ( see Methods , section “Mean-field theory” ) , we obtain the analogous of the self-consistent equation for the power spectrum ( Eq 8 ) for the general case S x ( f ) = G ˜ ( f ) ( S ϕ ( x 1 ) ( f ) + S I ( f ) ) , ( 22 ) where Sϕ ( x1 ) ( f ) is the power spectrum of ϕ ( x1 ) , i . e . the mean-field firing rate . As in the case of adaptation , G ˜ ( f ) is the squared modulus of the linear response function of single neurons in the frequency domain ( see Methods ) . By solving the mean-field theory , we find that , similarly to the case of adaptation , for small coupling the power spectrum converges to zero for all frequencies . The critical value of the coupling g is defined implicitly by ( see Methods , section “Fixed point stability in the mean-field network” ) g c 2 max fG ˜ ( f ) = 1 . ( 23 ) On the other hand , for g > gc we find that the network , also in this more general case , exhibits fluctuating activity , whose power spectrum results from a sharpening of the single-neuron linear response function G ˜ ( f ) . In Methods , section “Qualitative study of the iterative map” we show how this property can be understood from the mean-field equations and we provide two examples of network power spectra for higher-dimensional rate models .
We see four extensions to the work presented in the present manuscript . First , our study is limited to rate neurons while it would be interesting to extend the analysis to spiking neuron models . As a first step in this direction , previous work has already investigated the introduction of white noise in random rate networks [2 , 47 , 59] , which would be straightforward to include in the case of D-dimensional rate units . Second , our framework can readily be extended to multiple adaptation variables ( see Methods , section “Qualitative study of the iterative map” for two examples ) . This is a key feature in order to account for realistic SFA , which is known to have multiple timescales and it has been shown to have power-law structure [16–18 , 41] . Interestingly , our framework can be extended to power-law adaptation , since we require only the knowledge of the linear frequency-response function of the single neurons . We expect that in this situation the internal noise generated by the network will also have a power-law profile of the type fα , with α > 0 . With such a noise spectrum , the signal that maximizes information transmission should be dominated by low-frequencies in a power-law fashion [18 , 70] . Third , the introduction of additional structure in the connectivity , such as low-rank perturbations [12] , attractor structure [71] , or large scale connectivity of the brain [20] , could give rise to interesting dynamics when combined with single units with multiple adaptation variables . In particular , the state of resonant chaos may also arise from an interaction of excitatory and inhibitory spiking neurons in networks with partially random , and partially structured connections . Finally , while our study focused on neural networks , random network models are used in other areas of biology and physics [72] . By extending mean-field theory techniques to more complex node dynamics , our approach also contributes to understanding the interaction between node dynamics and network structure in more general settings . We hypothesize that our approach can be used in the future to provide and understanding the variability of single-neuron activity across trials in the presence of one or several peaks in the power spectrum at gamma and theta frequencies .
The traditional approach in the DMFT literature is to consider the time-domain version of Eq 22 [1] . Applying the inverse Fourier transform to Eq 22 would lead to a differential equation of order 2D . Unfortunately , by contrast with the case D = 1 , for the multi-dimensional case D > 1 the dynamics is no longer conservative , which precludes the determination of the initial conditions ( see [1] ) . We propose an alternative approach to find a self-consistent solution to Eq 22 in the Fourier domain . This approach is based on an iterative map , the fixed point of which is the self-consistent solution . Iterative methods have been proposed previously both in the context of spiking [62 , 64] and rate-based networks [75] using Monte-Carlo methods . Here , we use a semi-analytical iteration method that allows to rapidly solve for the self-consistent power spectrum , and hence to qualitatively understand several features of the network dynamics . In the frequency domain , the linear transform associated with G ˜ ( f ) is simple , whereas the nonlinearity ϕ ( x ) is difficult to handle . Concretely , we need to express Sϕ ( x1 ) as a functional of Sx ( f ) . This calculation can be performed semi-analytically for the piecewise-linear nonlinearity ( a detailed treatment of the nonlinear step is given in Methods , section “Effect of nonlinearities on second-order statistic” ) . The idea of our iterative method is to start with an arbitrary initial power spectral density S ϕ ( x 1 ) ( 0 ) ( f ) , which we choose to be constant ( white noise ) . We then apply multiple iterations each consisting of a linear step followed by a nonlinear one ( Fig 2f ) . At each iteration , the linear step is simply a multiplication by g 2 G ˜ ( f ) and it allows us to compute ( Sx ) ( n+ 1 ) ( f ) . The nonlinear step afterwards transforms ( Sx ) ( n+1 ) ( f ) into S ϕ ( x 1 ) ( n + 1 ) ( f ) . By studying the iterative map that defines the mean-field solution , we conclude that the power spectrum of the network activity emerges from a sharpening of the linear response function G ˜ ( f ) of single units . The sharpening mainly arises from repeated multiplications with the factor g 2 G ˜ ( f ) in the iteration , which however is balanced by cross-frequency interactions and saturation effects of the nonlinear steps ( see Methods , section “Qualitative study of the iterative map” for a detailed discussion ) . As a result , the network activity exhibits the same frequency bands that are preferred by single neurons , albeit much narrower . For a qualitative understanding of the effect of the iterations on the power spectral density , we exploit the fact that x1 is a Gaussian process , for which the following formula holds [76] C ϕ ( x 1 ) ( τ ) = ∑ n = 0 ∞ 1 n ! ( ⟨ d n ϕ d ( x 1 ) n ⟩ ) 2 C x 1 n ( τ ) , ( 37 ) where the angular brackets indicate the mean over the statistics of x1 . Eq 37 gives the effect of a nonlinearity ϕ on a the autocorrelation of a Gaussian process x1 . By truncating the series after the first term , we get C ϕ ( x 1 ) ( τ ) ≃ ( ⟨ ϕ ′ ( x 1 ) ⟩ ) 2 C x 1 ( τ ) . ( 38 ) Fourier transforming this equation we get an approximation of the power spectral density of ϕ ( x1 ) S ϕ ( x 1 ) ( f ) ≃ Ψ 1 ( ∫ - ∞ ∞ S x ( f ′ ) d f ′ ) S x ( f ) , ( 39 ) where we we introduced the function Ψ 1 ( ∫ - ∞ ∞ S x ( f ′ ) d f ′ ) ≔ ( 〈 ϕ ′ ( x 1 ) 〉 ) 2 to highlight the fact that the coefficient that multiplies Sx ( f ) depends on the area under the power spectral density , i . e . on the variance of x1 , and is therefore nonlocal in frequency space . We stress that retaining only the first term in Eq 37 is different than considering a linear approximation of ϕ , since the dependence of the coefficient on the variance would not appear in that case . Using this approximation , we can express the power spectral density at the nth iteration of the iterative method , as a function of the initial power spectral density S ϕ ( x 1 ) ( 0 ) ( f ) from which we started to iterate . We obtain ( S x ) ( n ) ( f ) = ( ∏ k = 1 n - 1 Ψ 1 ( k ) ) ( g 2 G ˜ ( f ) ) n S ϕ ( x 1 ) ( 0 ) ( f ) , ( 40 ) where Ψ 1 ( n ) ≔ Ψ 1 ( ∫ - ∞ ∞ ( S x ) ( n ) ( f ′ ) d f ′ ) . If we take S ϕ ( x 1 ) ( 0 ) ( f ) to be constant and we define a n = ( ∏ k = 1 n - 1 Ψ 1 ( k ) ) , we can rewrite the above expression as ( S x ) ( n ) ( f ) = a n ( g 2 G ˜ ( f ) ) n . ( 41 ) If g > gc , there will be a range of frequencies for which g 2 G ˜ ( f ) > 1 , which implies that its nth power diverges when n grows . In a purely linear network , this phenomenon would lead to a blow-up of the power spectral density , in agreement with the fact that activity in a linear network is unbounded for g > gc . If ϕ is a compressive nonlinearity however , the coefficient an will tend to zero for growing n , counterbalancing the unbounded growth of ( g 2 G ˜ ( f ) ) n . Based on Eq 41 , we would predict that all the modes for which G ˜ ( f ) > 1 / g 2 will get amplified over multiple iterations , while all the other modes will get suppressed . While this is a highly simplified description , the suppression and the amplification of modes is clearly visible when observing the evolution of the power spectrum over iterations ( Fig 2f ) and when comparing the dynamics of the self-consistent solution ( Fig 8c and 8f , parameters in Table 1 ) to the corresponding linear response function ( Fig 8b and 8e , parameters in Table 1 ) . When truncating the series after the first order however , the mean-field network does not admit a self-consistent solution , for which we need to retain also higher order terms . Such terms will balance the progressive sharpening of the power spectrum , allowing for a self-consistent solution . As an example of higher-order term , consider the next term in the series in Eq 37 , given by 1 2 ( ⟨ ϕ ′ ′ ( x 1 ) ⟩ ) 2 ( C x 1 ( τ ) ) 2 → F T → F T 1 2 Ψ 2 ( ∫ - ∞ ∞ S x ( f ′ ) d f ′ ) ( S x * S x ) ( f ) ( 42 ) where Ψ2 is defined analogously to Ψ1 . In general , higher-order terms will contain convolutions of the power spectral density with itself , which are responsible for the creation of higher harmonics . To qualitatively understand this effect , consider the case in which Sx ( f ) is a Dirac δ-function with support in f0 . In this case , the two-fold convolution of Sx ( f ) with itself is again equal to a Dirac δ-function , but centered in 2f0 . A similar argument can be given for resonant power spectral densities , which implies that a self-consistent solution should exhibit harmonics of the fundamental resonance frequency . Note that in this paper we considered odd functions , for which only odd terms in the series are nonzero . For higher values of g , the relative importance of higher-order terms in the series in Eq 37 will increase , leading to a broader power spectrum . The self-consistent power spectrum however , seems to be always narrower than the single neuron linear response function . For a possible explanation of this phenomenon , we consider the g → ∞ limit , which was already studied in [46] for the network without adaptation . Using the same technique , we conclude that in this limit the autocorrelation decay tends to be the same as one obtained for a single unit driven by white noise [46] . In the frequency domain , this is equivalent to say that the power spectral density of the network tends to the one of a single unit driven by white noise . In this section , we provide some additional details on how to compute the effect of nonlinearities on the second order statistics ( autocorrelation or power spectral density ) of a Gaussian process . We consider three cases of interest: polynomials , piecewise linear functions and arbitrary nonlinear functions . To simplify our notation , we drop the superscript of and consider a generic Gaussian process x . The effect of polynomial nonlinearities can be expressed in closed form in time domain . This can be seen by considering again the infinite series expression ( Eq 37 ) , valid for stationary Gaussian processes x C ϕ ( x ) ( τ ) = ∑ n = 0 ∞ ( ⟨ d n ϕ d x n ⟩ ) 2 C x n ( τ ) , ( 43 ) where the angular brackets indicate the average over the statistics of x . In the case in which ϕ is a polynomial of degree p , only the terms in the sum up to p are nonzero . As an example , we can compute the effect of a cubic approximation of the hyperbolic tangent , i . e . ϕ ( x ) ≃ ϕ 3 ( x ) ≔ x - x 3 3 C ϕ 3 ( x ) ( τ ) = ( 1 + C x 2 ( 0 ) - 2 C x ( 0 ) ) C x ( τ ) + 2 3 C x 3 ( τ ) . ( 44 ) As expected , the effect of the nonlinearity depends on Cx ( 0 ) i . e . on the variance of x itself . Notice that the coefficient of the first term is compressive ( i . e . smaller than one ) only if Cx ( 0 ) is smaller than one itself . This type of behavior is expected since ϕ3 is unbounded . Another interesting case are piecewise linear nonlinearities . In this case , we use Price’s theorem twice to get ∂ 2 C ϕ ( x ) ( t ) ∂ ( C x ( t ) ) 2 = C ϕ ′ ′ ( x ) ( t ) . ( 45 ) For a piecewise linear ϕ , the second derivative ϕ′′ is a sum of Dirac’s delta functions with variable coefficients . More precisely , we consider ϕ P L ( x ) = Θ ( x 1 - x ) c 0 x + ∑ p = 1 P - 1 Θ ( x - x p ) Θ ( x p + 1 - x ) c p x p + Θ ( x - x P ) c P x , ( 46 ) where xp are the points in which the first derivative is discontinuous , cp are some arbitrary coefficients and Θ ( ⋅ ) is the Heaviside function . The second derivative of ϕPL is given by ϕ P L ′ ′ ( x ) = ∑ p = 1 P ( c p - c p - 1 ) δ ( x - x p ) . ( 47 ) The delta functions allow us to compute the correlation function C ϕ P L ′ ′ ( t ) explicitly C ϕ P L ′ ′ ( t ) = ∑ p , p ′ = 1 P ( c p - c p - 1 ) ( c p ′ - c p ′ - 1 ) 2 π C x ( 0 ) 1 - ρ 2 ( t ) × × exp ( - x p 2 + x p ′ 2 - 2 ρ ( t ) x p x p ′ 2 C x ( 0 ) ( 1 - ρ 2 ( t ) ) ) , ( 48 ) where we defined ρ ( t ) ≔ C x ( t ) C x ( 0 ) . Inserting Eq 48 in Eq 45 and integrating twice with respect to Cx ( t ) we get C ϕ P L ( x ) ( t ) = f ϕ ( 0 ; C x ( 0 ) ) + f ϕ ′ ( 0 ; C x ( 0 ) ) C x ( t ) + ∑ p , p ′ = 1 P ∫ 0 C x ( t ) ∫ 0 σ ′ ( c p - c p - 1 ) ( c p ′ - c p ′ - 1 ) 2 π C x ( 0 ) 1 - σ 2 C x 2 ( 0 ) × × exp ( - x p 2 + x p ′ 2 - 2 σ C x ( 0 ) x p x p ′ 2 C x ( 0 ) ( 1 - σ 2 C x 2 ( 0 ) ) ) d σ d σ ′ . ( 49 ) In the case in which ϕ is an odd function , the term fϕ ( 0; Cx ( 0 ) ) is equal to zero . For the specific case of the piecewise linear approximation of the hyperbolic tangent considered in this paper , i . e . ϕ P L ( x ) = { - 1 for x < - 1 x for - 1 < x < 1 1 for x > 1 , ( 50 ) the expression in Eq 49 reduces to C ϕ P L ( x ) ( t ) = Erf 2 ( 1 2 C x ( 0 ) ) C x ( t ) + 2 π C x ( 0 ) ∫ 0 C x ( t ) ∫ 0 σ ′ 1 1 - σ 2 C x 2 ( 0 ) × × exp ( - 1 C x ( 0 ) ( 1 - σ 2 C x 2 ( 0 ) ) ) sinh ( σ C x 2 ( 0 ) ( 1 - σ 2 C x 2 ( 0 ) ) ) d σ d σ ′ . ( 51 ) For the piecewise linear function , an alternative approach is based on the infinite series in Eq 37 , which yields [77 , 78]: C ϕ P L ( x ) ( t ) = σ 2 ∑ n = 1 ∞ [ F ( n - 1 ) ( 1 σ ) - F ( n - 1 ) ( - 1 σ ) ] 2 C x n ( t ) n ! ( 52 ) with input variance σ2 = Cx ( 0 ) and cumulative Gaussian distribution function F ( x ) = 1 2 π ∫ - ∞ x e - y 2 / 2 d y . For the figures in this paper , we used the map in Eq 51 . For an arbitrary nonlinear function , we can use two methods . The first method is a semi-analytical approach that relies on the integral form of the autocorrelation of the rate Cϕ ( x ) ( τ ) as a functional of the autocorrelation Cx ( τ ) of x [45] C ϕ ( x ) ( τ ) = ∫ ∫ ϕ ( C x ( 0 ) - C x 2 ( τ ) C x ( 0 ) x + C x ( τ ) C x ( 0 ) z ) ϕ ( C x ( 0 ) z ) D x D z , ( 53 ) where 12π . Notice that a slightly different version of this formula was already proposed in [1] . Therefore , to obtain the effect of ϕ on the power spectral density , one should 1 ) inverse Fourier transform Sx ( f ) to get Cx ( τ ) 2 ) apply Eq 53 , by computing the two integrals numerically 3 ) Fourier transform Cϕ ( x ) ( τ ) to get Sϕ ( x ) ( f ) . Practically , this procedure requires the application of the fast Fourier transform algorithm and the numerical evaluation of two integrals . The second method is fully numerical and it can be useful in cases in which the integrals in the first method are expensive to evaluate numerically . This method consists in approximating the power spectral density Sϕ ( x ) via Monte Carlo sampling . More precisely , we sample multiple realizations in frequency domain of the Gaussian process with zero mean and power spectral density Sx ( f ) . We then transform each sample to time domain and apply the nonlinearity ϕ ( x ) to each sample x ( t ) individually . Finally , we transform back to Fourier domain and get Sϕ ( x ) by averaging . Despite being computationally more expensive than the closed form expressions , this sampling method provides a solution of the mean-field theory for an arbitrary nonlinearity and it is computationally much cheaper than running the full microscopic simulation . Moreover , this method can easily be extended to be used in the presence of a non-Gaussian sinusoidal input ( cf . section “Adaptation shapes signal transmission in the presence of internally-generated noise” and [4] ) . In this section , we extend the derivation of dynamical mean-field theory ( DMFT ) to the case of the network of multi-dimensional rate units . Since there are no additional complication with respect to the standard case , we report here only the main steps . For a review of the path-integral approach to DMFT , see e . g . [45 , 46] . The moment-generating functional corresponding to the microscopic system in Eq 21 is given by Z [ j , j ˜ ] ( J ) = ∫ D x D x ˜ exp [ S 0 [ x , x ˜ ] - ( x ˜ 1 ) T J ϕ ( x 1 ( t ) ) + j T x + j ˜ T x ˜ ] , ( 54 ) where S 0 [ x , x ˜ ]≔x ˜ T ( I D ∂ t - A ) x ( 55 ) and we introduced the notation x ˜ T x = ∑ α ∑ i ∫ x ˜ i α ( t ) x i α ( t ) d t . The integral is over paths and bold symbols indicate vectors , over both the network space and the rate model space , so that D x≔∏ α ∏ i D x i α . We are interested in properties that are independent of the particular realization of the coupling matrix J . In order to extract those properties , we average over the quenched disorder by defining the averaged generating function Z ¯ [ j , j ˜ ]≔∫ ∏ i j d J i j N ( 0 , g 2 N ; J i j ) Z [ j x , j ˜ x ] ( J ) . ( 56 ) The average over each Jij can be computed by noticing that the terms corresponding to different Jij factorize and the integral can be solved by completing the square . Since the details of this calculation are analogous to the one-dimensional case , we directly report the result Z ¯ [ j x , j ˜ x ] = ∫ D x D x ˜ exp [ S 0 [ x , x ˜ ] + j T x + j ˜ T x ˜ ] × × exp [ 1 2 ∫ - ∞ ∞ ( ∑ i x ˜ i 1 ( t ) x ˜ i 1 ( t ′ ) ) ( g 2 N ∑ j ϕ ( x j 1 ( t ) ) ϕ ( x j 1 ( t ′ ) ) ) d t d t ′ ] . ( 57 ) We now aim to decouple the interaction term in the last line by introducing the auxiliary field Q 1 ( t , s ) g 2 N ∑ j ϕ ( x j 1 ( t ) ) ϕ ( x j 1 ( s ) ) . ( 58 ) We introduce Q1 in the generating functional by inserting the following representation of the unity ∫ D Q 1 δ [ - N g 2 Q 1 ( s , t ) + ∑ j ϕ ( x j 1 ( s ) ) ϕ ( x j 1 ( t ) ) ] , ( 59 ) where δ[⋅] is the delta functional . Using the integral representation of the delta functional leads to the introduction of a second auxiliary field , which we call Q2 . We obtain Z ¯ [ j x , j ˜ x ] = ∫ D Q 1 D Q 2 D x D x ˜ exp [ S 0 [ x , x ˜ ] + j T x + j ˜ T x ˜ ] × exp [ 1 2 ∫ - ∞ ∞ ( ∑ i x ˜ i 1 ( t ) Q 1 ( t , t ′ ) x ˜ i 1 ( t ′ ) + ∑ i ϕ ( x i 1 ( t ) Q 2 ( t , t ′ ) ϕ ( x i 1 ( t ′ ) ) + - N g 2 Q 1 ( t , t ′ ) Q 2 ( t , t ′ ) ) d t d t ′ ] . ( 60 ) This expression has the advantage that any interaction between different units is removed and all the contribution coming from different units factorize . It is convenient to rewrite the averaged generating functional as a field theory for two auxiliary fields Q1 , Q2 , i . e . we remove the vectorial response terms j T x , j ˜ T x ˜ and we add two scalar response terms for the auxiliary fields . The result is Z ¯ [ j , j ˜ ] = ∫ D Q 1 D Q 2 exp ( - N g 2 Q 1 T Q 2 + N ln Z [ Q 1 , Q 2 ] + j T Q 1 + j ˜ T Q 2 ) Z [ Q 1 , Q 2 ]≔∫ D x D x ˜ exp ( S 0 [ x , x ˜ ] + 1 2 ( x ˜ 1 ) T Q 1 x ˜ 1 + ϕ ( x 1 ) T Q 2 ϕ ( x 1 ) ) , ( 61 ) where we extended our notation to Q 1 T Q 2≔∫ ∫ Q 1 ( s , t ) Q 2 ( s , t ) d s d t . The crucial observation to make is that essentially all factors associated to different units factorized yielding the factor N . For this reason , the integration is now not over all rate model indices but over only one unit index . The remainder is the problem of one unit , characterized by D variables , interacting with two external fields Q1 , Q2 . The final step is to perform a saddle-point approximation , i . e . replace Q1 , Q2 by their values that make the action stationary . To do this , we need to solve the two saddle-point equations δ δ Q { 1 , 2 } ( N g 2 Q 1 T Q 2 + N ln Z [ Q 1 , Q 2 ] ) = 0 ( 62 ) These equations are analogous to the ones in the one-dimensional case , and lead to the saddle-point solution Q 1 * ( s , t ) = g 2 C ϕ ( x 1 ) ( s , t ) Q 2 * ( s , t ) = 0 , ( 63 ) where Cϕ ( x1 ) ( s , t ) is the autocorrelation function of ϕ ( x1 ) evaluated at the saddle point solution . The averaged generating functional at the leading order in N can be written as Z ¯ * ∝ ∫ D x D x ˜ exp ( S 0 [ x , x ˜ ] + g 2 2 ( x ˜ 1 ) T C ϕ ( x 1 ) x ˜ 1 ) . ( 64 ) This is the statistical field theory corresponding to D linearly interacting variables , with x1 that receives a Gaussian noise whose autocorrelation is given by Cϕ ( x1 ) . Writing the corresponding differential equations results in our mean-field description ( Eq 29 ) . In this section , we will extend the derivation of the dynamic mean-field theory ( DMFT ) for the case of the network with heterogeneous adaptation . We consider the case in which each neuron has different parameters , sampled i . i . d from the same distributions , and different parameters of the same neuron are uncorrelated with each other . More precisely , we sample the elements of the matrix Ai for neuron i as A i α β ∼ N ( A ¯ α β , ( σ α β ) 2 ) , ( 65 ) where the subscript i runs over the neurons in the network . In deriving the mean-field theory , most of the steps are identical to those in Methods , section “Mean-field theory derivation” , so we will focus on the additional terms due to the new source of disorder . We separate the contribution of mean adaptation parameters A ¯ α β from the deviations , so that the generating functional reads Z [ j , j ˜ ] ( J ) = ∫ D x D x ˜ exp [ S 0 [ x , x ˜ ] - ( x ˜ 1 ) T J ϕ ( x 1 ( t ) ) - ∑ k x k T ( A k - A ¯ ) x k ++ j T x + j ˜ T x ˜] , ( 66 ) where S 0 [ x , x ˜ ]≔x ˜ T ( I D ∂ t - A ¯ ) x ( 67 ) and A ¯ is the matrix of the expected values of A . The action S0 is the same as for the network without heterogeneity , and when averaging over the connectivity disorder , we obtain the same result as for homogeneous network . In this case however , we need to also average over the disorder due to heterogeneity , i . e . over all the A k α β . The averaged generating functional will then result from the average Z ¯ [ j , j ˜ ] ≔ ∫ ( ∏ i j d J i j N ( 0 , g 2 N ; J i j ) ) ( ∏ α β k d A k α β N ( A ¯ k α β , σ α β ; A α β ) ) Z [ j , j ˜ ] ( J ) . ( 68 ) The new terms due to the heterogeneity result in integrations of the type 1 2 π ( σ α β ) 2 ∫ exp ( - 1 2 ( σ α β ) 2 ( A α β - A ¯ α β ) 2 - ( A α β - A ¯ α β ) ∫ x i ˜ α ( t ) x i β ( t ) d t ) , ( 69 ) that can be solved by completing the square . After averaging over both the connectivity disorder and the heterogeneity disorder , the generating functional reads Z ¯ [ j x , j ˜ x ] = ∫ D x D x ˜ exp [ S 0 [ x , x ˜ ] + j T x + j ˜ T x ˜ ] × × exp [ 1 2 ∫ ( ∑ i x ˜ i 1 ( t ) x ˜ i 1 ( t ′ ) ) ( g 2 N ∑ j ϕ ( x j 1 ( t ) ) ϕ ( x j 1 ( t ′ ) ) ) d t d t ′ ] × × exp [∑ i α β ( σ α β ) 2 2 ∫ x ˜ i α ( t ) x i β ( t ) x i β ( t ′ ) x ˜ i α ( t ′ ) ] . ( 70 ) The last term , which is due to the heterogeneity , factorizes into the contributions associated to different units . From this point on , in order to derive the mean-field equations , we follow exactly the same steps as in Methods , section “Mean-field theory derivation” , so we do not report those steps here . The mean-field equations read x ˙ α ( t ) = ∑ β = 1 D ( A ¯ α β x β ( t ) + η H α β ( t ) ) + δ α 1 ( η ( t ) + I ( t ) ) , ( 71 ) where η H α β are Gaussian processes associated to the heterogeneity , that all have mean zero and autocorrelation ⟨ η H α β ( t ) η H α β ( s ) ⟩ = ( σ α , β ) 2 ⟨ x β ( t ) x β ( s ) ⟩ . ( 72 ) For the particular case of adaptation with heterogeneity on the parameter β , as studied in section “Resonant chaos in random networks with adaptation” , we have the following mean-field equations x ˙ ( t ) =- x ( t ) - a ( t ) + η ( t ) + I ( t ) ( 73 ) a ˙ ( t ) =- γ a ( t ) + γ β ¯ x ( t ) + γ η H ( t ) , ( 74 ) where ηH ( t ) is a Gaussian process with mean zero and autocorrelation ⟨ η H ( t ) η H ( s ) ⟩ = σ β 2 ⟨ x ( t ) x ( s ) ⟩ . ( 75 ) From Eqs 73 and 74 , we can find the self-consistent equation for the power spectrum: S x ( f ) = G ˜ H ( f ) ( g 2 S ϕ ( x ) ( f ) + S I ( f ) ) , ( 76 ) where G ˜ H ( f ) is an effective filter given by G ˜ H ( f ) = G ˜ ( f ) 1 - γ 2 σ β 2 γ 2 + ω 2 G ˜ ( f ) , ( 77 ) where ω = 2πf . The effective filter G ˜ H ( f ) predicts a larger power at low frequencies , similar to what is observed in simulations ( cf . Fig 2d ) . Here we consider the full matrix of linear response functions ( see below ) , to conclude that the only quantity that matters for the stability at the fixed point is G ˜ ( f ) . Starting from the microscopic network equations ( Eq 21 ) , we derive a set of differential equations , that we write in matrix form ( I D ∂ τ - A ) χ i k ( τ ) = ∑ j = 1 N J i j Δ 1 χ j k ( τ ) + δ i k I D δ ( τ ) , ( 78 ) where Δ1 = δα1 δβ1 is a matrix whose only nonzero element is [Δ1]11 = 1 . χik ( τ ) is a D by D matrix , whose component are defined as χ i k α β ( τ ) = δ x i α ( τ ) δ h k β ( 0 ) , where h k β is a small perturbation given to the variable x k β at time τ = 0 . Notice that in deriving Eq 78 , we have assumed stationarity and that ϕ′ ( 0 ) = 1 . We now Fourier transform Eq 78 and get ( 2 π i f I D - A ) χ ˜ i k ( f ) = ∑ j = 1 N J i j Δ 1 χ ˜ j k ( f ) + δ i k I D . ( 79 ) Inverting the matrix ( 2πif ID − A ) and recognizing the linear response function of the single unit χ ˜ 0 ( f ) , we obtain χ ˜ i k ( f ) = ∑ j = 1 N J i j χ ˜ 0 ( f ) Δ 1 χ ˜ j k ( f ) + δ i k χ ˜ 0 ( f ) , ( 80 ) where χ ˜ 0 ( f ) is a D by D matrix whose elements are χ ˜ 0 α β ( f ) , defined in section “Mean-field theory” . Since in the mean-field approximation the mean of the linear response function is zero , we look for the second moments [2] . We multiply every element of the matrix equation ( Eq 80 ) by its complex conjugate and average over the quenched disorder . We obtain | χ ˜ ( f ) | 2 = g 2 | χ ˜ 0 ( f ) Δ 1 χ ˜ ( f ) | 2 + G ˜ ( f ) , ( 81 ) where the absolute value is intended element-wise . Due to the structure of the matrix Δ1 , we have that | χ ˜ 0 ( f ) Δ 1 χ ˜ ( f ) | 2 = G ˜ ( f ) Δ 1 | χ ˜ ( f ) | 2 , as it can be verified simply by using the definition of Δ1 . Finally , we can solve for | χ ˜ ( f ) | 2 | χ ˜ ( f ) | 2 = ( I D - g 2 G ˜ ( f ) Δ 1 ) - 1 ( G ˜ ( f ) ) . ( 82 ) Since the only nonzero eigenvalue of the matrix G ˜ ( f ) Δ 1 is | χ ˜ 0 11 ( f ) | 2 , the stability condition for the fixed point is given by g 2max f G ˜ ( f ) < 1 . ( 83 ) | Biological neural networks are formed by a large number of neurons whose interactions can be extremely complex . Such systems have been successfully studied using random network models , in which the interactions among neurons are assumed to be random . However , the dynamics of single units are usually described using over-simplified models , which might not capture several salient features of real neurons . Here , we show how accounting for richer single-neuron dynamics results in shaping the network dynamics and determines which signals are better transmitted . We focus on adaptation , an important mechanism present in biological neurons that consists in the decrease of their firing rate in response to a sustained stimulus . Our mean-field approach reveals that the presence of adaptation shifts the network into a previously unreported dynamical regime , that we term “resonant chaos” , in which chaotic activity has a strong oscillatory component . Moreover , we show that this regime is advantageous for the transmission of low-frequency signals . Our work bridges the microscopic dynamics ( single neurons ) to the macroscopic dynamics ( network ) , and shows how the global signal-transmission properties of the network can be controlled by acting on the single-neuron dynamics . These results paves the way for further developments that include more complex neural mechanisms , and considerably advance our understanding of realistic neural networks . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [
"resonance",
"frequency",
"neural",
"networks",
"engineering",
"and",
"technology",
"signal",
"processing",
"signaling",
"networks",
"neuroscience",
"research",
"design",
"mathematics",
"algebra",
"network",
"analysis",
"computational",
"neuroscience",
"white",
"noise",
"... | 2019 | How single neuron properties shape chaotic dynamics and signal transmission in random neural networks |
Anthocyanin is part of secondary metabolites , which is induced by environmental stimuli and developmental signals , such as high light and sucrose . Anthocyanin accumulation is activated by the MYB-bHLH-WD40 ( MBW ) protein complex in plants . But the evidence of how plants maintain anthocyanin in response to signals is lacking . Here we perform molecular and genetic evidence to display that HAT1 plays a new breaker of anthocyanin accumulation via post-translational regulations of MBW protein complex . Loss of function of HAT1 in the Arabidopsis seedlings exhibits increased anthocyanin accumulation , whereas overexpression of HAT1 significantly repressed anthocyanin accumulation . We found that HAT1 interacted with MYB75 and thereby interfered with MBW protein complex . Overexpression of HAT1 suppresses abundant anthocyanin phenotype of pap1-D plant . HAT1 is characterized as a transcriptional repressor possessing an N-terminal EAR motif , which determines to interact with TOPLESS corepressor . Repression activity of HAT1 in regulation of gene expression and anthocyanin accumulation can be abolished by deletion or mutation of the EAR motif 1 . Chromatin immunoprecipitation assays revealed that MYB75 formed a transcriptional repressor complex with HAT1-TPL by histone H3 deacetylation in target genes . We proposed that HAT1 restrained anthocyanin accumulation by inhibiting the activities of MBW protein complex through blocking the formation of MBW protein complex and recruiting the TPL corepressor to epigenetically modulate the anthocyanin late biosynthetic genes ( LBGs ) .
Anthocyanins , one kind of flavonoids , are vital secondary metabolites widespread throughout the plant kingdom [1] . As water-soluble pigments , anthocyanins confer widest colors to flowers , leaves , and fruits [2] . Anthocyanin accumulation is stimulated by a variety of endocellular signals such as sucrose and phytohormone [3 , 4] , as well as exogenous environmental stresses including high light [5] , drought [6] , and nutrient depletion [7] . Anthocyanins can protect plants against excessive light [8] and drought [9] , and defend from invasion by pathogens and herbivores [10] . Anthocyanin biosynthesis is derived from flavonoid synthetic pathway which is composed of multiple enzymes encoded by biosynthetic genes . Initially , the early flavonoid reactions catalyzed by early biosynthetic genes ( EBGs ) included chalcone synthase ( CHS ) , chalcone isomerase ( CHI ) , and flavonol 3-hydroxylase ( F3H ) , which are regulated by three redundant R2R3 MYB transcription factors ( TFs ) MYB11 , MYB12 , and MYB111 [11] . Then the expression of anthocyanin-specific biosynthetic genes encoding dihydroflavonol-4-reductase ( DFR ) , leucoanthocyanidin dioxygenase ( LDOX ) , and UDP-glucose: flavonoid-3-O-glycosyl-transferase ( UF3GT ) is regulated by the ternary MYB-bHLH-WD40 ( MBW ) protein complex , which is composed of R2R3-MYB , basic helix-loop-helix ( bHLH ) , and WD40-repeat proteins [2 , 12] . In Arabidopsis , the identified R2R3-MYB transcription factors include PRODUCTION OF ANTHOCYANIN PIGMENTATION 1 ( PAP1 ) /MYB75 , PAP2/MYB90 , MYB113 , and MYB114 [13 , 14] . The bHLH transcription factors include TRANSPARENT TESTA 8 ( TT8 ) and ENHANCER OF GLABRA 3 ( EGL3 ) , and and only one WD40-repeat protein , TRANSPARENT TESTA GLABRA 1 ( TTG1 ) , has been identified [14–16] . Notably , some MYB TFs such as TRANSPARENT TESTA 2 ( TT2 ) , GLABRA1 ( GL1 ) , and WEREWOLF ( WER ) act as a component of MBW protein complex to transcriptionally regulate expression of multiple gene involved in proanthocyanin accumulation , trichome development , and root epidermal cell fate in Arabidopsis thaliana [17–19] . Recent studies demonstrated that post-translational modification of MBW proteins modulated the transcriptional activity of MBW protein complex . Degradation of MYB75 in the dark was mediated by E3 ubiquitin ligase COP1 [20] , while MPK4 phosphorates MYB75 and increases its stability in response to light [5] . Ubiquitin protein ligase 3 ( UPL3 ) regulates anthocyanin and trichome development by mediating the proteasomal degradation of GL3 and EGL3 [21] . GSK3-like kinase BIN2 controls root epidermal cell fate through suppressing the activity of MBW protein complex via phosphorylating both TTG1 and EGL3 [22] . The other post-translational regulation is preventing the formation of MBW protein complex . Single repeat R3-MYB transcription factors , including MYBL2 , CAPRICE ( CPC ) , TRIPTYCHON ( TRY ) , ENHANCER OF TRY AND CPC 1 ( ETC1 ) , and ETC2 , suppress anthocyanin accumulation or trichome initiation by disturbing the formation of MBW protein complex [23–27] . SPL9 , a member of SQUAMOSA PROMOTER BINDING PROTEIN-LIKE family , negatively regulates anthocyanin accumulation by directly preventing expression of anthocyanin biosynthetic genes via destabilization of MBW complex [28] . Similarly , the JA-ZIM domain ( JAZ ) proteins suppress anthocyanin accumulation and trichome development by disturbing the MBW protein complex [4] . Additionally , MYBL2 and PhMYB27 transform the MBW complex from an activator to a repressor by replacing one of the R2R3-MYB components of MBW protein [23 , 24 , 29] . The transformation possibly depends on post-translational regulation . However , details of the regulation mechanism remain unclear . Anthocyanin accumulation in plants is modulated by light conditions [20] . Plants accumulate less anthocyanin under shade conditions [29] . It is well established that the class II homeodomain-leucine zipper family participate in shade avoidance and their expression is rapidly induced by shade conditions [30] . The members of class II homeodomain-leucine zipper family all contain a conserved DNA-binding homeodomain ( HD ) that is closed to a leucine zipper motif ( LZ ) , which is considered important to promote homo- or heterodimerization of HD-Zip protein [31 , 32] . The CPSCE motif adjacent to LZ motif is comprised of five conserved amino acids Cys , Pro , Ser , Cys , and Glu . This motif is thought to form high molecular weight multimers through intermolecular Cys-Cys bridges under oxidant environment , which can not possibly be transported to the nucleus to play its role [33] . It was reported that the class II HD-Zip proteins participated in modulation of plant development and multiple stress response . HAT2 is strongly induced by auxin and affects lateral root and hypocotyl elongation [34] . HAT3 and ATHB4 impact the leaf polarity in Arabidopsis thaliana by repressing MIR165/166 expression [35] . ATHB4 and HAT1 participate in Brassinosteroid signaling [36 , 37] . ATHB17 is involved in ABA response and plays an important role in protecting plants by adjusting expression of PhANGs and PEGs in response to abiotic stresses [38 , 39] . To explore the regulatory mechanisms in anthocyanin accumulation , here we identify a new regulator of MBW protein complex . HAT1 interferes with the formation of MBW protein complex by interacting with MYB75 . In Arabidopsis , TOPLESS ( TPL ) and TOPLESS RELATED ( TPR ) proteins generally mediate transcriptional repression in numerous pathways such as auxin and jasmonate signaling [40–42] , as well as developmental pathways including leaf polarity and flowering time regulation [43 , 44] . Meanwhile , we reveal that TPL can interact with HAT1 . We propose that HAT1 represses anthocyanin accumulation by inhibiting the activities of MBW protein complex through recruiting the TPL/TPR corepressors and histone modification . Our study exposes that HAT1 acts as a new regulator of anthocyanin accumulation , and proffers a mechanism for repression of anthocyanin accumulation .
Our previous research has proved that HAT1 participated in drought response [45] , and we noticed that transgenic plants overexpressing HAT1 ( 35S:HAT1 ) showed less anthocyanin accumulation compared with wild-type plants under drought stress . In Arabidopsis , the levels of anthocyanin accumulation are dependent on light intensity [46] . To understand how HAT1 influences anthocyanin accumulation , we maintained Arabidopsis seedlings under weak light of 40 μmol m-2 s-1 ( hereafter called Control ) , a control light intensity at which wild-type plants accumulate very low levels of anthocyanins [47] . To induce anthocyanin biosynthesis , seedlings were then shifted to moderate high light ( 180 μmol m-2 s-1 , hereafter called high light ) . Exposed to high light , two independent lines overexpressing HAT1 ( 35S:HAT1 #11 and #13 ) accumulated less anthocyanin than that of wild-type plants , while hat1 mutant showed higher anthocyanin accumulation ( Fig 1A and 1B ) . We also detected the expression levels of anthocyanin-specific biosynthetic genes , DFR , LDOX and UF3GT . The expression of these genes was also reduced in 35S:HAT1 plants but induced in the hat1 mutant ( Fig 1C–1E ) . Additionally , We also noticed that HAT1 transcription was notably repressed under high light conditions ( Fig 1F ) . 35S:HAT1 #13 transgenic plants were selected for subsequent experiment because the similar phenotype was observed between 35S:HAT1 #11 and 35S:HAT1 #13 transgenic plants . As shown in S1 Fig , 35S:HAT1 #13 transgenic plants grown in soil exhibited less anthocyanin accumulation compared with wild-type under high light conditions . Sucrose can specifically induce anthocyanin biosynthesis in Arabidopsis , thus we further investigate the anthocyanin accumulation in Arabidopsis seedlings grown with exogenous sucrose [3] . Similarly , the anthocyanin content of 35S:HAT1 #13 was significantly lower than that of wild-type under sucrose treatment ( S2A and S2B Fig ) . Compared with wild-type , DFR transcripts were decreased in 35S:HAT1 #13 but increased in hat1 mutant . LDOX and UF3GT showed the similar expression pattern to that of DFR ( S2C Fig ) . In Arabidopsis , some transcription factors modulate the expression of anthocyanin biosynthetic genes . We further examined the expression of several transcription factors including MYB75 , MYB90 , TT8 , EGL3 and TTG1 , which are characterized as regulators of anthocyanin biosynthetic genes ( S2D Fig ) . The expression levels of MYB75 and MYB90 obviously decreased in 35S:HAT1 #13 when compared with wild-type , while a nearly 1 . 5-fold increased was recorded in hat1 mutant than that in the wild-type ( S2D Fig ) . The transcript levels of TT8 and EGL3 were moderately increased in hat1 mutant ( S2D Fig ) . In summary , these results suggest that HAT1 may play as a negative regulator in anthocyanin biosynthetic pathway . To clarify how HAT1 regulates anthocyanin accumulation , we performed a yeast two-hybrid screen to identify its potential interaction partners . After screening , we identified MYB75 ( At1g56650 ) , an R2R3-MYB transcription factor , as its partner . Directed yeast two-hybrid assays validated that HAT1 interacted with MYB75 , but not with MYB90 , the paralog of MYB75 ( Fig 2A , S3 Fig ) . To further determine the domains required for the interaction , truncated HAT1 and MYB75 were used . As shown in Fig 2B , C-terminal fragment of HAT1 was required for the interaction . The R2 domain of MYB75 could strongly interact with HAT1 , but the C-terminal fragment of MYB75 weakly interacted with HAT1 ( Fig 2B ) . To verify whether HAT1 interacts with MYB75 in vivo , the bimolecular fluorescence complementation ( BiFC ) assay was performed for analysis . When MYB75-nYFP was coinfiltrated with HAT1-cYFP in tobacco ( Nicotiana benthamiana ) leaves , strong YFP fluorescence was observed in the nuclei ( Fig 2C ) . Further , the interaction between HAT1 and MYB75 was also confirmed by a coimmunoprecipitation ( Co-IP ) assay ( Fig 2D ) . These results suggest that HAT1 can interact with MYB75 in vivo . Previous evidence demonstrated that ternary MYB75-TT8/EGL3-TTG1 protein complex can activate the expression of LBGs [14] . The results above suggest that HAT1 interacts with MYB75 and represses the transcripts levels of LBGs , hence , we speculate that HAT1 competed with bHLH proteins for interaction with MYB75 since no interaction between HAT1 and TT8/EGL3 was detected by BiFC assay ( S4 Fig ) . To test this hypothesis , TT8-nYFP and MYB75-cYFP were transiently coexpressed in N . benthamiana leaves . As shown in Fig 3A , the strong YFP fluorescence were detected in N . benthamiana leaves , which is consistent with the previous study [48] . When HAT1-FLAG was coexpressed with TT8-nYFP and MYB75-cYFP , the fluorescence signal was visibly impaired ( Fig 3A and 3B ) , whereas coexpression of empty vector ( FLAG ) with TT8-nYFP/MYB75-cYFP did not reduce the fluorescence intensity . Similar results were observed when HAT1-FLAG was coexpressed with EGL3-nYFP and MYB75-cYFP ( Fig 3A and 3C ) . The expression of MYB75 , TT8 and EGL3 exhibited no obvious difference in these combination respectively ( Fig 3D and 3E ) . In vitro competitive binding assays demonstrated that interaction between HIS-MYB75 and MBP-TT8 was gradually impaired by an increased amount of HIS-HAT1 ( Fig 3F ) . Likewise , HIS-HAT1 also weakened the interaction between HIS-MYB75 and MBP-EGL3 ( Fig 3F ) . Using protoplasts from 35S:HAT1 #13 , we found that the interaction between TT8/EGL3 and MYB75 is counteracted by endogenous HAT1-GFP ( Fig 3G ) . We further tested whether HAT1 interferes with the interaction between MYB75 and TT8 or EGL3 in a yeast three-hybrid assay . When yeast transformant that carried both AD-MYB75 and pBridge-TT8-HAT1 plasmids were grown on plates with high methionine concentrations ( 200 μM ) , which restrain the expression of HAT1 , MYB75 strongly interacted with TT8 ( Fig 3H ) . When the level of methionine was reduced from 200 μM to 50 μM , hence permitting HAT1 expression , yeast growth were consumingly suppressed in these transformants ( Fig 3H ) . Similar results were observed when yeast transformed with AD-MYB75 and pBridge-EGL3-HAT1 ( Fig 3H ) . In summary , these results prove that HAT1 interferes with the formation of MBW protein complex . Next the transgenic line pap1-D overexpressing MYB75 was crossed with 35S:HAT1#13 to generate the pap1-D 35S:HAT1 #13 plants . As expected , overexpression HAT1 repressed anthocyanin accumulation in pap1-D background under control or high light conditions ( Fig 4A and 4B ) . We also crossed hat1 null mutant with myb75-c that MYB75 knockout mutant using the CRISPR-Cas9 system . Under high light conditions , the myb75-c hat1 double mutant exhibited similar phenotype with myb75-c . Consistent with the phenotype , transcript levels of anthocyanin biosynthetic genes DFR and LDOX were also lower in pap1-D 35S:HAT1 #13 than that of pap1-D mutant under normal or high light conditions , although the expression levels of HAT1 in pap1-D 35S:HAT1 #13 were similar with pap1-D ( Fig 4C–4E ) . Additionally , the anthocyanin content of the pap1-D hat1 exhibited no significant difference than that of pap1-D under high light conditions ( S5A and S5B Fig ) , it might be due to high light repressed HAT1 expression in pap1-D plants ( Fig 1D ) . Meanwhile , 35S:HAT1 #13 and 35S:HAT1 #13 myb75-c showed a similar anthocyanin accumulation phenotype ( S5C and S5B Fig ) . Taken together , our results suggest that HAT1 represses anthocyanin accumulation through interacting with MYB75 and at least partially by interfering with the formation of MBW protein complex . Although we have proved that HAT1 represses anthocyanin accumulation by sequestering the formation of MBW protein complex , we hypothesize that MYB75-HAT1 complex behaves as a repressor because overexpressing HAT1 in pap1-D represses anthocyanin accumulation when compared with pap1-D mutant . The previous study demonstrated that the members of HD-Zip II family act as a repressor in modulation of gene expression [31] . HAT1 protein possesses a leucine zipper motif ( LZ , between amino acids residue 190 and 233 ) followed by a DNA-binding homeodomain ( HD , between amino acids residue 134 and 188 ) ( Fig 2B ) . Interestingly , two typical ERF-associated-amphiphilic repression ( EAR ) motif ( DLGLSL and LQLNLK ) , which is proved as repression motif to repress transcription [49] , are located at the N-terminal end of the HAT1 protein ( Fig 5A ) . Many transcriptional repressors have been proved to regulate plant developmental process and signaling pathways by interacting with corepressor TPL/TRRs via EAR motif [40 , 41 , 43] . This promoted us to investigate whether HAT1 interacted with TPL/TRRs . Firstly , yeast two-hybrid assay showed that HAT1 interacted with TPL ( Fig 5B ) . Furthermore , the interaction relied on the HAT1 EAR motif 1 , because replacement of the three conserved Leu residues into Ala residues ( LxLxL to AxAxA , designated as HAT1mEAR ) abolished the interaction ( Fig 5B ) . We also found that HAT1 interacted with TPR3 , but not with TPR1 , TPR2 and TPR4 ( S6 Fig ) . Next we used Co-IP experiments to further check the interaction . Consistent with the yeast two-hybrid assays , Co-IP assays proffered that HAT1 interacts with TPL in plants , but HAT1mEAR1 could not form a complex with TPL ( Fig 5C ) , suggesting that EAR motif 1 determined the interaction between HAT1 and TPL . Finally , we investigated whether HAT1 behaved as a transcriptional repressor to inhibit VP16-mediated transcriptional activation . As expected , wild-type HAT1 significantly inhibit the VP16-promoted LUC activity , but deletion of the EAR motif 1 ( HAT1ΔEAR ) or HAT1mEAR1 abolished the inhibitory effects ( Fig 5D ) . Taken together , these results suggested that HAT1 interacts with TPL and plays as a repressor . We next investigated whether the HAT1-dependent repression of anthocyanin accumulation is released by the loss of function of TPL . To clarify this , we crossed 35S:HAT1 #13 to the tpl-1 mutant , which is an N176H substitution and has a dominant-negative on the rest of TPRs [50] . Loss of function of TPL in 35S:HAT1 #13 ( 35S:HAT1 #13 +/tpl-1 ) largely rescued the lower anthocyanin accumulation phenotype of 35S:HAT1 #13 under high light conditions ( Fig 5E and 5F ) . Consistently , repression of DFR , LDOX and UF3GT in 35S:HAT1 +/tpl-1 were largely released than that of 35S:HAT1 #13 ( Fig 5G–5I ) . These data suggest that HAT1 represses anthocyanin accumulation partially by interacting with TPL/TPRs . To further illustrate the mechanism how HAT1 regulated anthocyanin accumulation , we generated transgenic plants overexpressing variants of HAT1 protein ( 35S:HAT1mEAR1 ) . This approach was used successfully in the previous study [43] . Two independent transgenic lines , 35S:HAT1mEAR1 #6 and 35S:HAT1mEAR1 #10 , showed more anthocyanin accumulation than of 35S:HAT1 #13 under high light conditions ( Fig 6A and 6B ) . Intriguingly , we noticed that the anthocyanin contents of 35S:HAT1mEAR1 #6 and 35S:HAT1mEAR1 #10 could not completely restore to that of wild-type plants . Consistently , the expression levels of these three transgenes were similar , but the transcript levels of DFR , LDOX and UF3GT were higher in the 35S:HAT1mEAR1 transgenic plants under high light conditions compared with 35S:HAT1 #13 plants ( Fig 6C–6F ) . These data indicate the EAR motif 1 is required for HAT1-repressed anthocyanin accumulation . It has been reported that transgenic plants overexpressing MYB75-SRDX fusion protein ( 35S:MYB75-SRDX ) exhibited minimal anthocyanin under 3% sucrose [51] . Like EAR motif , the SRDX domain can convert transcriptional activators into dominant repressor when fused to transcription factors . In our study , the evidence that the C-terminal fragment of HAT1 ( 234 to 282 residues ) interacts with MYB75 and that the EAR motif in the N-terminal region interacts TPL suggests that HAT1 could link TPL to MYB75 and convert MYB75 into a repressor . To check this hypothesis , MYB75 was fused with the N-terminal fragment of HAT1 ( 1 to 90 residues containing EAR motif ) , and the fusion construct driven by a CaMV 35S promoter was transformed into wild-type Arabidopsis ( designated 35S:MYB75-HAT1N ) . As expected , two independent transgenic plants , 35S:MYB75-HAT1N #2 and 35S:MYB75-HAT1N #9 , exhibited minimal anthocyanin accumulation compared with 35S:MYB75 transgenic plants under high light conditions ( Fig 7A and 7B ) . Consistently , the levels of DFR , LDOX and UF3GT transcripts were also lower in 35S:MYB75-HAT1N #2 and 35S:MYB75-HAT1N #9 plants compared with 35S:MYB75 transgenic plants under high light conditions despite the similar expression levels of MYB75 in these transgenes ( Fig 7C–7F ) . Then we performed a transient transformation assay using the DFR promoter fused to the LUC gene as a reporter . HA-MYB75 , HA-TPL and HAT1-GFP construct acted as effector and transfected together with the reporter construct into myb75-c mesophyll protoplasts . The LUC expression of DFR promoter was very low without MYB75 expression , but was activated by expression of MYB75 ( S7 Fig ) . However , when HAT1 was coexpressed with MYB75 , this activation was markedly decreased ( S7 Fig ) . The relative luciferase activities were moreover decreased when TPL was co-transformed with HAT1 , but the repression was alleviated when TPL was coexpressed with mutational HAT1 , suggesting that HAT1 inhibits the transcriptional activity of MYB75 through TPL function . Taken together , these results indicate that HAT1 represses anthocyanin accumulation by connecting TPL with MYB75 . Previous studies have shown that TPL interacts with two histone deacetylase , HDA6 and HDA19 , which function in chromatin modification and epigenetic regulation of developmental and hormone-responsive genes [52 , 53] . The presence of ternary MYB75-HAT1-TPL protein complex led us to further analyze whether HAT1 represses the expression of anthocyanin biosynthetic genes via chromatin modifications . To prove this speculation , ChIP assays were performed by using antibody of acetylated H3 in different mutants . As shown in Fig 8A–8C , reduced histone H3 acetylation was verified in the transcription start sites ( TSSs ) of DFR , LDOX and UF3GT in 35S:HAT1 #13 under high light conditions , while elevated histone H3 acetylation was detected in hat1 mutant under the same condition . Simultaneously , 35S:MYB75-HAT1N #2 transgenic plants showed lower histone H3 acetylation in the TSSs of DFR , LDOX and UF3GT than that of 35S:MYB75 transgenic plants ( Fig 8D–8F ) . These fingdings suggest that MYB75-HAT1-TPL inhibits LBGs expression through recruiting a histone modification complex .
Although members of HD-Zip family have been well described in plants , the possible function in anthocyanin accumulation remains to investigate . GL2 , a member of HD-Zip subfamily IV , has been characterised as a negative regulator of anthocyanin accumulation in Arabidopsis [54] . In this study , our findings elucidate that HAT1 may function as a new repressor in anthocyanin accumulation through sequestering the MBW protein complex and epigenetic regulation in Arabidopsis . Additionally , MYB75 , MYB90 , TT8 and EGL3 transcript levels were increased in hat1 mutant and decreased in 35S:HAT1 plants ( S2D Fig ) , presenting the possibility that HAT1 regulates anthocyanin accumulation by modulating MBW protein complex expression . Our results indicated that HAT1 regulated anthocyanin accumulation via post-translational regulation and transcriptional regulation of MBW protein complex . Much work so far has evidenced that the expression of the members of HD-Zip II family is induced in response to simulated shade conditions . Interestingly , weak anthocyanin accumulation occurs under shade conditions in petunia plants and the conversion of light conditions change the members of HD-Zip II family abundance [30 , 55] . It seems that the members of HD-Zip II family are involved in modulation of anthocyanin accumulation in response to light/shade . However , MYB75 did not interact with HAT2 , HAT3 , ATHB2 , and ATHB4 in yeast ( S8 Fig ) , suggesting that HAT1 is a unique gene of HD-Zip II family to repress anthocyanin accumulation . It is likely that overexpression of one member will repress the expression of rest in HD-Zip family II [34] . Previous study described that the EAR motif exists in different members of HD-Zip II family [30] . We investigate the impact of EAR motif on anthocyanin accumulation . We proved that EAR motif is required for HAT1-repressed anthocyanin accumulation . Further , HAT1 suppresses anthocyanin accumulation via TPL-dependent histone deacetylation ( Fig 8A–8F ) . Interestingly , our previous study proved that HAT1 negatively regulates hormone synthesis and response gene [36 , 45] . It seems that HAT1 behaves as a transcriptional repressor in various hormonal signaling and metabolic process . Our recent study reported that phosphorylation of HAT1 by SnRK2 . 3 induced its degradation . And abscisic acid ( ABA ) positively regulated anthocyanin accumulation [56] . It raised the possibility that ABA induced anthocyanin accumulation by inhihiting HAT1 founction . Our previous studies also revealed that HAT1 was a positive regulator in BR pathway [36] . Under favorable conditions , HAT1 accumulated and interacted with BES1 to promote plant growth . Meanwhile , HAT1 also interacted with MYB75 and TPL to inhibit anthocyanin accumulation . It demonstrated that HAT1 was a key regulator in balancing plant growth and stresses adaption . Numerous EAR motif-contained transcription factors repress gene expression through directly binding to the cis element . The previous study showed HAT1 interacts with BES1 [36] . Intriguingly , BES1-TPL complex mediates the inhibitory action of brassinosteroids on ABA responses during early seedling development [57] . We speculate that the these proteins can form a HAT1-BES1-TPL protein complex to repress BR-response gene . A recent study revealed that ATHB4 regulates leaf polarity and hypocotyl elongation by interacting with TPL protein [58] . Several studies proposed that some transcription factors repressed gene transcription by recruiting TPL and HDAs to the promoter of target gene [44 , 53 , 59] . It is important to highlight that HAT1 inhibits the expression of anthocyanin biosynthetic-specific genes through bridging MYB75 and TPL rather than directly binding to the promoter of these genes . Anthocyanin accumulation occurs at the junction of the rosette leaves and stem during plant development in Arabidopsis [28] . We noticed the purple pigments of 35S:HAT1 #13 plants was much lower than that of wild-type , whereas hat1 mutant exhibited more purple pigments compared with wild-type ( S9A and S9B Fig ) . The expression of DFR is confined within basal regions of stems during the transition from leaves to flowers [28] . MYB75 showed high expression in the lower part of the inflorescence stem and leaves in 6-week-old Arabidopsis [48] . Our results showed that transcript levels of HAT1 was minimal in the lower part of the inflorescence stem and highest in the upper part of the inflorescence stem , while DFR and MYB75 exhibited an inverse pattern in 6-week-old Arabidopsis ( S10A Fig ) . Consequently , low expression of HAT1 eliminates the repression for MBW protein complex , thus resulting in anthocyanin accumulation in stem-rosette junction . On the other hand , HAT1 showed low expression in senescent leaves , while DFR and MYB75 displayed high expression in senescent leaves ( S10B Fig ) . Consistent with this , gene expression analysis using the Arabidopsis microarray data displayed in the eFP browser indicated opposite expression patterns of MYB75 and HAT1 , suggesting HAT1 serves as a repressor in senescent leaves during plant senescence ( S11 Fig ) . MBW protein complex activates anthocyanin biosynthetic-specific gene expression when HAT1 expression is limited and thereby induces anthocyanin accumulation . These data proved that HAT1 represses anthocyanin accumulation in stem-rosette junction and leaves during plant senescence . Present studies demonstrated that epigenetic modifications participated in regulation of anthocyanin accumulation [60 , 61] . Recent research suggested DELLA promoted anthocyanin accumulation via sequestering MYBL2 and JAZ suppressors of the MBW complex in Arabidopsis thaliana [62] . To test the relationship between HAT1 and MYBL2 in suppression of anthocyanin accumulation , we crossed 35S:HAT1 #13 with mybl2 mutant . The 35S:HAT1 #13 mybl2 plants showed less anthocyanin accumulation ( S12A and S12B Fig ) , indicating that HAT1 repressed anthocyanin accumulation is independent of the MYBL2-regulated pathway . We further observed that H3 acetylation levels in the promoter of LBGs were elevated in mybl2 mutant under high light conditions ( S12C–S12E Fig ) . These results suggest that epigenetic modifications are prevalent in regulation of anthocyanin accumulation in plants . Interestingly , maize bHLH transcription factor R interacts with RIF1 and thereby forms a complex with MYB transcription factor C1 . C1-R-RIF1 complex binds to A1 promoter and activates A1 expression by elevating the H3K9 and H3K14 acetylation levels in the promoter region [63] . Therefore , it remains challenging to identify the interaction between MBW protein complex and epigenetic regulators , which may explain why MBW protein complex activates the expression of biosynthetic genes . To date , much work has characterized many repressors in modulation of anthocyanin accumulation . Several repressors regulate anthocyanin accumulation without direct interaction with members of MBW protein complex [7 , 54 , 64] . Notably , it has been proved that LBD37 interact with TPL via Y2H screening [65] . On the other side , MYB75 , bHLH and TTG1 are expressed in various tissues of Arabidopsis under non-inductive conditions [15 , 16 , 19 , 48] . Therefore , some repressors suppress anthocyanin accumulation by directly interacting with members of MBW protein complex [4 , 28 , 29 , 66] . We prove that HAT1 suppresses anthocyanin accumulation by directly interacting with MYB75 . Intriguingly , the proteins that interact with MBW protein complex all contain EAR moif except SPL9 . Y2H screening proved that JAZ5/JAZ6/JAZ7/JAZ8 , MYBL2 and HAT1 interact with TPL respectively [65] . We speculate that EAR moif containd-repressors inhibit anthocyanin accumulation by forming a protein complex with components of MBW and TPL in plants . The repressive activity of MYBL2 might play a critical role in suppression of anthocyanin accumulation [24] . In our studuy , HAT1mEAR1 may still interact with MYB75 and interfere with MBW protein complex because the C terminus of HAT1 determines the interaction with MYB75 . We observe that anthocyanin level of 35S:HAT1mEAR1 transgenic plants was lower than that of wild-type . These results suggested that HAT1 inhibits anthocyanin accumulation partially by disturbing the formation of MBW protein complex . We suppose that interference of MBW protein complex ( Passive repression ) and epigenetic modification ( Active repression ) function synchronously in regulation of anthocyanin accumulation . Under non-inductive conditions , such as low light conditions , plants possess high levels of HAT1 . Meanwhile , HAT1 interferes with MBW protein complex by binding MYB75 , as well as transforming the active MBW protein complex into a repressive complex by recruiting EAR motif-dependent TPL corepressor , thus preventing the expression of LBGs ( Fig 8G ) . Under inductive conditions , such as high light conditions , HAT1 expression is suppressed . MYB75 and bHLH transcription factors are able to form MBW protein complex with TTG1 protein , which activates transcription of the target genes encoding LBGs ( Fig 8H ) . Overall , our work together with other studies suggests that plants restrict the expression of anthocyanin biosynthetic genes via the multiple and intricate mechanisms .
The Arabidopsis thaliana 35S:HAT1 #11 , 35S:HAT1 #13 and hat1 mutants were described as previously [36] . All wild-type , various mutants , and transgenic plants in this study are in Col-0 ecotype background . To avoid ecotype variability , the tpl-1 mutant , originally in Ler background [40] , was introgressed into the Col-0 . Arabidopsis seeds were placed on half-strength Murashige and Skoog medium . The plates were placed at 4°C for 3 d avant transfering to 22°C under different light conditions . Plates were put at control ( 40 μmol m-2 s-1 ) or high light ( 180 μmol m-2 s-1 ) conditions with a 16-h-light/8-h-dark photoperiod for high light-induced research [5] . For the BiFC assays , Nicotiana benthamiana was cultured in soil at 22°C under 16-h-light/8-h-dark conditions . Anthocyanin levels were measured as described previously [67] . Briefly , arabidopsis seedlings were incubated in extraction buffer ( methanol containing 1% HCl ) overnight at 4°C in the dark . After extraction and centrifuged , the supernatants were collected and absorbance calculated at 530 and 657 nm . Relative anthocyanin content was quantified by ( A530-0 . 25×A657 ) per gram fresh weight . HAT1mEAR1 was amplified from pZP211-HAT1 using primers indicated in S1 Table and cloned into pZP211 vector to generate 35S:HAT1mEAR1-GFP [68] . To generate MYB75 constructs , the 1500 bp genomic sequence of MYB75 contained the coding area was cloned into pCM1307 vector to create 35S:HA-MYB75 [69] . The sequence coding N terminus of HAT1 was amplified from pZP211-HAT1 and cloned into pCM1307-MYB75 to generate 35S:HA-MYB75-HAT1N . The coding sequences ( CDS ) of TPL were amplified and cloned into pCM1307 plasmid to create 35S:HA-TPL . Oligo primers used for cloning are listed in S1 Table . Col-0 plants were transformed with these constructs by using Agrobacterium tumefaciens ( strain GV3101 ) -mediated transformation [70] . The GAL4 reporter plasmid was generated from pUC19 [71] , which contains the firefly LUC reporter gene driven by the minimal TATA box of the 35S promoter plus five GAL4 binding elements . For transcriptional inhibition assays , HAT1ΔEAR and HAT1mEAR were amplified from pZP211-HAT1 and cloned into pRT-BD to generate GAL4-HAT1ΔEAR and HAT1mEAR respectively [72] . The positive control ( pRT-35S-BD-VP16 ) was constructed by insertion of VP16 , a herpes simplex virus-encoded transcriptional activator protein , into pRT-BD . Plasmid pTRL was used as internal control . For transcription activity assays in protoplast , a 512-bp DFR promoter was amplified from genomic DNA and fused with pGreenII 0800-LUC . The internal control , effectors and reporter were co-transformed into Arabidopsis protoplasts by PEG/CaCl2-mediated transfection [73] . All transfection cultured for 16 h , then luciferase assays were performed using the Promega dual-luciferase reporter assay system and a GloMax 20–20 luminometer ( Promega , http://www . promega . com ) . Relative LUC activity was defined as firefly LUC activity divided by Renilla LUC activity . For yeast two-hybrid assays , the full-length CDS of MYB75 and HAT1 were amplified and cloned into pGADT7 ( Clontech ) . The full-length CDS of MYB75 and N terminus of TPL were amplified and cloned into pGBKT7 ( Clontech ) . The yeast strain ( AH109 ) was transformed with pairs of plasmids and grown on Double DO supplement ( SD-Leu/-Trp ) for 3 days , then the cotransformants were shifted onto Quadruple DO supplement ( SD-Leu/-Trp/-Ade/-His ) to test for possible interactions . For yeast three-hybrid assays , the complete CDS of TT8 and EGL3 were amplified and fused with pBridge-HAT1 plasmid to generate pBridge-TT8-HAT1 and pBridge-EGL3-HAT1 respectively [74] . Yeast three-hybrid experiments performed as described previously [20] . Briefly , pBridge-TT8-HAT1 or pBridge-EGL3-HAT1 were used in co-transformation with pGADT7-MYB75 . pBridge contains a methionine ( Met ) suppressible promoter positioned upstream of a Gateway cassette . HAT1 expression was gradually suppressed using increasing methionine concentrations in Minimal Media Quadruple Dropouts ( SD-Leu/-Trp/-His/-Met ) . For each combination , 3 colonies selected on dropout medium ( SD -Leu/-Trp ) were resuspended in water , the OD600nm was adjusted to 0 . 7 and 20 μl was streaked out on the respective plates . For BiFC assays , full-length CDS of MYB75 and HAT1 was cloned into the pXY103-nYFP vector respectively [75] . The full-length CDS of MYB75 , TT8 , EGL3 and TTG1 were cloned into the pXY104-cYFP vector respectively [75] . The constructs were transformed into Agrobacterium tumefaciens strain ( GV3101 ) respectively , and the lower epidermis of Nicotiana benthamiana plants were used for injection of different combination . The transfected plants were grown in the green house for at least 36 hours and fluorescent signals were observed by using scanning microsystem ( Leica ) . Different version of MYB75 were cloned into the pMAL-C2X and pET28a vectors with MBP tag and 6×His tag respectively . TT8 and EGL3 were also cloned into the pMAL-C2X vector respectively . Different version of HAT1 were cloned into the pMALc-B and pET28a vectors with MBP tag and 6×His tag respectively . In vitro pull-down assays performed as described [76] . Ni-NTA beads containing 5 μg HIS-HAT1 proteins incubated with 5 μg MBP-MYB75 by using 500 μl pull-down buffer ( 150 mM NaCl , 20 mM Tris , 1 mM PMSF , 0 . 2% Triton X-100 , 1% protease inhibitor cocktail [pH 8 . 0] ) at 4°C for 2 h . Equally , MBP-HAT1 was incubated with Ni-NTA beads contained HIS-MYB75 . Beads were washed four times with the pull-down buffer and proteins were eluted from beads by boiling in 95°C with 30 μL SDS-PAGE loading buffer then separated by SDS-PAGE and analyzed by the anti-MBP antibies . Competitive HIS pull-down analysis performed as described previously [77] , 5 μg of MBP-TT8 or MBP-EGL3 mixed with 5 , 10 , or 20 μg HIS-HAT1 were incubated with Ni-NTA beads containing 5 μg HIS-MYB75 by using 500 μl pull-down buffer ( 150 mM NaCl , 20 mM Tris , 1 mM PMSF , 0 . 2% Triton X-100 , 1% protease inhibitor cocktail [pH 8 . 0] ) at 4°C for 2 h . Beads were washed four times with the pull-down buffer and proteins were eluted from beads by boiling in 95°C with 30 μL SDS-PAGE loading buffer then separated by SDS-PAGE and analyzed by the anti-MBP antibies . For the competing MBP pull-down assay [78] , samples from Col-0 or 35S:HAT1 protoplasts expressed HA-MYB75 were collected in protein extraction buffer containing 150mM NaCl , 50mM Tris-HCl ( pH 7 . 5 ) , 0 . 1% ( v/v ) NonidetP-40 , 10% ( v/v ) glycerol , and 1×complete protease inhibitor cocktail ( Roche ) . The lysate was centrifuged at 12 , 000 g for 5 min at 4°C , and the supernatant was taken for semi-in vivo pull-down assay . MBP-TT8 or MBP-EGL3 beads was added to 500 μL of total extracted protein and incubated at 4°C for 3 h . Beads were washed in extraction buffer five times , resuspended in SDS-PAGE loading buffer and analyzed using SDS-PAGE and immunoblotting with anti-HA antibody , anti-GFP and anti-MBP antibody . Plants expressing different proteins as indicated were extracted with protein extraction buffer containing 150mM NaCl , 50mM Tris-HCl ( pH 7 . 5 ) , 0 . 1% ( v/v ) NonidetP-40 , 10% ( v/v ) glycerol , and 1×complete protease inhibitor cocktail ( Roche ) [78] . After centrifuging at 12 , 000 g for 5 min at 4°C , and the supernatant was incubated for 3 h with Anti-HA Agarose Affinity Gel antibody at 4°C . Then the beads were washed six times using extraction buffer and then eluted with 50 μL of SDS-PAGE loading buffer for immunoblot analysis using Anti-HA and Anti-GFP antibody . ChiP assays perform as described [75] . The 4-week-old plants collected in 50 mL tubes , and 37 mL 1% formaldehyde solution was used for cross-linked under a vacuum for 20 min . The chromatin was collected and sheared by sonication to reduce the average DNA fragment size to around 500 bps , then the sonicated chromatin complex was immunoprecipitated by specific antibodies anti-acetyl-histone H3 ( Catalog no 06–599 , Millipore ) . After reverse cross-linking , the immunoprecipitated DNA fragment was analysed by qPCR with specific primers shown in S1 Table . Total RNA extraction , cDNA synthesis and qRT-PCR were performed as described before [79] . PCR analysis was carried out using SYBR Green PCR Master Mix was used as previously described . Three separate experiments were implemented , and technical triplicates of each experiment were implemented . Gene expression normalize to the transcript levels of ACTIN 8 . Samples were analyzed in triplicates , and the data are expressed as the mean ± SD unless noted otherwise . Statistical significance was determined using two-way ANOVA ( LSD’s multiple-range test ) or Student’s t-test . A difference at P<0 . 05 was considered significant . The Arabidopsis Genome Initiative identifiers for the genes described in this article are as follows: HAT1 ( At4g17460 ) , HAT2 ( At5g47370 ) , HAT3 ( At3g60390 ) , ATHB2 ( At4g16780 ) , ATHB4 ( At2g44910 ) , MYB75 ( At1g56650 ) , MYB90 ( At1g66390 ) , TT8 ( At4g09820 ) , EGL3 ( At1g63650 ) , TTG1 ( At5g24520 ) , MYBL2 ( At1g71030 ) , TPL ( At1g15750 ) , TPR1 ( At1g80490 ) , TPR2 ( At3g16830 ) , TPR3 ( At5g27030 ) , TPR4 ( At3g15880 ) , DFR ( At5g42800 ) , LDOX ( At4g22880 ) , UF3GT ( At5g54060 ) , ACTIN 7 ( At5g09810 ) and ACTIN 8 ( At1g49240 ) . | Anthocyanins , a class of flavonoids distributed ubiquitously in the plant kingdom , are induced by environmental stimuli and developmental signals , such as high light and sucrose . It is well established that anthocyanin accumulation is regulated by the MYB-bHLH-WD40 ( MBW ) protein complex in plants . But little is known about the regulation of MBW protein complex by other factors . Here , we show that an HD-ZIP II transcription factor HAT1 negatively regulates anthocyanin accumulation via post-translational regulation of MBW protein complex . Loss of function of HAT1 in the Arabidopsis seedlings exhibits increased anthocyanin accumulation , whereas overexpression of HAT1 significantly repressed anthocyanin accumulation . We reveal that HAT1 interacted with MYB75 and thereby sequestered MBW protein complex . Overexpression of HAT1 in pap1-D mutant suppresses abundant anthocyanin phenotype of the pap1-D mutant . HAT1 identified was as a transcriptional repressor possessing an N-terminal EAR motif , which determines the interaction with TOPLESS corepressor . The deletion or mutation of the EAR motif 1 of HAT1 partially eliminates the repression activity of HAT1 in regulation of gene expression and anthocyanin accumulation . Our results illustrate a new repressor HAT1 which helps plants fine-tune anthocyanin accumulation . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [
"biotechnology",
"plant",
"anatomy",
"medicine",
"and",
"health",
"sciences",
"engineering",
"and",
"technology",
"ears",
"gene",
"regulation",
"regulatory",
"proteins",
"brassica",
"dna-binding",
"proteins",
"dna",
"transcription",
"plant",
"science",
"model",
"organis... | 2019 | Regulation of anthocyanin accumulation via MYB75/HAT1/TPL-mediated transcriptional repression |
It has long been recognized that the modification of penicillin-binding proteins ( PBPs ) to reduce their affinity for β-lactams is an important mechanism ( target modification ) by which Gram-positive cocci acquire antibiotic resistance . Among Gram-negative rods ( GNR ) , however , this mechanism has been considered unusual , and restricted to clinically irrelevant laboratory mutants for most species . Using as a model Pseudomonas aeruginosa , high up on the list of pathogens causing life-threatening infections in hospitalized patients worldwide , we show that PBPs may also play a major role in β-lactam resistance in GNR , but through a totally distinct mechanism . Through a detailed genetic investigation , including whole-genome analysis approaches , we demonstrate that high-level ( clinical ) β-lactam resistance in vitro , in vivo , and in the clinical setting is driven by the inactivation of the dacB-encoded nonessential PBP4 , which behaves as a trap target for β-lactams . The inactivation of this PBP is shown to determine a highly efficient and complex β-lactam resistance response , triggering overproduction of the chromosomal β-lactamase AmpC and the specific activation of the CreBC ( BlrAB ) two-component regulator , which in turn plays a major role in resistance . These findings are a major step forward in our understanding of β-lactam resistance biology , and , more importantly , they open up new perspectives on potential antibiotic targets for the treatment of infectious diseases .
Decades after their discovery , β-lactams remain key components of our antimicrobial armamentarium for the treatment of infectious diseases . Bacterial resistance to them is generally driven either by the production of enzymes that inactivate them ( β-lactamases ) , or by the modification of their targets in the cell wall ( penicillin-binding Proteins , PBPs ) , sometimes in conjunction with mechanisms leading to diminished permeability or active efflux [1] . While the acquisition of modified PBPs showing low affinity for β-lactams is well known to be a major resistance mechanism in Gram-positive cocci , such as penicillin-resistant Streptococcus pneumoniae or the much-feared methicillin-resistant Staphylococcus aureus , this mechanism has not been thought to be important for most species of Gram-negative rods ( GNR ) [2] . The production of intrinsic or horizontally acquired β-lactamases is undoubtedly the predominant resistance mechanism in the latter organisms [3] . Among GNRs , the most widely distributed β-lactamases are chromosomally-encoded AmpC variants , produced by most Enterobacteriaceae and Pseudomonas aeruginosa , high up the list of pathogens causing life-threatening infections in hospitalized patients world-wide [4] . Although AmpC is produced at very low basal levels in wild-type strains , its expression is highly inducible in the presence of certain β-lactams ( β-lactamase inducers ) such as cefoxitin or imipenem [3] . In fact , the efficacy of the widely-used broad spectrum penicillins ( such as piperacillin ) and cephalosporins ( such as ceftazidime ) relies on the fact that they are very weak AmpC inducers , even though they are efficiently hydrolyzed by this enzyme [3] . Unfortunately , mutants showing constitutive high level AmpC production ( AmpC derepressed mutants ) are frequently selected during treatment with these β-lactams , leading to the failure of antimicrobial therapy [5] , [6] . In some natural strains of Enterobacteriaceae and P . aeruginosa [6]–[9] , the inactivation of AmpD ( cytosolic N-acetyl-anhydromuramyl-L-alanine amidase involved in peptidoglycan recycling [10]–[12] ) , and point mutations in AmpR ( LysR-type transcriptional regulator required for ampC induction [13]–[15] ) have been found to lead to AmpC overexpression , and thus to β-lactam resistance . In this paper we show that , in contrast to the current expectations , the mutations triggering β-lactam resistance in P . aeruginosa , whether in vitro , in vivo , or in the clinical setting , frequently arise within a PBP gene . Inactivation of the E . coli dacB ortholog , encoding the nonessential low molecular mass PBP4 [16] , [17] , is demonstrated to be the principal route to one-step high level ( clinical ) β-lactam resistance , by triggering the overexpression of ampC and the specific activation of the CreBC two component regulator [18] , which is also found to play a major role in resistance .
The mechanisms by which β-lactam resistance arises were studied in a previously described [19] collection of 36 independent ceftazidime resistant mutants . These mutants were obtained in vitro ( one-step spontaneous mutants ) or in vivo ( after 3 days of treatment with humanized ceftazidime regimen in mouse model of lung infection ) , at two ceftazidime concentrations ( 4 and 16 µg/ml ) , and from the wild-type strain PAO1 ( normal mutation rate supply ) or its mutS deficient hypermutable derivative PAOΔmutS ( high mutation rate supply ) . In the previous study , all the mutants were shown to be highly resistant to all tested penicillins , cephalosporins , and monobactams; overexpression of the chromosomal cephalosporinase AmpC ( 18 to 236-fold higher expression relative to wild-type ) was found to be the instrument of β-lactam resistance in all cases . In this work , in an attempt to find out the genetic mechanisms leading to AmpC hyperproduction , we sequenced and quantified the expression of all genes so far known to be involved in ampC regulation and overexpression ( ampD , ampE , ampDh2 , ampDh3 and ampR ) in the 36 mutants . We also performed complementation experiments with plasmids harboring the wild-type ampD gene ( pUCPAD ) or the complete ampDE operon ( pUCPADE ) . A complete report of the obtained results is provided in Table S1 . In contrast to present models , almost none of the mutants ( 32 of 36 ) showed mutations in any of the loci examined . The only exceptions were 4 mutants obtained in vivo from PAOΔmutS ( high mutation rate supply ) at the low ceftazidime concentration ( 4 µg/ml ) , each showing a different mutation in ampD . A modified expression of any of the studied genes was neither observed in the 32 mutants . As expected , only the four ampD mutants showed positive complementation with pUCPAD , but , intriguingly , all the 36 mutants showed positive complementation with pUCPADE ( Table S1 ) . Furthermore , positive complementation required the simultaneous presence of both ampD and ampE , since plasmids harboring ampDE operons with a non functional ampD and a wild-type ampE also failed to complement the resistance phenotype . These findings suggested that , contrary to current understanding , mutations in ampD or ampR are not at all the most common , in vitro or in vivo , leading to AmpC overexpression and high level ( clinical ) β-lactam resistance in P . aeruginosa . Furthermore , the results strongly suggest that one-step high-level ceftazidime resistance in P . aeruginosa mainly occurs through single mutations in a gene/genes previously unknown to be involved in β-lactam resistance or AmpC regulation . That a single mutation has to be responsible for the resistance phenotype is shown by the ceftazidime resistance mutation rates published previously [19] . At two different ceftazidime concentrations ( 4 and 16 µg/ml ) , spontaneous resistant mutants were obtained with a rate of 10−8 mutations per cell division for wild-type PAO1 and of 10−5−10−6 for PAOΔmutS . These mutation rates should rule out the involvement of more than one mutation in the resistance phenotype . In an attempt to detect the mutations in the gene ( s ) yet unknown to be involved in β-lactam resistance , we followed a whole-genome analysis approach . Four of the PAO1 ceftazidime resistant in vitro mutants were analyzed by comparative hybridization on a recently described microarray for the discovery of single nucleotide polymorphisms ( SNPs ) in P . aeruginosa [20] using the parental PAO1 strain as reference . As shown in Figure 1 , major decreases in hybridization ratios ( indicating deletions of 50–100 base pairs ) were detected for two of the mutants in the gene PA3047 , the E . coli dacB ortholog , encoding the nonessential low molecular mass PBP4 [16] , [17] . PCR and sequencing confirmed the presence of the deletions in this gene [ ( nts 1149–1231 for one of the mutants ( 1A5 ) and nts 1069–1138 for the other ( 1D7 ) ] ( Figure 1 , Table S1 ) . Furthermore , the two remaining mutants ( 1A1 and 1D4 ) also revealed a less pronounced decrease of hybridization ratio at a single position in gene PA3047 ( Figure 1 ) ; PCR and sequencing identified as well the mutations , a G to A change in nt 819 leading to a premature stop codon ( W273X ) for 1A1 and a A to C change in nt 235 leading to a missense mutation ( T79P ) for 1D4 ( Table S1 ) . The function and structure of PBP4 ( DacB ) has been characterized mainly in E . coli . The protein is a nonessential low molecular mass class C PBP with DD-carboxypeptidase and DD-endopeptidase activity , that is thought to play an auxiliary role in morphology maintenance , peptidoglycan maturation and recycling , and cell separation during division [17] , [21] , [22] . The crystal structure of E . coli PBP4 has been recently determined [16] and found to be organized in three domains . Domain I has the characteristic SXXK , SXN , and KTG motifs of PBPs and β-lactamases , and contains the other two extra domains embedded within it . PBP4 from P . aeruginosa shows a 27% identity with that of E . coli , and contains all conserved motifs . The alignment of E . coli and P . aeruginosa PBP4 sequences is included in Figure S1 . Following the discovery of mutations within the dacB ortholog , we sequenced this gene in the rest of the collection of the 36 ceftazidime resistant mutants , and the complete list of the mutations detected is provided as Table S1 . All but 2 of the 32 mutants not having mutations in ampD had mutations in dacB . A total of 28 different mutations were detected , and included deletions/insertions ( 9 ) , nonsense mutations ( 7 ) , and missense mutations ( 12 ) . Many of the missense mutations occurred in sequences encoding highly conserved motifs , including the catalytic serine , at position 72 in P . aeruginosa ( Figure S1 , Table S1 ) . In order to further confirm the role of dacB mutations in β-lactam resistance , we constructed the dacB knockout mutant of PAO1 ( PAΔdacB ) . As shown in Table 1 , the inactivation of dacB in PAO1 yielded an almost identical phenotype to that documented in the ceftazidime-resistant dacB spontaneous mutants , with high level β-lactam resistance and ampC overexpression . Therefore , it is for the first time demonstrated that the inactivation of a particular PBP ( which are supposed to be antibiotic targets ) produces high-level β-lactam resistance . In order to understand the role of PBP4 mutation in β-lactam resistance and upregulation of AmpC expression , we constructed the ampC , ampR ( transcriptional regulator of AmpC ) , ampD ( negative regulator of AmpC ) and ampE ( second component of the bicistronic ampDE operon , encodes an inner membrane-bound sensory transducer that modulates AmpD activity [23] ) , knockout mutants of strain 1A1 ( DacB W273X in vitro spontaneous mutant of PAO1 ) and of strain PAO1 as control . As shown in Table 1 , the inactivation of ampC completely restored ceftazidime susceptibility in 1A1 , showing that the overexpression of the β-lactamase is essential for the resistance phenotype . Furthermore , the inactivation of ampR restored ceftazidime susceptibility and basal ampC expression levels , thus demonstrating that the effect of PBP4 mutation requires a functional AmpR . Therefore , considering that PBP4 has been shown to be involved in peptidoglycan recycling [17] it seems reasonable to believe that ampC overexpression driven by dacB inactivation , as occurs in the classical ampD mutation pathway , should be consequence of the qualitative or quantitative modification of muropeptides , that are the effector molecules for AmpC induction through their interaction with AmpR [24] . Our results are also consistent with previous observations in the E . coli model , in which the strongest AmpC inducers ( such as imipenem ) were shown to be potent PBP4 inhibitors , suggesting a role of this PBP in the induction process [25] . Additionally , we show that the AmpDE pathway of AmpC repression is functional in the PBP4 mutants , since the inactivation of ampD dramatically increased further ampC expression and ceftazidime resistance in 1A1 ( Table 1 ) . Furthermore , while the inactivation of ampE in PAO1 ( or in its ampD mutant ) did not produce significant effects , it also determined a marked increase of ampC expression and ceftazidime resistance in the dacB mutant . These results suggest that both genes of the ampDE operon play a major role in the dacB mutant background . This conclusion is further supported by the positive complementation of the PBP4 mutants with the complete ampDE operon expressed from a multicopy plasmid ( Table 2 ) . Moreover , the expression of dacB from a multicopy plasmid ( pUCPdB ) also complemented both , the dacB and the ampD mutants ( Table 2 ) . Therefore , these results show that PBP4 and AmpDE are parallel synergic ampC regulatory pathways ( a defect in one of them can be complemented by increasing the amount of the other ) , both ultimately relying on a functional AmpR . While both pathways have a very similar effect on ampC expression , PBP4 mutation confers high level ( clinical ) β-lactam resistance ( i . e . resistant according to current breakpoints ) , while ampD inactivation confers only moderate resistance ( i . e . still susceptible according to current resistance breakpoints ) ( Table 1 ) . In fact , the resistance level conferred by PBP4 mutation is more similar to that conferred by the simultaneous inactivation of the three ampD genes of P . aeruginosa ( ampD plus the two additional homologous genes , ampDh2 and ampDh3 [26] ) , that produces a much higher increase in ampC expression ( Table 1 ) . Nevertheless , this mechanism of high-level resistance is not found among clinical strains [27] , [28] , because it requires the acquisition of several mutations and because it causes a marked reduction of fitness and virulence [27] . Here we show that in vivo ( murine systemic infection model ) fitness is not affected in the PAO1 dacB mutant , as shown by the competition index ( CI ) of 0 . 92 , in sharp contrast to the previously documented CIs of less than 0 . 01 for the double and triple ampD mutants [27] . Therefore , in contrast to the ampD inactivation pathway , PBP4 mutation in P . aeruginosa is a very efficient one-step trigger of high level β-lactam resistance mechanism of potentially enormous clinical relevance . In order to explore further the effects of PBP4 mutation , we performed a whole-genome analysis of gene expression in two selected mutant strains ( 1A1 and 2A2 ) compared to wild-type PAO1 , using the Affymetrix GeneChip P . aeruginosa genome array . In addition to ampC ( and co-transcribed PA4111 ) , which obviously was upregulated , only one further gene showed a significantly ( >2-fold change ) modified expression . This gene , creD , was upregulated in both mutants analyzed ( not shown ) . creD encodes an inner membrane protein of yet unknown function that is regulated by the CreBC two-component regulator . The CreBC system has been deeply studied in E . coli , and it is shown to be a global regulator involved in metabolic control [18] . Interestingly , the homolog of this system in Aeromonas Spp . ( designated BlrAB ) has been shown to play a role in the regulation of β-lactamase expression in these species [29]–[31] . As first approach , we analyzed whether creD overexpression was a signature feature of the PBP4 mutants . Indeed , real time RT-PCR experiments confirmed the overexpression of this gene in the two selected mutants ( Table 1 ) as well as in the complete collection of in vivo and in vitro PBP4 mutants , with creD mRNA levels ranging from 25 to 60-fold higher than those of wild-type PAO1 ( not shown ) . Moreover , the inactivation of dacB in wild-type PAO1 produced a similar creD overexpression ( Table 1 ) . The effect on creD expression seemed to be specific for the PBP4 mutants , rather than a direct consequence of AmpC overexpression , since this gene was not upregulated in the ampD mutants ( Table 1 ) . Interestingly , creD expression in wild-type PAO1 was found to be highly inducible by β-lactamase inducers ( cefoxitin ) ( Table 1 ) . Furthermore , creD inducibility was significantly reduced in the ampD mutants and the reduction in expression was dependent on a functional AmpR , since its inactivation restored the inducibility ( Table 1 ) . Overall , these results suggested a link between the CreBC regulator , PBP4 mutations , and the components of the regulatory system of ampC expression . The complete creB , creC and creD genes , as well as their promoter regions , were fully sequenced in 1A1 , 2A2 and five additional randomly selected mutants from the collection . The absence of mutations supported further the notion that the mutations in PBP4 are solely responsible for the complete β-lactam resistance response . The single mutation hypothesis is definitively confirmed by the fact that direct dacB inactivation produces the same phenotype ( i . e . the same MICs and ampC and creD expression levels ) observed in the spontaneous dacB mutants . To gain insights into the role of the CreBC system in β-lactam resistance , we constructed creBC and creD knockout mutants of the PBP4 mutants 1A1 and 2A2 , as well as of PAO1 and its single and triple ampD mutants as controls . Interestingly , the inactivation of creBC in the PBP4 mutants ( 1A1 and 2A2 ) not only decreased creD expression back to wild-type levels , but also drastically decreased β-lactam MICs , leaving them well within the susceptible ( treatable ) range according to current breakpoints ( Table 1 ) . Furthermore , the effect was specific to the PBP4 mutants , since β-lactam susceptibility was not affected by creBC inactivation in wild-type PAO1 or in its ampD mutants ( Table 1 ) . Overall , these results strongly suggest that PBP4 mutations specifically trigger the activation of the CreBC two-component regulator , leading to creD upregulation and β-lactam resistance . Nevertheless , CreD seems not to be the only CreBC-dependent driver of resistance , since the direct inactivation of creD in the PBP4 mutants decreased resistance slightly , but did not give the drastic reduction seen on CreBC inactivation ( Table 1 ) . The extra resistance of the PBP4 mutants compared to the ampD mutant ( despite showing similar levels of ampC expression ) is therefore apparently driven by the specific activation of the CreBC system in the PBP4 mutants . Indeed , the resistance level of 1A1 after CreBC inactivation was more similar to that of the ampD mutant ( Table 1 ) . Further evidence showing that the CreBC system is a key component in one-step high level β-lactam resistance development was provided by mutation rates experiments . While high level ceftazidime resistant mutants were readily selected from PAO1 wild-type strain [mutation rate to ceftazidime ( at breakpoint concentration , 16 µg/ml ) resistance of 3×10−8 mutants per cell division] , mutation rates were below the detection limit ( <1×10−11 ) for its CreBC knockout mutant ( PAΔCreBC ) , consistently with the interpretation that a functional CreBC system is required for one-step high level ( clinical ) β-lactam resistance development in P . aeruginosa . Nevertheless , in contrast to the previous experiences with the BlrAB system in Aeromonas Spp . , CreBC mediated resistance was not directly driven by an effect on ampC expression , since 1A1ΔCreBC ( despite showing a drastic reduction of resistance ) had similar ( still overexpressed ) ampC levels than the parent 1A1 ( Table 1 ) . Moreover , β-lactamase activity was also similar ( data not shown ) , showing that apparently the effect is neither produced by posttranscriptional modification of AmpC . Therefore , although we demonstrate that mutation of PBP4 specifically activates the CreBC two-component regulator , and that this event plays a major role in β-lactam resistance , the underlying mechanism is still uncertain . Despite only creD showed a modified expression greater 2-fold in the transcriptome analysis , even small modifications of expression of genes involved in outer membrane permeability , antibiotic efflux or general metabolism could significantly enhance the effect of AmpC overexpression and thus β-lactam resistance . In any case , our findings indicate that the nomenclature for this two-component system in P . aeruginosa should be changed to follow that used for Aeromonas Spp . ( Blr , standing for β-lactam resistance ) and not that used in E . coli ( Cre , standing for carbon-source responsive ) [18] , [29]–[31] . Figure 2 summarizes the current knowledge on P . aeruginosa ampC regulation , peptidoglycan recycling , and the described similarities and differences of the β-lactam resistance response driven by AmpD inactivation or PBP4 mutation . To find out whether PBP4 mutations and the linked CreBCD mediated resistance documented for in vitro and in vivo mutants occurred also in natural human infections , we investigated a previously described collection of clinical strains [6] . This collection included 10 isogenic pairs of ceftazidime susceptible and resistant clinical isolates obtained from 10 Intensive Care Unit patients . All patients had severe P . aeruginosa infections that were treated with β-lactams , experiencing therapy failure due to resistance development . In all cases , the subsequent ceftazidime resistant isolate showed AmpC hyperproduction , but only in four of them could the resistance phenotype be attributed to known mechanisms ( ampD mutations ) [6] . As shown in Table 3 , all 6 remaining ceftazidime isolates contained mutations in dacB ( not present in the preceding isogenic susceptible isolate ) . Interestingly , two of the isolates ( despite them being genetically distinct ) have sustained the same PBP4 mutation ( T428P ) ; this same mutation was found in one of the PAO1 in vivo mutants ( Table S1 ) and involves a conserved residue close to the KTG motif [16] . Furthermore , consistent with the findings for in vitro and in vivo mutants , all six natural PBP4 mutants overexpressed creD ( 2 . 8–38 fold higher expression than their respective wild-type isolates ) ( Table 3 ) . The inactivation of creBC significantly reduced ceftazidime resistance in all but one of the natural PBP4 mutants ( Table 3 ) . On the other hand , the expression of creD was not modified in the four clinical strains containing only mutations in ampD ( not shown ) . Using P . aeruginosa as a model organism , we have shown that the most prevalent mutations causing immediate onset of high level β-lactam resistance are found in the dacB gene , encoding the nonessential PBP4 . This is the first demonstration of the acquisition of β-lactam resistance through such a mechanism . All the previous examples by which PBP-mediated resistance develops involve modified enzymes showing low affinity for β-lactams ( target modification ) [2] . Even though inactivation of the classical AmpC negative regulator AmpD upregulates AmpC , only mutations in dacB confer high level ( clinical ) β-lactam resistance . The inactivation of PBP4 is found to trigger an AmpR-dependent overproduction of the chromosomal β-lactamase AmpC , and the specific activation of the CreBC ( BlrAB ) two-component regulator , which in turn plays a major role in the β-lactam resistance response . This interplay between mutation of supposed antibiotic targets , production of antibiotic inactivating enzymes , and global regulators , is an unexpected layer of complexity of β-lactam resistance biology , which provides new perspectives on potential antibiotic targets for the treatment of infectious diseases . Since all the components of the described resistance mechanism ( dacB , ampC , ampR , and creBCD ) are found in the genomes of many GNR ( E . coli might be an example of exception since ampR is not present in this species [14] ) , the results presented here are expected to have broad implications for the development of new antimicrobial compounds . Particularly , the CreBCD system is envisaged as an attractive candidate target to develop molecules capable of reducing the development of resistance when used together with β-lactam antibiotics .
A complete list of laboratory strains and plasmids used or constructed in this study is provided as Table S2 . A previously described [19] collection of 36 independent ceftazidime resistant mutants was used . These mutants were obtained either in vitro ( one-step spontaneous mutants ) or in vivo ( after 3 days of treatment with humanized ceftazidime regimen in mouse model of lung infection ) , at two ceftazidime concentrations ( 4 and 16 µg/ml ) , and from the wild-type strain PAO1 ( normal mutation rate supply ) or its mutS deficient derivative PAOΔmutS ( high mutation rate supply ) . Additionally , a previously reported [6] collection of 10 isogenic pairs of ceftazidime susceptible and resistant clinical isolates obtained from 10 Intensive Care Unit patients suffering from severe P . aeruginosa infections was used . P . aeruginosa single or multiple knockout mutants in ampD , ampE , ampR , ampC , creBC , creD , or dacB were constructed using the Cre-lox system for gene deletion and antibiotic resistance marker recycling [32] . Upstream and downstream PCR products ( primers used provided as Table S3 ) of each gene were digested with either BamHI or EcoRI and HindIII and cloned by a three way ligation into pEX100Tlink deleted for the HindIII site and opened by EcoRI and BamHI . The resulting plasmids were transformed into E . coli XL1Blue strain and transformants were selected in 30 µg/ml ampicillin LB agar plates . The lox flanked gentamicin resistance cassette ( aac1 ) obtained by HindIII restriction of plasmid pUCGmlox was cloned into the single site for this enzyme formed by the ligation of the two flanking fragments . The resulting plasmids were again transformed into E . coli XL1Blue strain and transformants were selected in 30 µg/ml ampicillin-5 µg/ml gentamicin LB agar plates . Plasmids were then transformed into the E . coli S17-1 helper strain . Knockout mutants were generated by conjugation followed by selection of double recombinants using 5% sucrose-1 µg/ml cefotaxime-30 µg/ml gentamicin LB agar plates . Double recombinants were checked first screening for carbellicin ( 200 µg/ml ) susceptibility and afterwards by PCR amplification and sequencing . For the recycling of the gentamicin resistance cassettes , plasmid pCM157 was electroporated into the different mutants . Transformants were selected in 250 µg/ml tetracycline LB agar plates . One transformant for each mutant was grown overnight in 250 µg/ml tetracycline LB broth in order to allow the expression of the cre recombinase . Plasmid pCM157 was then cured from the strains by successive passages on LB broth . Selected colonies were then screened for tetracycline ( 250 µg/ml ) and gentamicin ( 30 µg/ml ) susceptibility and checked by PCR amplification and DNA sequencing . In order to explore the β-lactam resistance mechanisms in the above described collection of bacterial strains , ampD , ampE , ampR , ampDh2 , ampDh3 , creB , creC , creD and dacB genes were amplified by PCR , using primers described in Table S3 , and fully sequenced . All mutations detected were checked by sequencing a fresh independent PCR product . Sequencing reactions were performed with the BigDye Terminator Kit ( PE Applied Biosystems , Foster City , Calif . ) and sequences were analyzed on an ABI prism 3100 DNA sequencer ( PE Applied Biosystems ) . Resulting sequences were then compared ( www . ncbi . nih . gov/BLAST ) with those of the wild-type PAO1 strain [33] , [34] . The levels of expression of ampC , ampD , ampE , ampDh2 , ampDh3 and creD were determined by real time RT-PCR with and without cefoxitin induction . Total RNA from logarithmic-phase-grown cultures ( grown with and without 50 µg/ml cefoxitin ) was obtained with the RNeasy Mini Kit ( QIAGEN , Hilden , Germany ) and treated with 2 U of TURBO DNase ( Ambion ) for 30 min at 37°C to remove contaminating DNA . The reaction was stopped by the addition of 5 µl of DNase inactivation reagent and the samples were adjusted to a final concentration of 50 ng/µl . A 500 ng sample of purified RNA was then used for one-step reverse transcription and real-time PCR amplification using the QuantiTect SYBR Green RT-PCR Kit ( QIAGEN , Hilden , Germany ) in a SmartCycler II ( Cepheid , Sunnyvale , CA ) . The primers listed in Table S3 were used for amplification of ampC , ampD , ampE , ampDh2 , ampDh3 , creD , and rpsL ( used as reference to normalize the relative amount of mRNA ) . In all cases , the mean values of relative mRNA expression obtained in at least three independent duplicate experiments were considered . Minimal inhibitory concentrations ( MICs ) for ceftazidime , cefepime , ticarcillin , piperacillin , piperacillin/tazobactam , aztreonam , imipenem , meropenem ciprofloxacin , tobramycin , tetracycline , chloramphenicol and colistin were determined in Müller-Hinton ( MH ) agar plates using E-test strips ( AB Biodisk , Sweden ) following the manufacturers recommendations . β-lactamase specific activity ( nanomoles of nitrocefin hydrolyzed per minute per milligram of protein ) was determined spectrophotometrically on crude sonic extracts as previously described [6] . To determine the β-lactamase specific activity after induction , before the preparation of the crude sonic extracts , the strains were grown in the presence of 50 µg/ml cefoxitin for 3 h . In all cases , the mean β-lactamase activity values obtained in three independent experiments were considered . Complementation experiments were performed following previously described protocols [6] . Briefly , plasmids pUCPAD ( harboring the wild-type ampD gene ) , pUCPADE ( harboring the complete wild-type ampDE operon ) or pUCP24 ( cloning vector ) were electroporated into the different β-lactam resistant strains or PAO1 ( as control ) . Additionally , plasmids pUCPADE2A1 , pUCPADE1C5 or pUCPADE2C2 containing a non functional ampD and a wild-type ampE were electroporated in selected mutants . Finally , complementation experiments using the cloned wild-type dacB gene ( pUCPdB ) were also performed in selected strains . Transformants were selected in 50 µg/ml gentamicin LB agar plates . Complementation was considered positive when the MICs of ceftazidime for the transformants were at least 3 two-fold dilutions lower than those of the parent strains . The rates of mutation to high level ( 16 µg/ml ) ceftazidime resistance were estimated following previously described protocols ( 19 ) . To determine the effect on bacterial fitness of particular resistance mutations , in vitro ( LB growth ) and in vivo ( murine model of systemic infection ) competition experiments were performed following previously described procedures [27] . Median Competition Indexes ( CIs ) , defined as the mutant/wild-type ratio , were calculated from at least 8 independent experiments . Three independent replicates of PAO1 and of two selected ceftazidime resistant mutants ( 1A1 and 2A2 ) were grown in 10 ml of LB broth in a 50-ml baffled flask vigorously shaken at 37°C to an optical density at 600 nm ( OD600 ) of 1 . The cells were collected by centrifugation ( 8 , 000 g for 5 min at 4°C ) and total RNA was isolated using the RNeasy minikit ( QIAGEN ) following the manufacturer's instructions . RNA was dissolved in water and treated with 2 U of TURBO DNase ( Ambion ) for 30 min at 37°C to remove contaminating DNA . The reaction was stopped by the addition of 5 µl of DNase inactivation reagent . Ten micrograms of total RNA were checked by running on an agarose gel prior to cDNA synthesis . cDNA synthesis , fragmentation , labeling and hybridization were performed according to the Affymetrix GeneChip P . aeruginosa genome array expression analysis protocol . Expression analysis was performed as described previously [35] . Only transcripts showing higher than two-fold increases or decreases were considered as differentially expressed . In all cases the PPDE ( posterior probability for differential expression ) was between 0 . 999 and 1 . In order to detect the presence of mutations in genes yet unknown to be involved in β-lactam resistance , a whole-genome analysis approach was followed . For this purpose , four ceftazidime resistant mutants were analyzed and compared with wild-type PAO1 using a recently described microarray for the discovery of single nucleotide polymorphisms ( SNPs ) in P . aeruginosa [20] . Cultures were grown in brain-heart infusion ( BHI ) medium for 12 h at 37°C in shaking glass flasks at 180 rpm and genomic DNA was isolated using the DNeasy Blood & Tissue Kit ( Qiagen ) . Cell lysates were treated with RNase I ( Qiagen ) to prevent accidental carryover of RNA to the microarray . Genomic DNA was partially digested with DNase I ( Amersham Biosciences , Piscataway , NJ ) to a fragment size of ∼50–250 bp , confirmed by gel electrophoresis , and fragments were labeled at the 3′-ends with biotin-ddUTP ( Roche Diagnostics , Indianapolis , IN ) using Terminal deoxynucleotidyl transferase ( Roche ) . Samples were hybridized to an identical lot of PATA1 arrays [20] for 16 hours at 50°C . After hybridization the microarrays were washed , stained with SA-PE and read using an Affymetrix GeneChip fluidic station and scanner according to Affymetrix standard protocols ( Affymetrix , Santa Clara , CA ) . Analysis of microarray data was performed using the Affymetrix GCOS 1 . 4 to generate the raw data files ( cel data ) . The raw data files were further analyzed using ‘Tiling Analysis Software’ ( TAS ) version 1 . 1 by Affymetrix . | Decades after their discovery , β-lactams remain key components of our antimicrobial armamentarium for the treatment of infectious diseases . Nevertheless , resistance to these antibiotics is increasing alarmingly . There are two major bacterial strategies to develop resistance to β-lactam antibiotics: the production of enzymes that inactivate them ( β-lactamases ) , or the modification of their targets in the cell wall ( the essential penicillin-binding proteins , PBPs ) . Using the pathogen Pseudomonas aeruginosa as a model microorganism , we show that high-level ( clinical ) β-lactam resistance in vitro and in vivo frequently occurs through a previously unrecognized , totally distinct resistance pathway , driven by the mutational inactivation of a nonessential PBP ( PBP4 ) that behaves as a trap target for β-lactams . We show that mutation of this PBP determines a highly efficient and complex β-lactam resistance response , triggering overproduction of the chromosomal β-lactamase AmpC and the specific activation of a two-component regulator , which in turn plays a key role in resistance . These findings are a major step forward in our understanding of β-lactam resistance biology , and , more importantly , they open up new perspectives on potential antibiotic targets for the treatment of infectious diseases . | [
"Abstract",
"Introduction",
"Results/Discussion",
"Materials",
"and",
"Methods"
] | [
"microbiology/microbial",
"physiology",
"and",
"metabolism",
"microbiology/microbial",
"evolution",
"and",
"genomics",
"microbiology/medical",
"microbiology",
"microbiology"
] | 2009 | β-Lactam Resistance Response Triggered by Inactivation of a Nonessential Penicillin-Binding Protein |
Plant pathogens and parasites are a major threat to global food security . Plant parasitism has arisen four times independently within the phylum Nematoda , resulting in at least one parasite of every major food crop in the world . Some species within the most economically important order ( Tylenchida ) secrete proteins termed effectors into their host during infection to re-programme host development and immunity . The precise detail of how nematodes evolve new effectors is not clear . Here we reconstruct the evolutionary history of a novel effector gene family . We show that during the evolution of plant parasitism in the Tylenchida , the housekeeping glutathione synthetase ( GS ) gene was extensively replicated . New GS paralogues acquired multiple dorsal gland promoter elements , altered spatial expression to the secretory dorsal gland , altered temporal expression to primarily parasitic stages , and gained a signal peptide for secretion . The gene products are delivered into the host plant cell during infection , giving rise to “GS-like effectors” . Remarkably , by solving the structure of GS-like effectors we show that during this process they have also diversified in biochemical activity , and likely represent the founding members of a novel class of GS-like enzyme . Our results demonstrate the re-purposing of an endogenous housekeeping gene to form a family of effectors with modified functions . We anticipate that our discovery will be a blueprint to understand the evolution of other plant-parasitic nematode effectors , and the foundation to uncover a novel enzymatic function .
The ability of nematodes to exploit living plants as a food resource has arisen independently in four of the twelve major lineages of the phylum Nematoda [1] . As a result , plant-parasitic nematodes occupy a diverse range of niches and climates , and infect a wide range of host species globally . Clade 12 of the phylum encompasses representatives of all major modes of parasitism; migratory ectoparasites , migratory endoparasites , and the most economically important and highly specialized obligate biotrophs—the sedentary endoparasites [2] . The latter induce the re-differentiation of root cells to form a unique nutrient-rich feeding site which is maintained for several weeks in a prolonged biotrophic interaction . For the cyst and reniform nematodes this takes the form of a large syncytium that arises through local cell wall dissolution and fusion of neighbouring protoplasts . Nematodes deploy hundreds of effector proteins to induce profound molecular and physiological changes associated with feeding site induction and maintenance . The majority of all described effectors are secreted from three pharyngeal gland cells ( one dorsal and two subventral ) through a hollow , protrusible needle-like stylet , into the plant . While the basis for the evolution of nematode parasitism is largely unresolved and widely debated [3 , 4] , it is likely that a series of evolutionary transitions gave rise to the biologically complex sedentary plant endoparasites [5 , 6] . Surprisingly little is known about the genetic changes that occurred with these transitions . In general , parasites lose functions as they further rely on their host . For the sedentary endoparasitic cyst nematodes , this is evidenced by a reduction in genes involved in detoxification of xenobiotic compounds , and the absence of whole classes of antibacterial and antifungal genes [7] . However , concurrent with this process , cyst nematodes have evolved a large repertoire of effectors that facilitate their remarkable abilities to suppress plant immunity and induce plant cells to re-differentiate into a novel tissue . The evolution of sedentary endoparasitism must therefore be additionally characterized by acquisition of novel functions . Both acquisition of new genes by horizontal transfer [8] and selective expansion of particular gene families [9] are associated with the evolution of a parasitic lifestyle . For example , a large expansion of the astacin protease and CAP gene families coincided with the emergence of parasitism in the animal parasitic Strongyloides nematodes [10] whilst gene duplication is proposed to have been an important driver of parasitism in the Orobanchaceae plant parasites [11] . Examples of such expansions are also present in the genome of the potato cyst nematode Globodera pallida . The most unusual expansion comprises more than 50 predicted genes with similarity to glutathione synthetase enzymes [12]: almost all eukaryotes possess only a single glutathione synthetase gene . Glutathione , the tri-peptide γ-L-glutamyl-cysteinyl-glycine , is the main low molecular weight thiol antioxidant in both plant and animal cells , often present at millimolar concentrations in vivo [13] . It is synthesised in a two-step , ATP-dependent , process: glutamate-cysteine ligase ( GCL ) catalyses the formation of γ-glutamylcysteine ( γ-EC ) from glutamate and cysteine , followed by the addition of glycine by glutathione synthetase ( GS ) to form glutathione . Glutathione has a fundamental , multifunctional role in modulating the redox status of cells , protecting them against oxidants and electrophiles , and detoxifying xenobiotics [14] . In both plants and animals , glutathione has a role in cellular defence responses against abiotic and biotic stress [15 , 16] . Glutathione has been particularly implicated in plant responses to pathogens . Its abundance decreases during compatible interactions but it accumulates in response to avirulent pathogens [17] and can induce the expression of plant defence genes [18 , 19] . Glutathione is also involved in biotic stress signalling [20] , in particular NPR1-dependent/independent salicylic acid ( SA ) -mediated pathways [21 , 22] and plants with reduced glutathione levels are generally more susceptible to pathogens and herbivorous insects [23 , 24] . Given the diverse functions of glutathione it is clear that an expanded family of glutathione synthetases could be involved in numerous aspects of the plant-nematode interaction . Here we demonstrate that the expansion is not restricted to G . pallida and multiple subsequent expansions of GS genes have co-occurred with the evolution of complexity in plant-nematode interactions . The most recent expansion has been coupled with a diversification of structure and biochemical function that has given rise to enzymes that are introduced into the host cell , and likely possess novel substrate specificities .
We analysed the genomes and/or transcriptomes of eleven plant-parasitic and non-plant-parasitic nematode species ( Caenorhabditis elegans , Bursaphelenchus xylophilus , Longidorus elongatus , Pratylenchus penetrans , Meloidogyne incognita , Nacobbus aberrans , Rotylenchulus reniformis , Globodera rostochiensis , G . pallida , Heterodera schachtii and H . avenae ) to explore the expansion of genes encoding glutathione synthetase-like domains ( Pfams PF03917 and PF03199 –Fig 1 ) . Although almost all plant and animal genomes encode only a single housekeeping glutathione synthetase gene ( GS ) , some plant-parasitic nematode genomes encode an unprecedented number ( up to ~70 in R . reniformis ) . A Bayesian phylogeny inferred from an alignment of all 180 GS-like loci from these eleven species reveals two major expansion events , resulting in three Clades ( Fig 1; complete GS-like gene sequences are available under Dryad accession doi:10 . 5061/dryad . 7vd0160 ) . Clade 1 contains one sequence from each nematode in the phylogeny ( with the exception of M . incognita that , due to its polyploid genome , contributes two genes to the Clade [25] ) . It includes the only sequence from the free living nematode C . elegans , and the relative topology of the sequences in Clade 1 is similar to the most recent single and multi-gene phylogenies of plant parasitic nematodes for those species included [1 , 26] . Taken together , this suggests that Clade 1 contains the housekeeping progenitor GS of each species . The first expansion of GS-like genes ( Clade 2 ) was present in the last common ancestor of present day migratory and sedentary plant endoparasites belonging to the order Tylenchida . Species vary in the number of Clade 2 genes from two ( P . penetrans and M . incognita ) to 12 ( R . reniformis ) , with a mean , mode , and median of ~5 . Clade 2 does not include sequences from the free living nematode C . elegans , the migratory ectoparasite L . elongatus or the non-Tylenchid migratory endoparasite B . xylophilus . We note that many Clade 2 GS share a short and somewhat variable C-terminal extension of the approximate consensus sequence P[A|S][A|S][E|Q][F|L] , which has no known function ( S1 Fig ) . This peptide is absent in all other clades and is not recognised as a canonical signal by TargetP . A second larger expansion of GS-like genes ( Clade 3 ) was present in the last common ancestor of the syncytia-forming cyst and reniform nematodes , which both lie on one side of a major bifurcation in the evolution of plant-sedentary endoparasitism ( Fig 1 ) . On average , sequences in Clade 3 share 36% protein identity with one another , and are much more diverse than those of Clades 1 and 2 ( 51% and 52% identity respectively ) despite encompassing a narrower range of species . The fact that 68% of G . pallida and G . rostochiensis Clade 3 GS-like genes ( 26/38 ) are more similar to a Clade 3 GS from another cyst nematode in the phylogeny than they are to another Clade 3 GS from their own species , suggests that the majority of the diversity in cyst nematode Clade 3 GS was probably present in their last common ancestor . The R . reniformis , H . avenae , and H . schachtii GS-like gene complements were largely or entirely assembled from de novo transcriptome data rather than genomic data , precluding similar conclusions . Nevertheless we note that 8/14 H . avenae Clade 3 GS are more similar to a Clade 3 GS from other cyst nematodes than they are to other Clade 3 GS from their own species . In contrast all R . reniformis Clade 3 GS are contained within two sub-clades , and are more similar to other R . reniformis Clade 3 GS than Clade 3 GS from the cyst nematodes . Given that species which have GS-like genes from the first expansion ( Clade 2 ) have retained the progenitor housekeeping gene ( Clade 1 ) , and those that have GS-like genes from the second expansion ( Clade 3 ) have retained both Clade 1 and 2 , this suggests that the role of each expansion does not supersede its predecessor ( s ) . They likely represent gain of novel function during the evolution of endoparasitism . Each expansion of GS-like genes has a distinctive temporal expression pattern compared to its predecessor ( Fig 1A ) . Clade 2 GS-like genes exhibit up-regulation predominantly in pre/non-parasitic stages , while Clade 3 GS-like genes are highly up-regulated , and often specifically expressed , during the plant-parasitic stages of endoparasitism . Given that the average within-species identity of G . pallida GS-like coding sequences is 49% , we can be confident in the distinct expression patterns of each Clade by RNAseq mapping . The disparity in differential temporal expression of GS-like genes further supports their functional diversification at different points of the life cycle . More detailed transcriptional analysis of all G . pallida GS-like genes throughout the lifecycle demonstrates that those of Clade 3 are highly up-regulated upon early interaction with the host plant and this expression is generally maintained throughout the parasitic stage ( 7 , 14 , 21 , 28 , and 35 days post infection; S2 Fig ) . This is strongly indicative that the most recently evolved Clade 3 GS-like paralogues are involved in prolonged sedentary endoparasitism . Analysing the promoters of G . pallida and G . rostochiensis GS-like genes , we discovered that specifically those in Clade 3 harbour multiple copies of the DOG box , a promoter motif associated with dorsal gland cell expression in Globodera spp . [27 , 28] . Globodera Clade 1 , 2 and 3 GS-like gene promoters contain on average 0 . 5 , 0 . 2 , and 1 . 31 DOG boxes respectively ( S1 Table ) . While there is no correlation between the number of DOG box motifs per promoter and temporal expression of the corresponding gene ( R2 = 0 . 03 and [28] ) , Clade 3 GS-like genes with >1 DOG box in the first 1000 bp upstream of the coding start site are approximately twice as highly up-regulated during parasitism as those in Clade 3 without DOG boxes ( 24-fold ( n = 16 ) , 13-fold ( n = 6 ) ) . The presence of the DOG box in the promoter of previously uncharacterised genes is used to predict dorsal gland cell-expressed effectors in cyst nematodes [27] . Consistent with DOG box enrichment , all GS-like genes from Clade 3 analysed by in situ hybridisation were highly and specifically expressed in the secretory dorsal gland cells in a range of parasitic stage nematodes ( Fig 2A and S3 Fig ) , while Clade 1 and Clade 2 GS-like genes were expressed in both female and male nematodes , with often preferential expression in the intestine ( Fig 2A and S3 Fig . Gland cell expression holds true for both major groups of nematodes that encode Clade 3 GS , the cyst and the reniform nematodes ( Fig 2A ) . The gland cells are the major effector-producing tissue , and secreted proteins expressed in the dorsal gland cell are delivered into the host plant [29 , 30] . Strikingly , almost all Clade 3 GS-like genes , but none of those from Clades 1 and 2 , encode a canonical N-terminal secretion signal ( Fig 1A , black bars ) . These 102 putatively secreted Clade 3 GS-like proteins are the only known GS from the plant or animal kingdoms that possess such a signal . While we cannot rule out the possibility that Clade 3 GS proteins that do not encode a signal peptide are in fact secreted by non-canonical pathways , as presumed for some nematode effectors ( eg . [20 , 31] ) , we restrict our analysis in Clade 3 to signal peptide-encoding GS-like genes . Taken together , a clear distinction can be made between Clade 3 GS-like genes and those of Clades 1 and 2 based on their promoters , their spatial expression patterns , the presence of a signal peptide , and the likelihood of the sequence being upregulated in parasitic stages . Tissue-specific expression of a putatively secreted protein in secretory dorsal gland cells is a strong indicator of a nematode effector that is delivered in planta . To confirm this , an affinity-purified polyclonal antibody was raised against recombinant Gpa-GSS17 and shown to be specific for this protein ( Fig 2C ) . Using this antibody we are able to detect the abundant presence of Gpa-GSS17 in the large dorsal gland cell of parasitic stage G . pallida nematodes , the cytoplasmic gland extension , and the ampulla at the base of the stylet where secretions accumulate prior to their release ( Fig 2B ) . The same native protein was delivered into the host across the plasma membrane , and is localized within the syncytial feeding site induced by the nematode in potato roots ( Fig 2B ) . No similar fluorescence was observed with the FITC-labelled 2o antibody control . Hereafter , Clade 3 genes are thus referred to as GS-like effector genes . While it is highly unusual for a 100 kilodalton pathogen effector to be translocated to the host cell cytoplasm ( GS-like effectors are obligate homo-dimers of 50 kDa per subunit , S4 Fig ) , plant parasitic nematodes clearly have the ability to construct organelle-sized structures inside the host cell at the plasma membrane where it meets the stylet orifice ( reviewed in [32] ) . The demonstrations that a native GS-like effector protein is secreted from the nematode , delivered into the host , translocated across the host plasma membrane , and localised within the host cell during parasitism are strong evidence of involvement in parasitism . Taken together , this illustrates how these plant-parasitic nematodes have exploited multiple gene gain events to deploy a novel effector family during parasitism . Interestingly , this programme of effector evolution is potentially more broadly applicable to other well-documented gene gain events in plant-parasitic nematodes: horizontally transferred genes must acquire a similar set of genetic attributes in order to be deployed as effectors during parasitism [33] . To explore the catalytic capacity of nematode GS-like genes , the G . pallida representative of Clade 1 ( Gpa-gss1 ) , one representative from Clade 2 ( Gpa-gss5 ) and a number from Clade 3 ( Gpa-gss12 , Gpa-gss17 , Gpa-gss22 , Gpa-gss24 and Gpa-gss30 ) were heterologously expressed in , and their proteins purified from , bacteria ( Fig 3A ) . For comparison , the only GS gene in the corresponding plant host of G . pallida ( Solanum tuberosum , St-gss1 ) , was similarly expressed and its product purified . All GS-like proteins were purified as obligate homo-dimeric pairs ( S4 Fig ) . Purified GS were tested for canonical glutathione ( Glu-Cys-Gly ) synthetic capacity using a spectrophotometric assay where phosphate release from ATP is used as a molar equivalent proxy for glutathione synthesis from the substrates γ-glutamyl-cysteine ( γ-EC ) and glycine . Using this approach , the maximum rate of the host GS ( St-GSS1 ) phosphate release was 7532 ( ±1358 ) umol/mg/min ( S2 Table ) , consistent with a previous report for Arabidopsis GS of ~7500 umol/mg/min [34] , demonstrating the suitability of the assay . Remarkably , each stage of the evolutionary process that gave rise to GS-like effectors has witnessed at least a 10-fold reduction in apparent glutathione synthetic rate , such that Clade 3 GS-like proteins have not retained canonical enzyme activity ( Fig 3 and S2 Table ) . We argue this loss of canonical activity is probably associated with a gain of non-canonical activity . We have analysed 180 GS-like protein sequences , each approximately 500 amino acids in length , from 11 species across the phylum . Within the context of an average sequence identity of only ~34% , there are just 5 residues that are absolutely conserved in all 180 sequences: three of these are in the ATP binding pocket , and one of these is required for catalysis . Perfect conservation of the catalytic residue in itself implies individual catalytic functionality . Given that GS effectors have no appreciable rate of ATP turnover when supplied with canonical substrates , yet all display perfect conservation at the catalytic arginine , we infer that GS-like effectors possess a distinct catalytic activity , which may involve alternative substrate ( s ) . GS-like enzymes that vary in substrate usage have been described in plants , yet in all cases this variation is restricted to the terminal amino acid: the same γ-EC backbone is universally used as a scaffold [35 , 36] . For example , the homo-glutathione synthetase ( hGS ) of Glycine max ( soya bean ) catalyses the addition of β-alanine to γ-EC giving rise to homoglutathione ( hGSH , γ-glu-cys-β-ala ) . The ability of purified GS-like proteins to incorporate a range of natural and non-natural terminal amino acids onto the γ-EC backbone was tested . St-GSS1 , Clade 1 Gpa-GSS1 , and Clade 2 Gpa-GSS5 all exhibited a strong preference for glycine , and a high affinity for γ-EC ( S5 Fig ) . In contrast , none of the Clade 3 GS-like effectors accepted any of the tested terminal amino acid substrates in combination with the γ-EC backbone . One possible explanation for this is that Clade 3 GS-like effectors represent a novel diversification at the site of the di-peptide acceptor . To create a structural basis for the exploration of novel substrate specificities , we initially solved the first crystal structure of a canonical plant GS ( St-GSS1 , S . tuberosum , host of G . pallida ) in complex with ADP and the canonical di-peptide substrate γ-EC ( 2 . 5 Å , PDB 5OES ) . The crystal structures of two non-canonical parasite GS-like effectors were subsequently solved: Gpa-GSS30 in its apoform ( 2 . 6 Å , PDB 5OET ) and Gpa-GSS22 in both its apoform ( 2 . 2 Å , PDB 5OEV ) and ADP-bound closed conformation ( 2 . 6 Å , PDB 5OEU , S3 Table ) . Comparison between Gpa-GSS22-open and–closed reveals a functioning ATP grasp fold ( Fig 4 ) . Residues in the ATP binding pocket of St-GSS1 are highly conserved in sequence and position in the Gpa-GSS22 structure ( 13/15 residues , S4 Table ) , and similarly conserved in sequence across G . pallida Clade 3 ( structure guided alignment , ~12/15 , n = 24 , S4 Table ) . In contrast , there is considerably more variation in the di-peptide binding pocket of G . pallida GS-like effectors , yet , this variation is not evenly distributed around the pocket . The position of γ-EC in St-GSS1 is coordinated at both the glutamate and the cysteine residue . Two of the three residues that coordinate the cysteine interact with the C-alpha backbone . Corresponding residues in all G . pallida GS-like effectors are not conserved in sequence but are preferentially small and uncharged ( Fig 3B inset ) , while the third residue , arginine , is 100% conserved ( Fig 3B and S4 Table ) : consistent with permitting interactions with a cysteine residue . However , the glutamate of the di-peptide is exclusively coordinated by interactions with charged side chains of residues in the binding pocket ( R , E , 2xQ , N , and Y ) . All six of these positions are substituted in Gpa-GSS22 and Gpa-GSS30 , thus suggesting that the lack of canonical activity is because γ-EC is not accepted despite the potential for cysteine to be accommodated ( Fig 3C and S4 Table ) . Variability in the glutamate portion of the di-peptide binding pocket is ubiquitous in G . pallida Clade 3 GS: of the 24 GS-like effectors there are 21 different amino acid compositions in these 6 positions , not one of which is canonical ( S4 Table ) . Such diversity is highly unusual among Eukaryotes: a canonical arrangement has been conserved in GS enzymes for ~1 billion years of evolution in three kingdoms of life ( Fig 5 , Plantae ( St-GSS1-closed , PDB 5OES ) , Fungi ( Saccharomyces cerevisiae , PDB 1M0T ) , and Animalia ( Homo sapiens GSS1 , PDB 2HGS ) ) . In summary , conserved amino acids are only absent from one half of the acceptor di-peptide binding pocket . The space that would accommodate the cysteine thiol—the “active residue” of glutathione—is well conserved in sequence and structure , and is non-variable in all GS-like effectors tested . Coupled with the high degree of conservation in the ATP-binding pocket , the functioning ATP-grasp fold , and the perfect conservation of a catalytic residue , these data support the hypothesis that the loss of canonical glutathione synthetic activity is associated with a gain of non-canonical activity . The reaction product is probably a thiol-containing compound: ultimately implicating novel thiol biology in plant-nematode parasitism . To determine the extent of thiol involvement in syncytia induced by cyst nematodes , we initially employed a qualitative analysis to specifically stain and visualize free thiols in plant tissue during infection . We used the Arabidopsis-Heterodera schachtii pathosystem because the thin and transparent host roots are amenable to such studies , whereas those of other syncytial-forming nematodes ( e . g . potato-G . pallida ) are not . The H . schachtii transcriptome encodes a number of putatively secreted Clade 3 GS-like effectors ( Fig 1 ) . Using ThiolTracker Violet we discovered that thiols are abundant in , and largely localized to , the cytoplasm of syncytia induced by H . schachtii , throughout infection ( Fig 6A ) . Following this support , short sections of infected potato root harbouring syncytia of G . pallida at 21 days post infection were collected , separated from their corresponding nematode , and both samples retained for analysis . Several hundred infection sites and nematodes were collected in this manner and pooled . Control uninfected adjacent root tissue was collected from the same plants . Low molecular weight ( LMW ) thiols present in the three samples were extracted , derivatized with mono-bromobimane and analysed by Hydrophilic Interaction Liquid Chromatography ( HILIC , Fig 6C ) . We note that any increase in syncytial thiol abundance is not explained by an increase in glutathione , but a series of other LMW thiols ( Fig 6C ) . Surprisingly , analysis of the area under each curve allowed us to roughly estimate that glutathione accounts for only approximately 2 . 5% of LMW thiols detected using this HPLC protocol in potato control roots , and 2 . 7% in syncytial segments . Furthermore , some of the novel LMW thiols are not present in uninfected potato root tissue , not present in nematodes , but only present in syncytia ( Fig 6D ) . Although the detected novel thiols were recalcitrant to analysis by mass spectrometry ( e . g . Fig 6D ) , their lower retention by HILIC allows us to infer they are more hydrophobic than glutathione . Our demonstration of GS-like effectors and abundant thiols in syncytia , coupled with the importance of rbohD-dependent ROS production in syncytial development [37] , point towards fine-tuned redox homeostasis during parasitism . We cannot target GS-like effectors in the nematode to disrupt redox homeostasis in planta . The lack of a transformation system precludes the generation of gene knockouts whilst the likely functional redundancy of GS-like effectors and their low sequence similarity would require the combined use of one RNAi construct per effector ( n≈20 ) in order to achieve host-induced gene silencing . We therefore exploited the availability of Arabidopsis plants mutant in the first step of glutathione synthesis ( GSH-1 ( pad2-1 ) ) and thus compromised in the endogenous glutathione portion of the syncytial thiol pool by approximately 50–80% [38] . Both cyst nematode and syncytial development is significantly retarded at 10–12 days post infection in pad2-1 compared to wild type ( WT ) plants ( Student’s T-test p ≤ 0 . 001 , n = 147 and 82 respectively , Fig 7A , 7B and 7D ) . Specifically , syncytia and nematodes supported by pad2-1 plants are both on average ~50% the size of WT . However , despite that feeding site size and nematode size significantly co-vary in pad2-1 and WT ( Pearson’s correlation , p ≤ 0 . 05 and 0 . 001 , n = 49 and 66 respectively ) , the correlation is weak: most of the variation in nematode size ( 83–89% ) is not explained by syncytium size ( Fig 7C ) . This suggests that the lack of plant glutathione synthesis is associated with at least two linked but largely unrelated processes during infection which individually contribute to nematode size , and feeding site size . Furthermore , in pad2-1 Arabidopsis , localised necrosis could often be seen surrounding syncytia , and aborted syncytia were common ( Fig 7D and S6 Fig ) . Given that it is technically intractable to measure aborted syncytia and those obscured by localised necrosis , the effect may be even greater than we report . In a parallel approach , we reduced glutathione synthetic capacity to approximately 16% of wild-type levels by the exogenous application of 1 mM L-buthionine-sulfoximine ( BSO ) . BSO is an irreversible chemical inhibitor of the first step in the glutathione synthesis pathway [38] . At this dramatically reduced level , infective nematodes fail to maintain a compatible interaction . Under these conditions nematodes are able to penetrate root tissue , migrate to the vascular cylinder and initiate the formation of a syncytium . However , unlike the interaction with pad2-1 , syncytia induced in plants treated with BSO are always aborted ( Fig 7E ) . Necrosis occurs in the area of the developing syncytium , as evidenced by propidium iodide staining , and often spreads non-distal to the site of infection ( Fig 7F ) . This phenotype is similar to , but more frequent and severe than , the necrosis seen in pad2-1 plants . Necrotic patches on both pad2-1 plants and BSO treated wild-type plants were associated only with nematode infection sites , while uninfected root was comparable to that of untreated wild-type plants , albeit with reduced proliferation ( S7 Fig ) . We cannot rule out a direct effect of BSO on the nematodes during infection , however , infective stage nematodes incubated for 48 hours on water agar plates containing 1 mM BSO were largely unaffected in mortality or motility ( Mann-Whitney U Test , p = 0 . 408; n = 20 . Fig 7G and 7H ) .
Here we report a paradigm of effector gene birth for a plant pathogen of global economic importance . Cyst nematodes have exploited a series of gene gain events to redeploy glutathione synthetase-like enzymes during parasitism , within the syncytial feeding cell formed in the host root . We predict that the attributes acquired by GS-like paralogs that allow them to be deployed as effectors ( e . g . DOG box promoter motif , change in spatial expression , change in temporal expression , gain of signal peptide ) constitute a programme of effector evolution common to the genesis of other plant-parasitic nematode effectors from endogenous loci ( e . g . SPRY-SECs [39] ) . The programme likely also applies to well-documented gene gain events in plant-parasitic nematodes ( e . g . effectors derived from horizontal gene transfer events [33] ) , and perhaps even other pathosystems ( e . g . aphids [40] ) . Studying effector gene birth may therefore contribute towards addressing a priority in the field: characterising effector repertoires of diverse plant-pathogens [41] . The structures of the GS-like effectors led us to employ non-biased approaches to measure and analyse thiol biology during parasitism . Ultimately this resulted in the unexpected discovery of a range of novel thiols associated with the nematode feeding sites in host roots . Whilst cysteine , γ-EC and glutathione are major LMW thiols common to both plants and animals , analysis here of potato , and previously of other species [42] , reveals a diverse array of unidentified LMW thiols in plants . Many of the very large number of undescribed compounds ( ~200 ) discovered in the Arabidopsis sulfur metabolome [43] could be LMW thiols , representing a pool of potential novel cysteine-containing substrates for the nematode GS-like effectors . The thiol moiety , a nucleophile occurring predominantly in cysteine residues , is one of the most chemically reactive groups in biological systems and plays a major role in maintenance of cellular redox homeostasis . In most other plant-pathogen interactions described to date , the strong nucleophile glutathione is a negative regulator of pathogenicity [18–22 , 24 , 44 , 45] . For example , the Arabidopsis pad2-1 mutant that has reduced glutathione content is more susceptible to a range of pathogens including Pseudomonas syringae and Phytophthora brassicae [24] , whilst increased glutathione enhances plant defence responses [21] . The RipAY effector from the bacterial pathogen Ralstonia solanacearum has recently been shown to specifically target host glutathione in order to promote pathogenicity . It acts as a γ–glutamyl cyclotransferase to deplete intracellular glutathione , further emphasising the important role of this thiol in plant immunity [46–48] . A notable exception is the discovery that homoglutathione deficiency impairs root-knot nematode development in Medicago truncatula [49] . Here we show that Arabidopsis plants deficient in endogenous glutathione synthesis are less susceptible to cyst nematodes . This initially would thus seem unsurprising , however the necrosis and aborted feeding sites that result from depletion of glutathione during cyst nematode parasitism are not apparent for root-knot nematodes [49] , suggesting different roles for glutathione in the two interactions . It is also important to note that cyst and root-knot nematodes have independent evolutionary origins of sedentary endo-parasitism , have almost no overlap in effector complement [39] , and produce feeding sites that are different in structure and ontogeny [50] . Taken together , we cannot draw clear parallels between these superficially similar discoveries in two dissimilar pathosystems . Nevertheless , we show that plant-derived glutathione is a positive regulator of cyst nematode parasitism . Interestingly , the expansion of Clade 2 GS enzymes , which preceded that of the GS-like effectors , is common to cyst nematodes , root-knot nematodes , and indeed all endoparasitic nematode species within the Tylenchida ( including those that do not establish a feeding site within the host ) . Clade 2 GS enzymes do not encode a secretion signal and are expressed in the intestine . While the Clade 2 GS enzyme tested clearly has a slower rate of canonical enzyme activity than the Clade 1 GS , it can nevertheless synthesise glutathione: It has a high affinity for the γ-EC dipeptide substrate , and a preference for glycine as the terminal amino acid despite a lack of conservation in the two residues that apparently contribute to this specificity . While we cannot rule out the existence of other substrates , Clade 2 GS-like enzymes have retained canonical activity . Many Clade 2 GS share a short and somewhat variable C-terminal extension that is absent from all other clades and is not recognised as a canonical signal by TargetP . We can assume that the conservation of this C-terminal extension implies the existence of some functional constraints that remain to be elucidated . In conclusion , we implicate a positive role for novel nucleophiles in parasitism of cyst nematodes . We show three discoveries that are functionally independent but grouped under the banner of redox homeostasis in the plant cell , 1 ) Nematode-derived GS-like effectors likely accept a thiol substrate , but do not produce glutathione ( Figs 3 , 4 and 5 ) ; 2 ) Syncytia are abundant in novel thiols of unknown origin ( Fig 6 ) ; and 3 ) Plant-derived glutathione is a positive regulator of cyst nematode parasitism ( Fig 7 ) . In contrast to this , it was shown that rbohD-dependent ROS production is also integral to feeding site development , and is necessary to limit cell death and promote cyst nematode parasitism [37] . Taken together , these data collectively support the hypothesis that nematode-induced syncytia operate within a narrow redox “Goldilocks zone” . The focus of future research will be to determine if any of these discoveries are dependent on one another biochemically , what is the cross talk between the various aspects of redox regulation , and how , together , they contribute to parasitism .
RNAseq reads for Heterodera avenae [51] , H . schachtii ( doi:10 . 5061/dryad . 7vd0160 . ) and Rotylenchulus reniformis [52] were trimmed according to previously described methods except that HEADCROP was set to 11 , 10 and 12 respectively [26] . Trimmed reads were assembled into de novo transcriptomes using the Trinity pipeline [53] with a minimum Kmer coverage of 2 . Proteins were predicted using the Trinity wrapper scripts for transdecoder using the Pfam A and B library . GS genes were predicted from the assemblies generated above , existing transcriptome assemblies for Nacobbus aberrans [26] , Longidorus elongatus [54] , and Pratylenchus penetrans [55] and existing genome assemblies for Globodera rostochiensis [28] , Globodera pallida [12] , R . reniformis ( doi:10 . 5061/dryad . 7vd0160 . ) , Meloidogyne incognita [25 , 56] , Bursaphelenchus xylophilus [57] , and Caenorhabditis elegans [58] using Pfam ( PF03917/PF03199 ) . For R . reniformis , additional GS-like sequences were identified in the genome and transcriptome by sequence similarity searches ( BLAST v 2 . 4 . 0; [59] ) using all G . pallida GS-like proteins as queries . The results of these two identification pipelines were merged and a list of unique R . reniformis GS-like sequences was compiled from both the genome and transcriptome . Drastically truncated sequences identified following alignment of all encoded proteins ( MUSCLE v3 . 8 . 3 . 1; [60] ) , were removed or , for genomic predictions , manually curated where possible based on transcript coverage and/or homology . Any incomplete genomic predictions that could not be corrected due to missing sequence were also removed from further analysis . For G . pallida , upstream regions ( 2 kb ) of all genomic predictions were analysed for the presence of additional exons that could encode signal peptides . The majority of G . pallida full-length coding regions , including all those where manual curation conflicted with the original gene prediction , were amplified using Phusion polymerase from cDNA prepared from early parasitic stage nematodes . Primers used are detailed in S5 Table . The number of predicted G . pallida GS genes that encoded a signal peptide increased following the manual curation and subsequent cloning . The amino acid sequences of those genes from all species remaining after curation were aligned and refined using MUSCLE v3 . 8 . 3 . 1 [60] . The alignment was trimmed using TrimAL ( -gappyout ) [61] and subject to model selection ( WAG+GAMMA with invariable sites ) and Bayesian phylogeny construction ( Mr Bayes ) with two million five hundred thousand generations , a sample frequency of 10% , and a burn in rate of 30% carried out in TOPALi v2 . 5 [62] . The phylogeny was out-group routed by the Clade containing the C . elegans and L . elongatus sequences [63] using FigTree v1 . 4 ( http://tree . bio . ed . ac . uk/software/figtree/ ) . Where available ( G . rostochiensis , G . pallida , H . avenae , R . reniformis , H . schachtii , and N . aberrans ) , RNAseq reads were mapped back to either the relevant assembly , or in the case of G . rostochiensis and R . reniformis the manually curated GS-like transcripts , and normalized expression values ( TMM ) were calculated using the Trinity wrapper scripts for RSEM and EdgeR using default parameters . Expression fold change was calculated by dividing average normalized expression at parasitic life stages ( any stages recovered from roots ) by that of non-parasitic ( eggs , second stage juveniles and , for G . pallida , adult males ) . Multiple biological replicates were available for G . pallida ( two ) , G . rostochiensis ( two ) , R . reniformis ( three ) and N . aberrans ( three ) . In depth transcriptional analyses across the life cycle of G . pallida were performed using normalized expression values available [12] . Signal peptides were predicted using SignalP v4 . 0 [64] . For the cyst nematodes G . pallida and G . rostochiensis , in situ hybridisation was carried out on 3rd ( J3 ) and 4th ( J4 ) stage juveniles and young adult females extracted from roots of potato ( Solanum tuberosum ) according to previously described methods [28] . For sedentary stage female R . reniformis nematodes extracted from roots of cotton the same methods were followed except that the proteinase K treatment was reduced to 1 hour at room temperature . Single-stranded 100–200 base pair DNA probes corresponding to sequences of interest were prepared as described [24] using the oligonucleotide primers detailed in S5 Table . For each gene of interest , an equivalent sense-strand probe acted as a negative control . More than 100 individual nematodes were examined for each probe and the results presented are representative of the staining patterns observed . Purified , recombinant Gpa-GSS17 protein was used to raise a polyclonal antibody in rabbit that was affinity-purified against the original antigen by Cambridge Research Biochemicals ( Billingham , UK ) . For detection of the protein in nematodes , mixed parasitic stages of G . pallida were recovered from potato roots , fixed , cut , permeabilized and dehydrated as for in situ hybridisation . Rehydrated nematodes were washed with maleic acid buffer then incubated in the same buffer containing 1% blocking reagent ( Roche ) for 30 mins at room temperature . Following an overnight incubation at 4°C in blocking buffer containing Gpa-GSS17 antibody at a dilution of 1 in 200 , nematodes were washed , reblocked for 30 mins and incubated with FITC-conjugated goat anti-rabbit 2o antibody ( Sigma ) at a dilution of 1 in 200 for 2h at room temperature . After three washes in maleic acid buffer containing 0 . 01% Tween-20 , nematodes were resuspended in anti-fadent ( PBS/glycerol; Citifluor ) and visualized using a Leica DMRB microscope with GFP filter set . The experiment was carried out on separate occasions with two batches of fixed nematodes and >100 individual nematodes were observed on each occasion . Images were captured with a QIcam camera ( QImaging ) and Q-Capture software . The images presented are representative of all those that displayed hybridisation of the antibody . Control nematodes were processed in the same manner with the omission of primary antibody . For detection of Gpa-GSS17 in syncytia , lengths of potato root 14 days post infection with J2 of G . pallida were fixed in 4% paraformaldehyde in PEM buffer ( 50 mM PIPES , 10 mM EGTA , 10 mM MgSO4 pH 6 . 9 ) for 3 days at 4°C . Samples were dehydrated , resin embedded , sectioned and applied to microscope slides according to Davies et al . [65] Transverse sections through the nematode feeding site were blocked with 5% milk powder in PBS for 3 h , then incubated in primary Gpa-GSS17 antibody at a dilution of 1 in 50 in 0 . 5% milk powder/PBS overnight at 4°C . After washing in PBS , primary antibody was detected with a FITC-conjugated anti-rabbit secondary antibody at a dilution of 1 in 100 . At least 20 sections through each of three separate syncytia were analysed . Control sections were treated identically except for the omission of primary antibody . Plant cell walls were stained by incubation in Calcofluor-White ( 1 mg/ml ) for 5 mins , followed by copious washes with PBS . Antibody localisation was visualized and recorded as described above for nematodes . All GS-like genes analysed were cloned ( without their predicted signal peptide if appropriate ( SignalP v4 . 0 ) ) into the pOPINS3C vector [66] in frame with an N-terminal poly-Histidine tag , a SUMO chaperone to promote protein solubility , and a 3C protease cleavage site ( S5 Table ) . GS-like genes were expressed in , and their encoded proteins purified from , E . coli strain Shuffle to allow disulphide bond formation [67] . Cell cultures were grown in Luria Bertani media at 30°C until an optical density of 0 . 6–0 . 8 at A600 was reached . Cell cultures were cooled to 18°C and expression of the GS-like proteins of interest induced with addition of IPTG to a final concentration of 1 mM . Proteins were allowed to express for 14 hours at 18°C and the cells were collected by centrifugation and lysed immediately . Cell pellets were re-suspended in 50 mM Tris-HCl , 500 mM NaCl , 50 mM glycine , 5% ( v/v ) glycerol and 20 mM imidazole , pH 8 . 0 with the addition of one EDTA-free protease inhibitor tablet per 50 ml , and lysed by sonication . Cell lysate was clarified by centrifugation and applied to a 5 ml Ni2+-NTA column on an AKTA Xpress . His-tagged proteins were step eluted in resuspension buffer + 500 mM imidazole and injected onto a Superdex 200 26/60 gel filtration column equilibrated to 20 mM HEPES and 0 . 15 M NaCl , pH 7 . 5 . Fractions containing the protein of interest were pooled , concentrated to ~5 ml , and the His+SUMO tag cleaved by overnight digestion with 3C protease at 4°C at a ratio of 100:1 ( protein:protease ) . Mature GS-like proteins were separated from the His+SUMO tag by passing the solution over a 5 ml Ni2+-NTA column and injecting the flow-through onto a Superdex 200 26/60 gel filtration column . The concentration of each protein was measured by direct detection of the peptide bond ( Direct detect ) , and protein aliquots were stored at -80°C until needed . All enzyme assays were carried out at 30°C in a typical reaction buffer of 100 mM HEPES ( pH 7 . 5 ) , 20 mM MgCl2 , and 5 mM dithiothreitol , with the addition of ATP , γ-EC , and glycine at varying concentrations . The hydrolysis of ATP in the presence of each protein , relative to the control , was used as a molar equivalent proxy for the production of glutathione , and measured by detection of free phosphate using malachite green absorbance at 630 nm . After determining the linear range of the reaction over time , pkat values of each protein in the presence of 1 mM γ-EC , 2 . 5 mM ATP and 100 mM glycine were measured in triplicate , and compared to a standard curve of free phosphate . To estimate the Michaelis-Menten kinetics of those enzymes with a rate appreciably above their negative control , γ-EC was varied in serial dilution and data analysis carried out in Sigmaplot . Experiments to explore the substrate specificity at the terminal amino acid of all GS-like enzymes were carried out using 2 . 5 mM ATP and 1 mM γ-EC with the following amino acids at 100 mM: glycine , β-alanine , D-alanine , GABA , AABA , diaminopropionic acid , D-serine ) . Purified GS-like proteins were concentrated to between 5 and 10 mg/ml in 20 mM Tris ( pH 7 , 5 ) and 200 mM NaCl . Sitting drop vapour diffusion crystallization experiments at 20°C were carried out in 96 well format using an OryxNano robot . Gpa-GS22 crystallized readily in a number of conditions in several screens at 5 mg/ml final concentration . Screen JCSG condition D10 , was optimized to produce crystals in 0 . 2 M tri-methylamine N-oxide , 0 . 1 M Tris pH 9 and 20% w/v PEG 2000 MME . Several crystals were transferred to cryoprotectant ( mother liquor with the addition 20% ethylene glycol final concentration ) and frozen in a loop in liquid nitrogen . X-ray diffraction data were collected at the Diamond light source beamlines i04-1 , processed using the xia2 pipeline [68] , and the structure was solved by molecular replacement with 3KAL and named Gpa-GSS22-apo . The submitted structure was obtained through an iterative process of manual building and refinement using COOT [69] and REFMAC5 respectively . Tools of COOT and MOLPROBITY [70] were used for structure validation . Further , Gpa-GSS22 , at a final concentration of 2 . 5 mg/ml , crystallized in the same condition with the addition of ADP ( 2 . 5 mM ) , MgCl2 ( 5 mM ) and glutathione ( 2 . 5 mM ) . Diffraction data ( beamline i02 ) , structure solution , refinement , and validation were carried out as for Gpa-GSS22-apo , using the solved structure of Gpa-GSS22-apo in molecular replacement . Protein crystals for Gpa-GSS30 were obtained in JCSG E1 ( 0 . 2 M magnesium formate di-hydrate and 20% PEG 3350 ) and the structure solved/refined/validated as described above , using Gpa-GS22-apo for molecular replacement . Protein crystals for St-GSS1 were obtained by optimisation of JCSG D9 ( 1 . 4 M D-malic acid ) with the addition of 2 . 5 mM γ-EC , 2 . 5 mM ADP , 5 mM MgCl2 and St-GSS1 at 3 . 75 mg/ml final concentration . The structure was solved/refined/validated as described above , using molecular replacement with the structure of hGS of Glycine max ( 3KAL ) . Intra-cellular free thiols were visualized in syncytia produced by H . schachtii 7 , 14 and 21 days post infection by incubating lengths of Arabidopsis root in 5 mM ThiolTracker Violet ( Life Technologies ) for 2 hours at room temperature . This fluorescent dye reacts with any reduced thiols , including glutathione , in live cells . Samples were rinsed twice in thiol-free PBS ( Life Technologies ) prior to imaging with a Zeiss LSM700 confocal microscope ( excitation at 405 nm and collection from 410–500 nm ) . For thiol analysis , roots of potato plants cv Desiree were harvested 28 days after planting tubers into sandy loam soil infested with cysts of G . pallida . Plants were grown at 20°C in a glasshouse with a 16h:8h light:dark cycle . Roots were washed to remove adhering soil particles and maintained in water while infection sites were identified using a stereobinocular microscope . Young adult female nematodes at approximately 21 days post infection were carefully removed intact from the root and collected . The length of root harbouring the syncytium ( 3–4 mm ) was excised and collected separately . Equivalent , uninfected root lengths were harvested from the same root system . A total of ~200 feeding sites/nematodes/control root sections were amassed on each experimental occasion . Tissue in 1 . 5 ml tubes was flash frozen in liquid N2 and stored at -70°C until use . Samples were thawed in 130 ul of 0 . 1 M HCL and homogenized with a micro pestle . The cell lysate was clarified by centrifugation twice ( 10 minutes , 4°C , 20 , 000 RPM ) and 50 μl was removed for derivatisation of thiols by addition of 1 . 5 μl 1 M DTT , 1 . 5 μl Mono-bromobimane and 45 μl 1 M CHES pH 9 . The reaction was incubated at room temperature for 20 minutes , and stopped by the addition of 50 μl of 50% acetic acid . Samples were analysed using a Shimadzu LC/MS system comprising Nexera UHPLC binary pumps and autosampler , Prominence fluorescence and UV diode array detectors and LCMS-2020 single quadrupole mass spectrometer with ESI/APCI dual ion source . Two microliters of each sample was injected onto a Accucore 150-Amide–HILIC 100mm x 2 . 1mm column held at 25°C , and low molecular weight thiols were separated by a gradient of 0 . 1% v/v formic acid ( Buffer A ) and acetonitrile ( Buffer B ) : a linear gradient from 90–85% B over 6 minutes followed by 85% B for 2 minutes , 60% B for 1 . 5 minutes , and 90% B for 2 . 5 minutes . Samples were eluted at 0 . 4 ml/minute and mono-bromobimane derivatives were detected by fluorescence at excitation/emission 397/480 nm . Mass spectra were collected continuously using a DL temperature of 250°C , a nebulizing gas flow rate of 1 . 5 L/min , a heat block temperature of 400°C , a drying gas flow rate of 15 L/min and in both ESI and APCI ionisation mode . Mass spectra were collected in scan mode with both +ve ( 4 kV ) and–ve ( -3 . 5 kV ) across a range of 100–1200 Da . To determine the elution time of glutathione under these conditions , samples were spiked with 1 μl of 10 mM glutathione standard ( Sigma ) after clarification of the HCL extract by centrifugation . Surface sterilised Arabidopsis thaliana ( Col-0 ) or pad2-1 seeds were grown for 15 days on 9 cm vertical plates containing ½ strength Murashige and Skoog medium supplemented with 10 g/l sucrose and 1% Phytagel ( ½ MS10 ) . Hatched second stage juveniles ( J2 ) of H . schachtii were sterilised [71] and resuspended at a concentration of approximately one nematode/μl . 20 μl of suspension was pipetted onto each of two root points per plant with two plants per plate . Infection points were covered with GF/A paper ( Whatman ) for two days to facilitate invasion . At 10–12 days post infection , nematode ( excluding un-emerged adult males ) and syncytium size were estimated from the projected cross-section as viewed under a microscope ( Olympus BH2 ) and measured using Image-Pro Analyser v7 ( MediaCybernetics ) . Lengths of root containing aborted syncytia were incubated in propidium iodide ( 10 μg/ml ) for 5 minutes at room temperature , washed twice in PBS , and imaged with a Zeiss LSM700 ( excitation at 488 nm and collection from 590–700 nm ) . For L-Buthionine-Sulfoximine ( BSO ) treatment , wild-type seedlings were transferred to ½ MS10 plates containing 1 mM BSO two days prior to infection . BSO toxicity to nematodes over the course of the invasion period was tested separately by incubating J2s on water agar plates with or without 1 mM BSO in the absence of any plants . Nematode mortality was scored by observing each nematode for 5–10 seconds to record movement . Nematode motility was assessed by measuring nematode speed over 2 minutes using the ImageJ ( http://imagej . nih . gov/ij/ ) plugin wrMTrck ( http://www . phage . dk/plugins/wrmtrck . html ) . | Plants and their pathogens/parasites are locked in an evolutionary arms race , with considerable attention directed towards the specific functions of the parasites’ “weapons”: the effectors . While we are beginning to understand these functions , we have very little understanding of how plant parasitic nematodes have bolstered their effector repertoire . Here we provide an example of how plant parasites of global economic importance have populated their effector repertoire by the unprecedented duplication and subsequent re-deployment of the endogenous housekeeping gene , glutathione synthetase . We hypothesise that many aspects of the “weaponization” programme detailed here will be common to the genesis of other plant-parasitic nematode effectors . Given that parasitic nematodes deploy a battery of effectors , many arising from the adaptation of either endogenous genes or loci acquired by horizontal gene transfer , this paradigm will have substantial impact on the effort to understand and ultimately undermine devastating USDA and EPPO quarantine organisms . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] | [
"invertebrates",
"medicine",
"and",
"health",
"sciences",
"ecology",
"and",
"environmental",
"sciences",
"chemical",
"compounds",
"parasite",
"evolution",
"parasitic",
"diseases",
"organic",
"compounds",
"animals",
"parasitology",
"nematode",
"infections",
"plant",
"scien... | 2018 | Effector gene birth in plant parasitic nematodes: Neofunctionalization of a housekeeping glutathione synthetase gene |
Rift Valley fever ( RVF ) is a zoonotic arbovirosis for which the primary hosts are domestic livestock ( cattle , sheep and goats ) . RVF was first described in South Africa in 1950–1951 . Mechanisms for short and long distance transmission have been hypothesised , but there is little supporting evidence . Here we describe RVF occurrence and spatial distribution in South Africa in 2008–11 , and investigate the presence of a contagious process in order to generate hypotheses on the different mechanisms of transmission . A total of 658 cases were extracted from World Animal Health Information Database . Descriptive statistics , epidemic curves and maps were produced . The space-time K-function was used to test for evidence of space-time interaction . Five RVF outbreak waves ( one in 2008 , two in 2009 , one in 2010 and one in 2011 ) of varying duration , location and size were reported . About 70% of cases ( n = 471 ) occurred in 2010 , when the epidemic was almost country-wide . No strong evidence of space-time interaction was found for 2008 or the second wave in 2009 . In the first wave of 2009 , a significant space-time interaction was detected for up to one month and over 40 km . In 2010 and 2011 a significant intense , short and localised space-time interaction ( up to 3 days and 15 km ) was detected , followed by one of lower intensity ( up to 2 weeks and 35 to 90 km ) . The description of the spatiotemporal patterns of RVF in South Africa between 2008 and 2011 supports the hypothesis that during an epidemic , disease spread may be supported by factors other than active vector dispersal . Limitations of under-reporting and space-time K-function properties are discussed . Further spatial analyses and data are required to explain factors and mechanisms driving RVF spread .
Rift Valley fever ( RVF ) is a vector-borne zoonotic disease caused by infection with a Phlebovirus ( Family Bunyaviridae ) . The main vectors are mosquitoes from the genera Aedes and Culex; primary hosts are domestic livestock ( cattle , sheep and goats ) , but the disease can also affect camels , buffaloes and other wild animals [1] . Since its first description in Kenya in 1931 [2] , RVF has been reported in several African countries , and in the Arabian Peninsula [3] . Transmission to humans is mainly through contact with infectious animals or animal tissues , and symptoms vary from a flu-like illness to more severe conditions such as meningoencephalitis , haemorrhagic fever or death . In animals , RVF is of economic importance , causing waves of abortions at all stages of pregnancy and high mortality in newborn animals [1] , [4] . Rift Valley fever epidemics have been reported following inundation of floodplains and dambos due to unusually heavy rainfall , allowing a large number of infected Aedes eggs to hatch , like in Kenya [5] or following the introduction of infected vectors or animals in flooded areas as hypothesized in Saudi Arabia and Yemen [6] . Animals are infected via bites from infectious vectors , and the sustainability of local transmission is supported by the presence of more permanent bodies of water in the environment which creates suitable conditions for Culex mosquitoes to breed and act as secondary vectors [7]–[11] . The extent of virus spread in time and space during RVF epidemics is believed to be attributed to active or passive vector dispersal , but also to the movements of infectious animals , either wild or domestic [12] . Although practically challenging to study because of data scarcity , knowledge on the relative importance of vector dispersal versus movements of infectious animals would be useful to inform disease control . For infectious diseases , the presence of space-time interaction between cases , which is the extent to which cases are spatially and temporally proximate , can be interpreted as an indicator of an underlying contagious process [13]–[17]; and measuring and quantifying it may assist in generating hypotheses on the different mechanisms of transmission involved in disease spread . The analysis of space-time interactions using the space-time K-function , has previously been explored for a variety of animal infectious diseases , such as sheep scab [18] , foot-and-mouth disease [19] , [20] and equine grass sickness in Great Britain [21]; tuberculosis in cattle farms in New Zealand [22] , infectious bursal disease in broilers in Denmark [23] , and recently foot-and-mouth disease in Tanzania [24] and porcine high fever disease in Viet Nam [25] . In South Africa , three major country-wide epidemics occurred in 1950–1951 [26] , in 1973–1975 [27] and lately in 2008–2011 . As of April 2012 , very few descriptions of these epidemics have been published [26]–[29] . This paper presents a first step to improve our understanding of the space-time pattern of RVF in South Africa using the 2008–2011 dataset collated from World Animal Health Information Database [30]–[34] . During these four years , a total of 690 farms were confirmed RVF positive . About 95% ( n = 658 ) of the farms contained the most susceptible species to RVF infection , that is , domestic livestock including cattle , small ruminants ( sheep or goats ) or both; the remaining farms raising Camelidae or wild animals . In the present paper , we used the RVF domestic livestock data subset to describe the spatial and temporal pattern of RVF in 2008–2011 , and , by using the space-time K-function , to quantify the presence of a potential transmission process , in order to generate hypotheses on the different mechanisms of RVF transmission .
The dataset contained 658 RVF cases , defined as reports from farms raising only cattle , small ruminants ( sheep or goats ) , or both , in South Africa , between 2008 and 2011 , collated from the World Animal Health Information Database [30]–[34] . Available information comprised the GPS coordinates of the affected farms , the starting date of the outbreak ( day precision ) , the host species , and where available the number of susceptible animals , cases and animal deaths on the farm . Since RVF is an “OIE ( World Organisation for Animal Health ) Listed Disease” , diagnosis was made using standardised RVF diagnostic tests [35] . Epidemic curves showing the daily number of cases for the years 2008 , 2009 , 2010 and 2011 were produced , and cases were mapped . Descriptive on-farm statistics were calculated , including on-farm morbidity and case fatality proportions . On-farm morbidity was obtained by dividing the number of cases by the number of susceptible animals present on farm; and case fatality was the number of deaths divided by the number of cases . Space-time interaction was investigated using the space-time K-function , K ( s , t ) , defined as the expected number of cases that occur within separating distance s and time t of a previously randomly selected case , divided by the mean number of cases per unit space per unit time , also termed “intensity” [14] . In the absence of space-time interaction , that is , when cases occur independently in time and space , K ( s , t ) is the product of two K-functions in space K1 ( s ) and in time K2 ( t ) ; such as: K ( s , t ) = K1 ( s ) K2 ( t ) ( Eq 1 ) . If we define D ( s , t ) the difference D ( s , t ) = K ( s , t ) −K1 ( s ) K2 ( t ) ( Eq 2 ) , then positive values of D ( s , t ) indicate the presence of space-time interaction; and the higher D ( s , t ) , the stronger the evidence . Because D ( s , t ) naturally increases with space and time , we calculate D0 ( s , t ) = D ( s , t ) /K1 ( s ) K2 ( t ) ( Eq 3 ) , which is the number of events attributable to space-time interaction divided by the number of events in the absence of a space-time interaction . D0 ( s , t ) is therefore interpreted as the proportional increase , or excess risk , attributable to the space-time interaction [14] . D0 ( s , t ) >1 indicates that the number of observed events was greater than twice the number of expected events [23] . Under the null hypothesis of no space-time interaction , the dates of case reports are randomly permuted on the fixed set of case locations using Monte Carlo simulation , therefore generating a distribution for D ( s , t ) values . This distribution is compared with the D ( s , t ) calculated from the observed data , and if it exceeds 95 percent of the simulated D ( s , t ) values , then it can be concluded that there is less than 5% probability that the observed space-time interaction occurred by chance [17] , [36] . The space-time K-function was calculated separately for the years 2008 , 2009 ( for each distinct wave ) , 2010 and 2011 . Maximum separation distances of 300 km and 60 days were used for s and t dimensions to investigate long-distance transmission mechanisms , and to allow farms' infectiousness to persist twice as long as the 30 days assumed at the animal level by the OIE [35] . D ( s , t ) was estimated from 999 Monte Carlo random date permutations . The analysis was implemented using the splancs library [37] from the statistical package R version 2 . 14 . 0 [38] .
Between 2008 and 2011 , 658 RVF cases were reported in five distinct waves of varying size and location . Over 70% ( n = 471 ) of the cases were reported in 2010 ( Table 1 ) . The occurrence of RVF was seasonal , with most cases occurring between January and April , and reported until July ( Figure 1 ) ; except in 2009 when RVF cases resumed in October . Figure 2 shows the spatial distribution of RVF cases reported during the period 2008–2011 . In 2008 , Mpumalanga , North West , Gauteng and Limpopo provinces were affected ( Figure 2A ) . In 2009 , cases from the first wave were located in the east of the country , mostly in KwaZulu-Natal province; and the second wave occurred in the Northern Cape , near the Namibian border ( Figure 2B ) . In 2010 , the epidemic was almost country-wide , except for the eastern low-lying areas ( Figure 2C ) . Finally , in 2011 , cases were mostly distributed in southern South Africa , mainly in the Western Cape and Eastern Cape provinces ( Figure 2D ) . Across the four years , the mean on-farm morbidity varied from 0 . 02 to 0 . 23 in 2008–2009 , 0 . 07 to 0 . 09 in 2010 , and 0 . 07 to 0 . 21 in 2011 . The mean on-farm case fatality ranged from 0 . 29 to 1 . 00 in 2008–2009 , 0 . 66 to 0 . 79 in 2010 , and 0 . 85 to 1 . 00 in 2011 ( Table 2 ) . Finally , for the four years , the mean morbidity and case fatality proportions for cattle farms were 0 . 08 and 0 . 74 respectively; 0 . 10 and 0 . 81 for small ruminant farms , and finally 0 . 07 and 0 . 67 for farms raising both ( Table 2 ) . Table 3 presents the spatiotemporal distances at which an excess risk ( Do ( s , t ) >1 ) was attributable to space-time interaction , together with their p-values . No space-time interaction was present during the second 2009 wave and only weak evidence was found in 2008 ( p-value = 0 . 091 , Table 3 ) . Do ( s , t ) plots were produced for the waves that showed significant space-time interaction ( p-value<0 . 05 ) , that is , the first 2009 wave and the 2010 and 2011 ones ( Figures 3A , 3B and 3C ) . Detailed examination of the Do ( s , t ) values for 2009 showed evidence of a short ( 1 day ) and intense contagious process ( excess risk >3 ) up to 20 km . The intensity of the space-time interaction decreased but remained for a month , up to 40 km ( Table 3 and Figure 3A ) . Initial and localised transmission processes were observed in the 2010 and 2011 waves ( 3 days over 5 km and 3 days over 15 km , respectively ) , although the intensity of the transmission seemed to be more important in 2011 ( maximum excess risk = 5 . 88 ) compared with 2010 ( maximum excess risk = 3 . 20 ) . However , although reduced ( 1<Do ( s , t ) <2 ) , the spatial extent of the transmission was almost 3 times more important in 2010 ( 90 km ) than in 2011 ( 35 km ) within the same time-window of 13 days ( Table 3 , Figures 3B and 3C ) .
Rift Valley fever has been reported in South Africa over the last four years , showing a seasonal pattern mainly between January and July . About 70% of the cases reported between 2008 and 2011 occurred in 2010 . Each year , a different part of the country has been affected , with the 2010 epidemic being almost country-wide . In other years , cases were confined to a few provinces . No strong evidence of space-time interaction was found in 2008 and in the second wave in 2009 . In the first wave of 2009 , a significant space-time interaction was detected for up to one month and over 40 km . In 2010 and 2011 a significant intense , short and localised space-time interaction ( up to 3 days and 15 km ) was detected , followed by one of lower intensity ( up to 2 weeks and 35 to 90 km ) . The season between January and April ( mid-summer to autumn ) , brings rain in most parts of the country , and corresponds to the period when Culex theileri , Aedes juppi , Aedes mcintoshi and other members of the Aedes ( Neomelaniconion ) genus , the main RVF epidemic vectors in South Africa , are the most prevalent mosquitoes [39] . Our results , showing significant contagious processes during these seasons for the years 2009 , 2010 and 2011 , are in line with the hypothesis that mosquito bites are the principal infection mechanism of RVF in South Africa . While these results are to be expected for a vector-borne disease , the absence of contagious process in 2008 and the second 2009 wave , and the various extents and intensities of the space-time interactions found across the different years could support further evidence that other transmission mechanisms may also exist . Active dispersal for most RVF vectors is short , and although little information is available , it is estimated to be about 1 km , varying from less than 150 m for Aedes to approximately 2 km for Culex theileri [12] , [40] . In addition , the analysis of spatial and space-time clusters for dengue , a human disease mainly transmitted by Aedes aegypti , showed a local transmission varying between 800 m and 4 km [41]–[43] , and spatio-temporal clusters over short distances from 400 m to 2 . 8 km , sustained over 2 to 13 weeks [41] , [44] , [45] . These vector-borne transmission patterns share some similarities with the initial and localised contagious processes observed during RVF epidemics in 2010 and 2011 , but our study detected the presence of an additional spatiotemporal process , with RVF potentially spreading to distances up to 40 to 90 km , within about 2 weeks . This appearance of long-distance spread could be explained by the existence of several RVF virus emergences; defined as distinct hatchings of infected Aedes eggs or multiple re-introductions of infected vectors from areas external to our study area . However , similar extended spatio-temporal patterns as those observed in this study have been described for foot-and-mouth disease in Tanzania , reaching 50 km to hundreds of kilometres over a 2 week period [24] and for avian influenza in Bangladesh up to 150–300 km [46]; both diseases for which the movements of animals were likely to play a major mechanism of spread [47]–[49] , [50] . Therefore , this suggests that RVF spread over distances larger than the assumed range of active vector dispersal could be explained by the movement of domestic or wild viraemic and therefore infectious animals . Other mechanisms of long-distance spread could also be incriminated , such as wind-borne vector dispersal , which has been described up to 100 km for some Aedes and Culex species [40] . Finally , in early 2009 in KwaZulu-Natal province , space-time interaction was present up to 20 km within 1 day . Such a pattern probably allowed ruling out active vector dispersal in favour of animal movements , or multiple local emergences . Several limitations in these analyses may have affected our results and their interpretation . Firstly , this study relies on RVF cases that were reported to the World Organisation for Animal Health ( OIE ) and are likely to represent only a subset of the total number of infected farms in South Africa . From a statistical perspective , the type I error of the space-time K-function has been shown to remain low with under-reporting of cases [51] , which means we can be confident that the space-time interactions found in 2009 , 2010 and 2011 actually existed . Also , Fenton et al . 2004 [51] showed that the K-estimate was a good reflection of the underlying contagious process , when the probability of a farm not being reported increased proportionally with increasing distances from a random point , assumed to be a regional laboratory centre , which is likely to be the case for a notifiable disease . However , the study power , i . e . the ability of the test to detect a space-time interaction when there is one , was more dependent on sample size [51] , which makes it difficult to know whether the absence of space-time interaction in 2008 was likely to be true or resulted from the small number of reported positive farms . While no published outbreak investigation has been identified for this 2008 outbreak , Anyamba et al 2010 [52] reported that the current early warning system , based on climatic factors , forecasted suitable conditions for virus re-emergence on a regional scale ( Southern Africa ) in February 2008 . However , no larger epidemic followed , suggesting an absence of suitable environmental conditions for producing significant populations of secondary vectors to amplify the virus to epidemic proportions . The absence of contagiousness for the second wave of 2009 is easily explained by the fact that 89% ( 17/19 ) of the cases were reported on the same day ( October 19 , 2009 ) . If cases truly occurred in different locations on the same day , this would suggest that the virus was evenly distributed in the environment and emerged at the same time . For this wave , one outbreak investigation was published [53]; reporting no abnormal climatic conditions that could explain high mosquito densities , but hypothesized flood irrigation techniques as a factor for virus emergence , and a low number of animals precluding to sustain an epidemic . In addition , recent genome sequencing revealed that RVF viruses from the same lineage H caused the outbreaks in Namibia in 2004 , these late 2009 cases , and the 2010/2011 ones in South Africa , suggesting an epidemiological link between them [54] . Secondly , the definition of the space-time K-function is based on several assumptions that may have affected our results . For example , the space-time K-function assumes that the underlying first-order effects are constant across the space-time study environment [14] , [15] , [17] , therefore considering that all cases arose from second-order effects . In our study , this means that cases within 300 km and 60 days of any arbitrary case were treated as if resulting from transmission only and none were due to emergence . Since the existence of multiple foci of RVF virus emergence cannot be totally excluded , by artificially decreasing the number of potential ‘parent cases’ in the dataset , that is RVF foci , we tended to overestimate the study power [51] . Further environmental data would be necessary to identify potential RVF foci resulting from Aedes hatching , although infected farms located in such suitable environment could also have been infected by transmission from neighbouring infected farms . Another assumption of the space-time K-function is that the density of the population at risk does not vary , or varies evenly over time [13]–[15] , [17] . In practice , the population at risk is likely to have reduced over time due to animal vaccination or life-long immunity induced by natural infections [4] , and to culling procedures that removed previously diagnosed animals . The timing and location in which these activities ( i . e . vaccination and culling ) were implemented are both difficult to estimate since they depended on farmers' decisions . However , a decrease in the number of susceptible farms over time would have resulted in under-estimating the intensity of the space-time interaction , which makes our results conservative . Thirdly , it is acknowledged that vaccination could have been applied in some affected farms or areas during the different waves [30]–[33] , but since RVF is not an officially controlled disease , vaccination coverage is not reported by the government [55] . Nevertheless , vaccination was widely advised during the 2010 epidemic [34] , and it is therefore possible that some part of the areas affected in 2011 were vaccinated prior to the 2011 wave itself; leading to a possible underestimation of the D0 ( s , t ) values . Finally , the analysis was conducted using animal and not human cases . Whereas humans acquire infection by close contact with infected animals or their infected organs , domestic livestock are the primary hosts for RVF virus , and get infected directly from mosquito bites . Therefore the dynamics of disease in those species should better reflect vector transmission . In conclusion , by providing a description of the spatiotemporal patterns of RVF in South Africa between 2008 and 2011 , this study supports the hypothesis that during an epidemic , disease spread may be supported by factors other than active vector dispersal . To optimize disease control , these mechanisms underlying disease spread should be disentangled and quantified . This would require the use of spatiotemporal modelling tools in combination with environmental , virus genotyping , vaccination , animal movement and population at risk data . | The factors explaining Rift Valley fever ( RVF ) spread in domestic livestock during an epidemic are attributed to short and long distance mechanisms , including active vector dispersal , passive vector dispersal and movements of infectious animals . However , because of data scarcity , quantifying and disentangling these mechanisms remains challenging . Here , we generate hypotheses on the possible mechanisms involved in RVF spread in South Africa between 2008 and 2011 . We use descriptive statistics and estimate the space-time K-function to explore the presence of space-time interactions , being interpreted as an indicator of an underlying transmission process . Our results confirm the presence of an intense , short , initial transmission process that could be attributed to active vector dispersal; but also highlight the presence of another transmission mechanism of a lower intensity and over further distances that could be explained by the movements of infectious animals , passive vector dispersal or emergence of other foci . Further data collection and modelling tools are required to confirm these hypotheses . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"veterinary",
"epidemiology",
"veterinary",
"science"
] | 2012 | Exploratory Space-Time Analyses of Rift Valley Fever in South Africa in 2008–2011 |
Infection with Burkholderia pseudomallei or B . thailandensis triggers activation of the NLRP3 and NLRC4 inflammasomes leading to release of IL-1β and IL-18 and death of infected macrophages by pyroptosis , respectively . The non-canonical inflammasome composed of caspase-11 is also activated by these bacteria and provides protection through induction of pyroptosis . The recent generation of bona fide caspase-1-deficient mice allowed us to reexamine in a mouse model of pneumonic melioidosis the role of caspase-1 independently of caspase-11 ( that was also absent in previously generated Casp1-/- mice ) . Mice lacking either caspase-1 or caspase-11 were significantly more susceptible than wild type mice to intranasal infection with B . thailandensis . Absence of caspase-1 completely abolished production of IL-1β and IL-18 as well as pyroptosis of infected macrophages . In contrast , in mice lacking caspase-11 IL-1β and IL-18 were produced at normal level and macrophages pyroptosis was only marginally affected . Adoptive transfer of bone marrow indicated that caspase-11 exerted its protective action both in myeloid cells and in radio-resistant cell types . B . thailandensis was shown to readily infect mouse lung epithelial cells triggering pyroptosis in a caspase-11-dependent way in vitro and in vivo . Importantly , we show that lung epithelial cells do not express inflammasomes components or caspase-1 suggesting that this cell type relies exclusively on caspase-11 for undergoing cell death in response to bacterial infection . Finally , we show that IL-18’s protective action in melioidosis was completely dependent on its ability to induce IFNγ production . In turn , protection conferred by IFNγ against melioidosis was dependent on generation of ROS through the NADPH oxidase but independent of induction of caspase-11 . Altogether , our results identify two non-redundant protective roles for caspase-1 and caspase-11 in melioidosis: Caspase-1 primarily controls pyroptosis of infected macrophages and production of IL-18 . In contrast , caspase-11 mediates pyroptosis of infected lung epithelial cells .
Burkholderia pseudomallei is a Gram-negative flagellated bacterium that causes melioidosis , a diseases endemic to South-East Asia and other tropical regions and the most common cause of pneumonia-derived sepsis in Thailand [1 , 2] . Due to global warming and increased international travel , cases of melioidosis are increasingly being reported outside the endemic areas . B . pseudomallei infection can be contracted through ingestion , inhalation , or subcutaneous inoculation and leads to broad-spectrum disease forms including pneumonia , septicemia , and organ abscesses . Although not pathogenic to humans , Burkholderia thailandensis possesses several of the B . pseudomallei virulence factors , causes morbidity and mortality in mice , and is often used as a model for melioidosis [3–5] . Following infection of macrophages and other non-phagocytic cell types , Burkholderia is able to escape the phagosome and invade and replicate in the host cell cytoplasm . Macrophages and IFNγ have been shown to play a critical role in protection from melioidosis [6–8]and several B . pseudomallei virulence factors have been identified . Analysis of mouse strains with different susceptibility to B . pseudomallei infection indicates that the early phases of the infection are crucial for survival , emphasizing the necessity for better understanding of innate immune responses during melioidosis . Burkholderia has been shown to activate TLR2 , TLR4 , and TLR5 in epithelial reporter cell line [9] . Interestingly , while Myd88-/- mice are highly susceptible to B . pseudomallei infection [10] , Tlr4-/- mice have similar resistance to wild type ( WT ) mice but Tlr2-/- mice showed reduced mortality [11] indicating that MyD88-dependent pathways may play opposite role in melioidosis . This notion is supported by our previous works that showed that IL-18 was protective in melioidosis while IL-1β was deleterious because of excessive neutrophils recruitment to the lung and tissue damage due to release of neutrophil elastase [12 , 13] . Caspase-1 has been shown to be protective against Burkholderia infections [14] . Production of IL-1β and IL-18 in melioidosis is regulated by activation of caspase-1 downstream of the NLRP3 inflammasome while activation of the NLRC4 inflammasome triggers the pyroptotic cell death process [12 , 15] . A potentially confounding factor that affects all the works that examined the role of caspase-1 in melioidosis is that those studies relied on caspase-1-deficient mice that also lacked caspase-11 . The non-canonical inflammasome composed of caspase-11 ( encoded by Casp4 ) has also been shown to play a protective role in melioidosis [16] by recognizing cytoplasmically-located LPS [17 , 18] . This process is dependent on priming of macrophages with interferons or TLR ligands . The mechanism through which caspase-1 and caspase-11 initiate pyroptosis is by cleaving gasdermin D , a cellular protein that open pores in the cell membrane [19 , 20] . As for infections by most intracellular bacteria , IFNγ is an essential component of the innate immune response to B . pseudomallei and its absence results in severely decrease resistance to the infection [6–8] . The antimicrobial properties of IFNγ are mediated by several effector mechanisms operating in a variety of cell types . Among the thousand IFN-stimulated genes , IFN-induced GTPases , iNOS , and NADPH oxidase are the most studied and effective antimicrobial effector mechanisms of macrophages . Recently it has been proposed that in the early phase of Burkholderia infection caspase-11 may act as an IFNγ-inducible effector mechanism because of its reliance on the IL-18-IFNγ axis for priming , which would place non canonical inflammasome downstream of canonical caspase-1 activation [15] . The function of IFN-inducible effector mechanisms such as iNOS , ROS , Guanylate binding proteins , and caspase-11 in melioidosis has been examined previously [15 , 16 , 21 , 22] . While iNOS and several GBPs do not seem to be required to survive a lethal infection with B . pseudomallei or B . thailandensis , absence of NADPH oxidase or caspase-11 renders mice significantly more susceptible . For this reason we decided to determine the relative contribution of caspase-11 and NADPH oxidase to the protection conferred by IFNγ against B . thailandensis infection . We also revisited the role of caspase-1 independently of caspase-11 using recently generated bona fide caspase-1-deficient mice [23] . The results presented here support the following conclusions: first , the main protective role of caspase-1 in melioidosis is to trigger pyroptosis in macrophages and production of IL-18 . Second , caspase-11’s function during B . thailandensis infection is to mediate pyroptosis in lung epithelial cells , rather than in macrophages . Finally , the protective action of IFNγ is mediated by ROS independently of caspase-11 .
We and others have previously shown that caspase-1 plays a critical role during intranasal or intraperitoneal infection with Burkholderia [12–14] . However , those results were obtained using caspase-1-deficient mice that also lacked caspase-11 due to a passenger mutation in the 129 mouse strain . “Pure” caspase-1-deficient mice have been recently generated [23] and for this reason we decided to reexamine the role of caspase-1 and caspase-11 in melioidosis . Analysis of the survival of mice infected intranasally with B . thailandensis showed that both Casp1-/- and Casp11-/- mice were significantly less resistant than WT mice and became moribund within 5 days of infection and had to be euthanized ( Fig 1A ) . The bacterial burdens in different organs were measured at different time points post-infection ( p . i . ) and in mice infected using different doses and revealed the relative susceptibility of Casp1-/- , Casp11-/- , and Casp1-/-/Casp11-/- mice ( Fig 1B ) . Mice lacking both caspases had the highest amount of bacteria while mice deficient in caspase-11 were clearly more susceptible than Casp1-/- mice . This pattern has been previously observed in a study that compared Casp11-/- mice to mice that lacked both ASC and NLRC4 ( used as surrogate for pure caspase-1 deficient mice ) [15] . In agreement with the function of caspase-1 in the generation of the mature form of IL-18 , this cytokine was absent in BALF or serum of Casp1-/- mice while was detected at the same level in Casp11-/- or WT mice ( Fig 1C ) . The levels of IL-1β , a deleterious factor in melioidosis [12 , 13] , were decreased in Casp11-/- mice compared to WT mice . However , when Casp11-/- mice were infected with a lower inoculum and examined at different time points p . i . , IL-1β ( like IL-18 ) was detected in their BALF at the same level , or even higher , than in WT mice ( Fig 1D ) . Interestingly , neutrophils were present in the BALF of Casp11-/- mice in significantly reduced number than in WT mice 48 and 72 hours p . i . ( Fig 1E ) . This phenotype , however , was not observed 24 hours p . i . , was not due to impaired production of TNFα , IL-6 , KC , or MCP-1 , and correlated with the increased bacterial burdens in organs ( S1 Fig ) . Whether absence of caspase-11 negatively affects recruitment , survival , or permanence of neutrophils in the infected lung cannot be determine at present and will be the focus of future studies . Impaired neutrophil recruitment has been previously observed in Casp11-/- mice during lung infection with K . pneumoniae [24] . Analysis of bone marrow-derived macrophage ( BMM ) cultures infected with B . thailandensis confirmed that IL-1β and IL-18 production is mediated by caspase-1 and not caspase-11 ( Fig 2A ) . Induction of pyroptosis in these cells was primarily dependent on caspase-1 with negligible contribution of caspase-11 ( Fig 2B ) . For these experiments BMM were primed O/N with IFNγ . However , when bone marrow-derived dendritic cells ( S2 Fig ) or BMM were primed with IFNγ concomitantly to infection ( see below ) , caspase-11 contribution to pyroptosis was modest but statistically significant . Intracellular B . thailandensis replication in BMM inversely correlated to occurrence of pyroptosis with maximal bacterial count in Casp1-/- and Casp1/Casp11-/- cells ( Fig 2B ) . B . thailandensis replication in Casp11-/- cells was moderately higher than in WT cells . Interestingly , IFNγ priming of cells significantly decreased bacteria replication in all strains emphasizing the importance of inflammasomes-independent microbicidal mechanisms activated by IFNγ ( see below ) . Although B . thailandensis has been shown to serve as a useful model for melioidosis , it is not pathogenic in humans and differs in many aspects from the virulent B . pseudomallei . Therefore , it was important to determine the role of caspase-11 even during infection with B . pseudomallei . As shown in S3 Fig , two light emitting B . pseudomallei clinical isolates robustly replicated in Casp1-/-/Casp11-/- macrophages but to a much lower degree in WT and Casp11-/- cells , again indicating the prominent role of caspase-1 , rather than caspase-11 , in restriction of intracellular replication of Burkholderia in macrophages . Taken together , these results conclusively indicate that the increased susceptibility of Casp1-/- mice to melioidosis is likely due to their inability to produce IL-18 and trigger pyroptosis in myeloid cells . In contrast , the reason for the susceptibility of Casp11-/- mice remains unclear but does not appear to be due to lack of IL-18 or gross inability to trigger pyroptosis in macrophages . The fact that pyroptosis of B . thailandensis-infected macrophages and cytokine processing in these cells appeared mostly dependent on caspase-1 with minor involvement of caspase-11 raised the question of why Casp11-/- mice appeared so susceptible to melioidosis . Caspase-11 function has been extensively studied in myeloid cells but its role in non-hematopoietic cells has been mostly neglected . To increase our understanding of the role played by Caspase-11 during melioidosis we performed bone marrow transplant experiments . As shown in Fig 3A , the bacterial burden in different organs of WT mice reconstituted with Casp11-/- bone marrow cells was significantly higher compared to WT mice receiving WT cells , as expected for a hematopoietic role for caspase-11 . However , Casp11-/- mice reconstituted with WT bone marrow cells still had organ bacteria burdens much higher than WT mice indicating that caspase-11 also plays a protective role in the radio-resistant cell compartment . Analysis of bone marrow and spleen cells indicated complete and equally effective reconstitution by both genotypes . ( S4 Fig ) . However , confirming what observed in Casp11-/- mice ( Fig 1E ) , severely decreased neutrophil numbers were detected into the lung of mice reconstituted with caspase-11-deficient bone marrow ( Fig 3B and S4 Fig ) . These results suggest that caspase-11 plays a protective role not only in hematopoietic cells but also in radio-resistant cell types . B . thailandensis can infect several cell types including lung epithelial cells [25 , 26] . The mouse lung epithelial cell line TC-1 is often used to study the lung innate immune response to bacteria [27 , 28] . TC-1 cells were readily infected with B . thailandensis ( S5A Fig ) . To test whether caspase-11 can be activated in lung epithelial cells infected with B . thailandensis , TC-1 cells were incubated with Biotin-VAD-FMK , a cell permeable caspase pseudosubstrate that irreversibly binds to active caspase [29] . As shown in Fig 4A , caspase-11 could be pulled down from B . thailandensis-infected TC-1 cell lysates using streptavidin agarose . The B . thailandensis bsaZ mutant , which is unable to escape the phagosome , activated caspase-11 to a much lower degree . Caspase-11 expression in TC-1 cells was strongly induced by TNFα/IFNγ while expression of the canonical inflammasome components NLRP3 , NLRC4 , ASC , Caspase-1 was not detectable ( S5B Fig ) suggesting that caspase-11 may be the only pathway available in TC-1 cells to trigger pyroptosis . This was confirmed by knocking-out caspase-11 gene in TC-1 cells using CRISPR/CAS9 technology ( Fig 4B ) . While the control TC-1 cells underwent pyroptosis upon infection with B . thailandensis , the caspase-11-deficient TC-1 cells were resistant to B . thailandensis-induced cell death ( Fig 4C ) . Intracellular B . thailandensis replication proceeded unrestrained in TC-1 cells lacking caspase-11 but was significantly restricted in control TC-1 cells ( Fig 4C ) . TC-1 cells expressed IL-18 mRNA upon treatment with TNFα/IFNγ though IL-18 secretion could not be detected in conditioned culture supernatants ( S5C Fig ) . To test whether cell death of lung epithelial cells occurs in vivo , we exposed WT or Casp11-/- infected mice to a green fluorescent compound that stains the nucleus of necrotic cells in situ . Mice were euthanized shortly thereafter and the lungs were fixed in paraformaldehyde . Histological sections were counterstained with the epithelial marker EpCAM and analyzed by immunofluorescence microscopy to visualize necrotic epithelial cells in situ . As shown in Fig 4D and 4E , significantly decreased cell death of lung epithelial cells was observed in lung sections of Casp11-/- mice compared to WT mice . It was recently shown that pyroptotic cells trap intracellular bacteria and are phagocytosed by neutrophils [30] . Our preliminary results ( S5D Fig ) suggest that pyroptotic epithelial cells encounter the same fate and are phagocytosed by neutrophils and macrophages . Taken together , these results suggest that one of the most critical functions of caspase-11 in melioidosis is to control pyroptosis of lung epithelial cells . IL-18 is a potent inducer of IFNγ , a cytokine required to survive infection with B . thailandensis [12 , 31] . In agreement with this activity of IL-18 and with previous works from our and others labs , IFNγ production was severely decreased in Casp1-/- mice but not Casp11-/- mice ( Fig 1C ) suggesting lack of IFNγ as a possible mechanism to explain the protective effect of IL-18 . Confirming this hypothesis , administration of recombinant IFNγ significantly reduced organ bacteria burdens in Il18-/- mice infected with a lethal dose of B . thailandensis ( Fig 5A ) showing that IFNγ is sufficient to mediate the protective action of IL-18 . However , IL-18 performs other functions and , therefore , we asked whether IFNγ was necessary for the protective action of IL-18 . As shown in Fig 5B , administration of recombinant IL-18 significantly decreased organ bacterial burdens in WT mice but not in Ifngr1-/- mice . IL-18 treatment induced IFNγ in both mouse strains ( Fig 5C ) . Taken together , these results demonstrate that IFNγ is necessary and sufficient to mediate the protective action of IL-18 . We next turned our attention on the role of IFNγ in melioidosis . A number of studies including from our group [6–8 , 12] , have demonstrated the protective role of IFNγ during B . thailandensis infection though the mechanism of protection remains undefined . IFNγ is known to activate several microbicidal mechanisms that are critical for killing intracellular bacteria , including different families of GTPases , NRAMP1 , NADPH oxidases and iNOS . Recent work suggested that caspase-11 should also be considered as an IFNγ-inducible mechanism [15] . Because both ROS and caspase-11 have been shown to be protective against B . thailandensis infection while iNOS , NRAMP1 , or GBP do not appear to play a significant role in the innate immune response against this bacterium [15 , 21 , 22 , 32] , we decided to determine to what degree the protective action of IFNγ in melioidosis is mediated by either ROS or caspase-11 . As shown in Fig 6A , administration of recombinant IFNγ to intranasally infected mice significantly reduced the organ bacterial burdens in WT , Casp1-/-/Casp11-/- and Casp11-/- mice but not in Cybb-/- mice ( deficient in the gp91 subunit of the NADPH oxidase ) , suggesting that production of ROS is an essential microbicidal mechanism triggered by IFNγ against B . thailandensis . The importance of ROS as mediator of IFNγ protection was also observed in culture of BMMs infected with B . thailandensis ( Fig 6B ) . Intracellular bacteria replication was drastically higher in Casp1-/-/Casp11-/- and Casp11-/- macrophages compared to WT cells and correlated with the decreased pyroptosis in cells lacking either caspase . Intracellular B . thailandensis replication was also elevated in Cybb-/- macrophages but this was not due to decreased pyroptosis , which was not significantly different than in WT cells . Treatment with IFNγ significantly restricted bacteria replication in WT , Casp1-/-/Casp11-/- , and Casp11-/- cells but , importantly , not in Cybb-/- cells . The decreased bacteria replication was not due to higher rate of pyroptosis , which was not significantly affected by treatment with IFNγ . Taken together , these results suggest that production of ROS plays a predominant role in the antimicrobial action of IFNγ . Confirming the protective role of NADPH oxidase downstream of IFNγ , ex vivo generation of ROS was significantly impaired in neutrophils or macrophages obtained from the BALF of infected Il18-/- or Ifngr1-/- mice 14 hours or 48 hours p . i . ( Fig 7A ) . The organ bacterial burdens in these mice inversely correlated with the amount of ROS produced ( Fig 7B ) . Importantly , administration of the antioxidant N-acetyl-cysteine ( NAC ) dissipated the protective effect of exogenous IFNγ administration ( Fig 7C ) , again reinforcing the notion that ROS is a major mediator of the protective effect of IFNγ in melioidosis . NAC treatment had no effect on the production of IL-1β or IL-18 , whose BALF levels correlated with the bacterial burdens ( Fig 7D ) .
The innate immune response to lung infection with Burkholderia species has been examined in a few papers but much remains to be learned . Here we have analyzed the role of the canonical and non-canonical inflammasomes and of the IL-18-IFNγ axis in a mouse model of melioidosis . We and others have previously shown that processing and secretion of the mature form of IL-1β and IL-18 in response to Burkholderia infection was dependent on caspase-1 [12 , 14 , 15] . The caveat of those studies is that they were performed using mice that also lacked caspase-11 . The recent generation of bona fide caspase-1-deficient mice [23] allowed us to examine for the first time the role of this caspase independently of concomitant absence of caspase-11 . Our data conclusively demonstrate that processing and secretion of IL-1β and IL-18 in response to B . thailandensis infection in vivo or in vitro is completely dependent on caspase-1 but unaffected by absence of caspase-11 . The fact that IL-1β secretion is not reduced in absence of caspase-11 also indicates that activation of the NLRP3 inflammasome , which we previously showed exclusively controls IL-1β and IL-18 secretion in response to Burkholderia species infection [12 , 13] , does not occur as a consequence of caspase-11-mediated pyroptosis and potassium efflux , as in other circumstances [33] . Our results also show that pyroptosis of macrophages infected with B . thailandensis or B . pseudomallei and restriction of intracellular bacteria replication is primarily mediated by caspase-1 with minor involvement of caspase-11 . Previous works have attributed a more prominent role to caspase-11 in the pyroptosis of B . thailandensis-infected BMM [15] . It should be noted that those studies relied on Asc-/-Nlrc4-/- cells or Casp1-/-/Casp11-/- cells reconstituted with transgenic human caspase-4 as proxy of bone fide caspase-1 deficient cells , two models that may not faithfully represent caspase-1 absence . Our results also indicate that experimental variables , such as the length of IFNγ priming , may lead to discordant conclusions regarding the involvement of caspase-11 in the pyroptosis of myeloid cells . In fact , it has been proposed that caspase-11 may function as a back-up mechanism to trigger pyroptosis in situations where caspase-1 may be inactivated [34] . The most important result of our study was obtained through bone marrow adoptive transfer experiments and the analysis of the role of caspase-11 in epithelial cell . Our data show that while cell death in B . thailandensis-infected macrophages occurred primarily through caspase-1 , with caspase-11-dependent pathway playing a secondary role , pyroptosis of lung epithelial cells was exclusively dependent on caspase-11 and efficiently restricted intracellular B . thailandensis replication in these cells . Interestingly , lung epithelial cells do not express canonical inflammasome components and therefore depend exclusively on caspase-11 for induction of pyroptosis . It is surprising to observe that Casp1-/- mice that are unable to release mature IL-18/IL-1β or trigger pyroptosis in myeloid cells appear as susceptible ( if not more , Fig 1B ) as Casp11-/- mice that are sufficient for both functions . At face value , this result would attribute equal importance to pyroptosis triggered by caspase-11 in epithelial cells and to that triggered by caspase-1 in myeloid cells plus IL-18 production , a notion previously underappreciated . Thus , caspase-11 dependent pyroptosis of infected lung epithelial cells may be the main protective mechanism triggered by the non-canonical inflammasome in melioidosis . The non-canonical inflammasome was recently shown to restrict S . typhimurium replication in intestinal epithelial cells [35] . Caspase-11 has also been shown to control pyroptosis of endothelial cells during endotoxemia-induced lung injury [36] . Thus , it is conceivable that activation of caspase-11 in cell types other than myeloid or epithelial cells may also play a protective role in melioidosis , an issue we will examine in future studies . Extending our previous work , we show here that IL-18’s protective action in melioidosis is exclusively dependent on its ability to induce IFNγ . This is an important result because IL-18 , in addition to being a strong inducer of IFNγ , also has many other activities . In fact , it has been shown that IL-18 can protect from Streptococcal infections independently of IFNγ [37] . The fact that IFNγ appeared to be indispensable to survive B . thailandensis infection prompted us to investigate the downstream effector mechanisms triggered by IFNγ and responsible for the protection . We concentrated on caspase-11 and the NADPH oxidase because both pathways were already known to provide protection from B . thailandensis infection and because both are IFNγ-inducible , though it was unclear which one contributed more prominently to the IFNγ protective effect . Our data show that in vitro and in vivo the protective action of IFNγ is dependent on production of ROS through the NADPH oxidase system while caspase-11 was dispensable . A previous study has concluded that IFNγ primes caspase-11 in vivo to protect from melioidosis [15] . Although that study ruled out contributions from iNOS and GBP encoded on chromosome 3 , the role of NADPH oxidase was not examined . Moreover , that study used a strain of B . thailandensis that has been passaged into Casp1-/-/Casp11-/- mice to acquire higher virulence and used the intraperitoneal infection route , rather than the intranasal one , as in our study . Although it is clear that caspase-11 priming is a necessary step for the function of this molecule , it should be pointed out that several inflammatory stimuli , including TLR agonists produced by B . thailandensis , can prime caspase-11 as efficiently as IFNγ . Interestingly , it was shown that human caspase-4 does not require IFNγ priming in vivo [15] . The results presented here also indicate that caspase-11 may control recruitment of neutrophils to the infected lung . A similar observation has been previously reported during infection with K . pneumoniae [24] . The reason for the impaired inflammatory response of Casp11-/- mice is unclear and actively pursued in our lab . Preliminary analysis failed to detect defective production of the main neutrophil-specific chemotactic factors . It is conceivable that chemotactic alarmins released by pyroptotic epithelial cells may be the missing factor in Casp11-/- infected mice . Whether neutrophils are effective against Burkholderia species is an unresolved issue . We and others have shown that neutrophils are not very effective against this bacterium and that excessive neutrophil recruitment to the infected lung becomes deleterious due to tissue damage caused by release of neutrophil elastase [12 , 13 , 38] . For these reasons , we think it is unlikely that the high susceptibility to melioidosis of Casp11-/- mice is primarily due to the observed defective neutrophil recruitment . In conclusion , our results identify non-redundant mechanisms activated by the canonical and the non-canonical inflammasomes that confer host protection in melioidosis: Caspase-1-dependent activation of the IL-18-IFNγ-NADPH oxidase axis and pyroptosis in myeloid cells and caspase-11-dependent pyroptosis of infected lung epithelial cells .
All the animal experiments described in the present study were conducted in strict accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health . All animal studies were conducted under protocols approved by the Rosalind Franklin University of Medicine and Science Institutional Animal Care and Use Committee ( IACUC #B14-17 ) . All efforts were made to minimize suffering and ensure the highest ethical and humane standards . C57BL/6J , B6 . SLJ , Il18-/- , Casp1-/-/Casp11-/- , Cybb-/- , and Ifngr1-/- mice were purchased from Jackson lab . Casp11-/- mice were provided by Vishva Dixit ( Genentech ) and Casp1-/- mice by Mohamed Lamkanfi ( VIB Belgium ) . All mouse strains were on C57BL/6J genetic background and were bred under specific pathogen-free conditions in the RFUMS animal facility . Age- ( 8–12 weeks old ) and sex-matched animals were used in all experiments . Experimental groups were composed of at least 5 mice , unless stated otherwise . B . thailandensis E64 was obtained from ATCC . B . pseudomallei K96243 and 390b are clinical virulent isolates . Bacteria were grown in Luria broth to mid-logarithmic phase , their titer was determined by plating serial dilutions on LB agar , and stocks were maintained frozen at -80°C in 20% glycerol . For mice infections , frozen stocks were diluted in sterile PBS to the desired titer . Mice were anesthetized using isoflurane and the infectious doses were applied to the nares in 50 μl total volume PBS . Recombinant murine IL-18 ( MBL , Nagoya , Japan ) was delivered intranasally ( 1 μg ) 6 hours prior to bacterial infection . Two additional IL-18 treatments ( 1 μg/each ) were administered by intraperitoneal injections at 12–15 hours intervals before euthanasia . Recombinant murine IFNγ ( Pepro Tech , NJ , USA ) was administered by intraperitoneal injections ( 2 μg ) once daily for two days . In other experiments , 1 μg IFNγ was administered by intraperitoneal injections once daily for two days in the presence or absence of 10 mg N-acety-L-cysteine ( NAC , Sigma ) delivered at 12–15 hours intervals for 2 days . All cytokines were diluted to desired concentrations with PBS and PBS alone was applied as control . Organs aseptically collected were weighted and homogenized in 1 ml PBS . Serial dilutions were plated on LB agar plates containing Streptomycin ( 100 μg/ml ) using the Eddy Jet Spiral Plater ( Neutec ) . Bacterial colonies were counted 24 hours later using the Flash & Grow Automated Bacterial Colony Counter ( Neutec ) . BALF were collected from euthanized mice by intratracheal injection and aspiration of 1 ml PBS . Cytokines levels in tissue culture conditioned supernatants , BALF , or sera were measured by ELISA using the following kits: MCP-1 , IFNγ , TNFα , KC , IL-1α , IL-1β , IL-6 ( eBioscience ) , and IL-18 ( MBL Nagoya , Japan ) . Cells obtained from BALF were counted and stained with anti-CD11b , anti-CD11c , anti-F4/80 , anti-Ly6G , anti-NK1 . 1 and acquired with a LSRII BD flow cytometer . For reactive oxygen species ( ROS ) measurement , BALF were collected and immediately spun down at 300 × g for 10 minutes to collect cells . Cells were loaded with 7 μM freshly prepared 2' , 7'-dichlorodihydrofluorescein diacetate , H2DCFDA ( Molecular Probes ) in PBS at 37 °C for 30 minutes . Cells were stained with anti-CD11b , anti-CD11c , anti-Ly6G , and anti-F4/80 for 10 minutes , washed twice with PBS , and immediately acquired using a LSRII BD flow cytometer . The ROS level was assessed by MFI of DCF ( an oxidized product of H2DCFDA ) using FITC-channel . Data was analyzed using FlowJo ( TreeStar , OR , USA ) software . Cell lysates were separated by SDS-PAGE , transferred to PVDF membranes , and probed with anti-Caspase-11 antibody ( Abcam , ab180673 ) , anti-β-Actin antibody ( Cell signaling , 4967 ) , anti-Cyclophilin B antibody ( Abcam , ab178397 ) . HRP-conjugated anti-Rabbit IgG antibody ( Sigma , A0545 ) was used as secondary antibody . Immunoblots were developed using ECL method and exposed to X-ray film . Release of LDH in tissue culture media , a reflection of pyroptosis , was measured using the Roche Cytotoxicity Detection Kit ( Roche Applied Science , 11644793001 ) . BMM or TC-1 cells were plated in 48-well plates . Bacteria were added to the cell culture and the plates were centrifuged at 300x g for 10 minutes to maximize and synchronize infection and incubated for 30 minutes ( BMM ) or 2 hours ( TC-1 ) at 37°C . Cells were washed with PBS to remove extracellular bacteria and medium containing kanamycin and gentamicin ( 200 μg/ml each ) was added to inhibit extracellular bacteria growth . Media were collected at 4 and 8 hours post infection for LDH measurement . Cells were lysed in PBS-2% saponin-15% BSA and serial dilutions of the lysates were plated on LB agar plates containing streptomycin ( 100 μg/ml ) . For real time cell culture infection with B . pseudomallei , BMM infected with strain 390B or K96193 ( MOI 10 ) in black 96-well plates were incubated in a 37° C , humidified , 5% CO2 atmosphere IVIS Spectrum camera system . Images were captured every 10 min for 10 hr using capture settings of 1 min with medium binning . Grid ROI measurements of Total Flux ( p/s ) per well were extracted for plotting luminescence of viable bacteria as a function of infection time . All work with B . psuedomallei was performed under biosafety level-3 ( BSL3/ABSL3 ) containment according to policies and standard operating procedures approved via the University of Louisville Committee on Biocontainment and Restricted Entities , The University of Louisville has been approved for select agent work by the Centers for Disease Control and Prevention . Bone marrow from 8-weeks old B6 . SJL ( CD45 . 1 ) mice or Casp-11-/- mice ( CD45 . 2 ) was harvested and 106 bone marrow cells were injected intravenously into lethally irradiated ( 1040 rad ) B6 . SJL or Casp11-/- mice ( 8-weeks of age ) . Chimeric mice were infected five weeks later . Peripheral blood , bone marrow cells , and splenocytes were stained with anti-CD45 . 1 and anti-CD45 . 2 ( Biolegend ) and analyzed by flow cytometry to confirm the efficiency of bone marrow reconstitution . TC-1 were kindly provided by Thomas Kawula ( Washington State University ) and grown in RPMI1640-10% FCS . To target caspase-11 gene , TC-1 cells were transfected using Effectene reagent ( Quiagen ) with Caspase-11 CRISPR/CAS9 KO plasmid and caspase-11 HDR plasmid ( Santa Cruz sc-419462 and sc-419462-HDR ) . TC-1 cells were also transfected with Control CRISPR/CAS9 plasmid ( sc-418922 ) as control . Cells positive for GFP and RFP expression were sorted using BD FACSAria II cell sorter and single cell clones isolated . Expression of caspase-11 in different clones was measured by RT-PCR and immunoblot . One clone was selected that lacked caspase-11 protein expression and produced an aberrant Casp4 mRNA that lacked exons 3 , 4 , and 5 ( S6 Fig ) . TC-1 cells were seeded into 10 cm tissue culture dishes and infected when confluent with B . thailandensis or the bsaZ mutant ( MOI 500 ) in a final volume 5 ml in the presence or absence of 100 ng/ml IFNγ . Three hours later , cell monolayers were extensively washed with D-PBS to remove extracellular bacteria and incubated for another 4 hours in medium containing gentamicin and kanamycin ( 200μg/mL ) . Biotin-VAD-FMK ( 15 μM , Santa Cruz ) was added to the culture medium one hour before lysing cells in RIPA buffer . Active caspase-11 was pulled down by incubation overnight at 4 °C with streptavidin agarose ( Sigma Aldrich ) and analyzed by Western Blot with rabbit anti-caspase-11 antibody ( Abcam , ab180673 ) Mice infected with B . thailandensis ( 5x105 CFU ) for 48 hours were administered intranasally with the Image-iT DEAD Green viability stain ( Invitrogen , 1 nmole in 50 μl saline ) and euthanized 30 minutes later . Lung were perfused and fixed in 4% paraformaldehyde/PBS and embedded in paraffin . Four-micron sections were stained with anti-EpCam antibody ( Abcam , ab213500 ) followed with Alexa Fluor-647 to label Clara cells and Alveolar Type II pneumocytes and visualized with a Nikon Eclipse 80i Microscope equipped with photometrics coolsnap ES2 imaging system . For quantification of dead cells , up to 400 EpCAM-positive cells were counted in six random fields and scored for nuclear positivity to the green fluorescent viability stain . All data were expressed as mean ± S . D . Survival curves were compared using the log rank Kaplan-Meier test . Mann-Whitney U test , One-way ANOVA Tukey Post-test , or unpaired t-test were used for analysis of the rest of data as specified in the figure legends . Significance was set at p<0 . 05 . | Burkholderia pseudomallei is a bacterium that infect macrophages and other cell types and causes a diseases called melioidosis . Inflammasomes are multiprotein complexes that control activation of the proteases caspase-1 and caspase-11 resulting in production of the inflammatory mediators IL-1β and IL-18 and death of infected cells . Mice deficient of caspase-1 or caspase-11 are more susceptible to infection with B . pseudomallei or the closely related B . thailandensis . Here we show that absence of caspase-1 completely abolished production of IL-1β and IL-18 as well as death of macrophages infected with B . thailandensis . In contrast , in the highly susceptible caspase-11-deficient mice , IL-1β and IL-18 production and macrophages death were not significantly affected . Rather , absence of caspase-11 abolished death of infected lung epithelial cells . Taken together , our results show that caspase-1 and caspase-11 have non-redundant protective roles in melioidosis: Caspase-1 primarily controls cell death of infected macrophages and production of IL-18 . In contrast , caspase-11 mediates cell death of infected lung epithelial cells . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] | [
"blood",
"cells",
"medicine",
"and",
"health",
"sciences",
"immune",
"cells",
"melioidosis",
"burkholderia",
"infection",
"immunology",
"epithelial",
"cells",
"bone",
"marrow",
"cells",
"animal",
"models",
"bacterial",
"diseases",
"model",
"organisms",
"inflammasomes",
... | 2018 | Caspase-11-dependent pyroptosis of lung epithelial cells protects from melioidosis while caspase-1 mediates macrophage pyroptosis and production of IL-18 |
Virus-infected cells secrete a broad range of interferon ( IFN ) subtypes which in turn trigger the synthesis of antiviral factors that confer host resistance . IFN-α , IFN-β and other type I IFNs signal through a common universally expressed cell surface receptor , whereas IFN-λ uses a distinct receptor complex for signaling that is not present on all cell types . Since type I IFN receptor-deficient mice ( IFNAR10/0 ) exhibit greatly increased susceptibility to various viral diseases , it remained unclear to which degree IFN-λ might contribute to innate immunity . To address this issue we performed influenza A virus infections of mice which carry functional alleles of the influenza virus resistance gene Mx1 and which , therefore , develop a more complete innate immune response to influenza viruses than standard laboratory mice . We demonstrate that intranasal administration of IFN-λ readily induced the antiviral factor Mx1 in mouse lungs and efficiently protected IFNAR10/0 mice from lethal influenza virus infection . By contrast , intraperitoneal application of IFN-λ failed to induce Mx1 in the liver of IFNAR10/0 mice and did not protect against hepatotropic virus infections . Mice lacking functional IFN-λ receptors were only slightly more susceptible to influenza virus than wild-type mice . However , mice lacking functional receptors for both IFN-α/β and IFN-λ were hypersensitive and even failed to restrict usually non-pathogenic influenza virus mutants lacking the IFN-antagonistic factor NS1 . Interestingly , the double-knockout mice were not more susceptible against hepatotropic viruses than IFNAR10/0 mice . From these results we conclude that IFN-λ contributes to inborn resistance against viral pathogens infecting the lung but not the liver .
Viral infection of vertebrate cells triggers innate immune responses , which result in rapid synthesis of IFN and other pro-inflammatory cytokines [1]–[4] . Virus-induced IFN represents a complex mixture of IFN subtypes which act on target cells by engaging two distinct cell surface receptors [5] . All members of the type I IFN family which , in the mouse , includes 14 different IFN-α subtypes , IFN-β , IFN-κ , IFN-ε and limitin , use the same heterodimeric IFN-α/β receptor complex ( IFNAR1/2 ) for signaling [6] . By contrast , signaling by type III IFN family members ( in the mouse IFN-λ2 and IFN-λ3 ) occurs through the heterodimeric interleukin-28 receptor α/interleukin-10 receptor β ( IL-28Rα/IL-10Rβ ) complex [7] , [8] . Although activating distinct receptor systems , IFN-λ and type I IFNs trigger strikingly similar responses in target cells which mostly result from phosphorylation-induced activation of transcription factors STAT-1 and STAT-2 [9] , [10] . The IFNAR1/2 complex is present on most if not all nucleated cells , whereas expression of the IL-28Rα subunit seems to be cell type-restricted [11] , [12] . Consequently , antiviral protection by type I IFN is observed in most cell types , whereas antiviral protection mediated by IFN-λ is restricted to cells that express functional IL-28R complexes . The spectrum of cell types that respond to IFN-λ in vivo is poorly defined . Recent experiments suggested that epithelial cells are the main targets of IFN-λ in the mouse [13] . Information on the contribution of IFN-λ to virus resistance at the level of the whole organism is very limited as mice lacking functional IFN-λ receptors ( IL28Rα0/0 ) were generated only recently [14] . Unlike knockout mice lacking functional type I IFN receptors ( IFNAR10/0 ) that are highly susceptible to a broad spectrum of different viruses [15] , IL28Rα0/0 and wild-type mice did not differ significantly in resistance to a large panel of pathogenic viruses [14] . The only observed difference between wild-type and IL28Rα0/0 mice was that treatment of knockout mice with toll-like receptor ( TLR ) 3 and TLR9 agonists failed to induce resistance to vaginal infection with herpes simplex virus type 2 [14] . Here we used Mx1+/+ mice to investigate the relative contributions of IFN-λ and type I IFN in immunity toward influenza A virus . Mx1+/+ mice differ from standard mouse strains in being fully IFN-competent . They carry functional alleles of the influenza virus resistance gene Mx1 , which is defective in standard laboratory mice [16] . Consequently , in Mx1+/+ mice , virus-induced IFN activates the Mx1 gene in addition to other antiviral genes , leading to a more complete innate immune response and more robust resistance to influenza and influenza-like viruses [17] , [18] . The Mx1+/+ mouse model system has the power to reveal even subtle defects in antiviral immunity against orthomyxoviruses . It has recently been used to uncover the beneficial effect of IFN-β in influenza virus defense [19] . It was further used to demonstrate that IFN-α might be used to prevent disease induced by highly lethal human H5N1 influenza viruses [17] . Using this experimental system we now demonstrate that IFN-λ contributes to innate immunity against influenza virus but not against two different hepatotropic viruses . These differences in virus susceptibility correlated with the differing ability of virus-induced IFN-λ to activate the Mx1 gene in lung and liver of IFNAR10/0 mice .
Since virus-induced activation of IFN genes requires positive feedback through the IFN-α/β receptor in certain cell types [20] , we first determined whether the major IFN subtypes are induced in lung and liver of IFNAR10/0 mice after infection with viruses that strongly activate the innate immune system . As can be seen in Figure 1 , we observed strong transcriptional activation of genes for the IFN-α family , IFN-β and IFN-λ2 in the lung of mice infected intranasally with the influenza A virus mutants SC35M-ΔNS1 and PR8-ΔNS1 that are known to induce large amounts of type I IFN [21] , [22] . Similarly , strong transcriptional activation of IFN-α , IFN-β and IFN-λ genes was observed in the liver of IFNAR10/0 mice infected with a mutant of hepatotropic Thogotovirus ( THOV ) that lacks the IFN-antagonistic factor ML ( THOV-ΔML ) [23] . In a first experiment , IFNAR10/0 mice were treated with exogenous IFN-λ by the intranasal route to determine whether this cytokine might contribute to protection from pneumonia induced by pathogenic influenza viruses . Groups of IFNAR10/0 mice were treated with 7 , 500 units of either recombinant IFN-λ2 or IFN-λ3 . Control animals received corresponding volumes of a mock preparation . Ten hours later , the animals were infected with 100 plaque-forming units ( pfu ) ( ∼20 LD50 ) of mouse-adapted H7N7 influenza A virus strain SC35M [22] . The control animals quickly lost weight and had to be killed between days 7 and 9 post infection due to clinical signs of pneumonia , whereas all animals treated with either IFN-λ2 or IFN-λ3 remained healthy ( Fig . 2A ) . Since standard IFNAR10/0 mice lacking functional Mx1 alleles developed severe disease under identical experimental conditions in spite of treatment with IFN-λ3 ( data not shown ) , we concluded that the protective effect of IFN-λ that we observed in our Mx1+/+ mice was mainly mediated by the IFN-induced resistance factor Mx1 . In a second experiment , 15 , 000 units of IFN-λ3 were applied by the intraperitoneal route to IFNAR10/0 mice carrying functional Mx1 alleles . Ten hours later the animals were challenged with 100 pfu ( ∼20 LD50 ) of THOV . Animals treated with IFN-λ3 as well as control animals treated with a mock preparation developed severe disease between 48 and 96 hours post infection ( Fig . 2B ) . Thus , IFN-λ exhibited effective antiviral activity in the lung , but seemed to be inactive in the liver . Like type I IFN , IFN-λ exhibits antiviral activity by binding to a specific cell receptor complex that can activate latent STAT transcription factors [24] . After activation , the STAT proteins move to the nucleus where they activate transcription of a large number of IFN-responsive genes , including Mx1 . To determine whether exogenous IFN-λ activates IFN-responsive genes in our IFNAR10/0 mice , we harvested lung and liver at 20 hours post onset of treatment with IFN-λ3 and analyzed the tissue homogenates for Mx1 protein by western blotting . Easily detectable levels of Mx1 were present in lungs of mice that were treated intranasally with 3 , 500 units of IFN-λ3 ( Fig . 3A ) . The lungs of mock-treated mice did not contain detectable levels of Mx1 . Mx1 protein could not be detected in the liver of IFNAR10/0 mice treated with 15 , 000 units of IFN-λ3 by the intraperitoneal route ( Fig . 3B ) . If , as a control , a cross-reactive variant of human IFN-α was injected by the same route into wild-type mice , Mx1 was prominently induced in the liver ( Fig . 3B ) . To distinguish between the possibility that liver cells lack functional receptors for IFN-λ and the possibility that the recombinant IFN-λ failed to reach the liver under our experimental conditions , we analyzed the Mx1 protein levels in IFNAR10/0 mice infected with THOV-ΔML which strongly induces IFN-λ in the liver ( Fig . 1 ) . The liver of mice with severe THOV-induced disease contained no detectable amounts of Mx1 protein ( Fig . 3B ) . Similarly , no Mx1 protein could be detected in the liver of terminally ill IFNAR10/0 mice infected with Rift Valley fever virus clone 13 ( Fig . 3B ) , another hepatotropic virus with strong IFN-inducing activity [25] . Thus , differential IFN-λ receptor expression in lung and liver seemed to explain why exogenously applied IFN-λ protected IFNAR10/0 mice from virus-induced disease of the lung but not the liver . To directly assess the contribution of IFN-λ to the protection from influenza virus-induced lung disease , we generated IL28Rα0/0 mice carrying functional Mx1 alleles by crossbreeding of appropriate mouse strains and compared the fate of wild-type and IL28Rα0/0 mice after challenge with 5×104 plaque-forming units ( pfu ) of SC35M . Survival of IL28Rα0/0 mice was slightly reduced compared to wild-type mice ( Fig . 4A ) , but the difference was not statistically significant . Viral titers in lungs of IL28Rα0/0 mice were slightly but significantly higher at 72 h post infection than in lungs of wild-type mice ( Fig . 4B ) . To determine the relative contributions of IFN-α/β and IFN-λ in antiviral defense we generated Mx1+/+ mice that lack functional receptors for both of these two classes of IFN ( IFNAR10/0IL28Rα0/0 ) and compared them to mice that lack receptors for IFN-α/β only . We previously demonstrated that IFNAR10/0 mice with intact Mx1 alleles are highly susceptible to challenge infections with wild-type SC35M [19] . However , intranasal infection with 105 pfu of SC35M-ΔNS1 did not induce disease in IFNAR10/0 mice ( Fig . 5A ) . Similarly , all wild-type and IL28Rα0/0 mice remained healthy when challenged with up to 106 pfu of SC35M-ΔNS1 ( data not shown ) . In marked contrast , all IFNAR10/0IL28Rα0/0 double-knockout mice infected with 105 pfu of SC35M-ΔNS1 developed severe disease and had to be killed around day 5 post infection ( Fig . 5A ) . Additional experiments in which we used lower doses of challenge virus demonstrated that the LD50 of SC35M-ΔNS1 in IFNAR10/0IL28Rα0/0 double-knockout mice was approximately 103 pfu ( Fig . 5A ) . A similar picture emerged when the mice were challenged with a NS1-deficient variant of the H1N1 human influenza A virus strain PR8 ( PR8-ΔNS1 ) . At a dose of 106 pfu , all infected IFNAR10/0IL28Rα0/0 double-knockout mice developed severe pneumonia within 4–6 days post infection , whereas all IFNAR10/0 single-knockout mice remained healthy ( Fig . 5B ) . Importantly , our single- and double-knockout mice did not differ in susceptibility to infection with the two hepatotropic viruses THOV-ΔML ( Fig . 5C ) and RVFV clone 13 ( Fig . 5D ) , strongly supporting the above-formulated conclusion that IFN-λ is not active in the liver of IFNAR10/0 mice . Virus replication in lungs of wild-type and mutant mice was assessed at 48 hours post infection with 105 pfu of SC35M-ΔNS1 . Virus titers in lungs of wild-type mice were below the detection limit in four of five animals , and they were only slightly above the detection limit in lungs of IL28Rα0/0 mice at 48 h post infection ( Fig . 6A ) . Remarkably , SC35M-ΔNS1 did not grow much better in lungs of IFNAR10/0 mice , whereas it replicated to very high titers in lungs of IFNAR10/0IL28Rα0/0 double-knockout mice ( Fig . 6A ) . At 20 hours post infection with SC35M-ΔNS1 the Mx1 protein levels in lungs of IL28Rα0/0 mice were about 2-fold lower than in the wild-type animals ( Fig . 6B ) . Lungs of infected IFNAR10/0 mice contained about 10-fold lower levels of Mx1 protein than wild-type mice , whereas Mx1 levels were below the detection limit in IFNAR10/0IL28Rα0/0 double-knockout mice ( Fig . 6B ) . Thus , after infection with SC35M-ΔNS1 , the extent of Mx1 gene expression in lungs of mice with defective receptors for IFN-α/β , IFN-λ or both correlated inversely with virus titers .
The intracellular signaling pathways activated by IFN-λ and IFN-α/β are quite similar [9] , [10] , suggesting that both IFN types are contributing to virus resistance . Surprisingly , however , mice lacking functional receptors for IFN-λ did not differ from wild-type mice when challenged with a panel of different pathogenic viruses [14] . A mild deficiency of IFN-λ-deficient mice became only apparent in an experimental setting in which resistance to herpes simplex virus type 2 was induced by treating the animals with TLR3 or TLR9 agonists [14] . This phenotype is in marked contrast to that of mice lacking functional receptors for IFN-α/β which are highly susceptible to many viruses [15] . We reasoned that the different phenotypes of the knockout mice might be explained by the different expression patterns of the receptors for IFN-α/β and IFN-λ in the organism . Receptors for IFN-α/β are rather uniformly expressed on most if not all nucleated cells [26] , whereas receptors for IFN-λ are preferentially expressed on epithelial cells [13] . If our reasoning was correct , one would predict that the protective effect of IFN-λ should be restricted to organs with a high percentage of cells expressing the IFN-λ receptor and that the protective effect of IFN-λ in these organs might be most obvious when the IFN-α/β system is defective . In this report we provide evidence that strongly supports this view . We observed that intranasal application of IFN-λ protected the mice from lethal challenge with influenza A virus , whereas systemic application of IFN-λ failed to mediate protection from disease induced by a hepatotropic virus ( Fig . 2 ) . It should be noted that the mice employed here lacked functional IFN-α/β receptors , excluding the possibility that the protective effect in the lung was indirect and resulted from IFN-α/β that might have been induced by undefined contaminating substances in our IFN-λ preparations . Protection against influenza virus correlated with the presence of the IFN-induced Mx1 protein in the lung tissue ( Fig . 3 ) , suggesting that lung epithelial cells carry functional IFN-λ receptors . By contrast , no Mx1 protein was found in liver tissue of mice treated with IFN-λ ( Fig . 3 ) . The liver tissue also failed to respond to IFN-λ synthesized in the virus-infected liver ( Fig . 1 ) , suggesting that mouse liver cells do not express functional receptors for IFN-λ . This latter conclusion is in agreement with results from recent quantitative RT-PCR analyses which showed that the alpha chain of the IFN-λ receptor ( IL28R-α ) is expressed only at very low levels in liver of mice [13] . However , our results appear to be in conflict with a previous report in which IFN-λ was successfully used to inhibit hepatitis B virus replication in a murine hepatocyte cell line expressing the viral genome as a transgene [27] . However , these authors observed no induction of IFN-responsive genes in the liver of mice treated with large amounts of IFN-λ , and they observed no inhibition of hepatitis B virus replication in vivo [27] . In this respect , hepatocyte cell lines may not mirror the normal behavior of hepatocytes in intact liver tissue . Since the virus challenge studies in a former report [14] were carried out with IFN-λ receptor knockout mice lacking the IFN-induced influenza virus resistance factor Mx1 , it remained possible that the beneficial effect of IFN-λ against influenza virus had previously been underestimated . Yet , our new experiments with Mx1-positive mice revealed that the lack of IFN-λ system has indeed a much less drastic effect on virus resistance than the lack of the IFN-α/β system . The protective role of IFN-λ became only apparent in Mx1-positive mice that lack a functional IFN-α/β system , and it was most prominent if influenza virus mutants with high IFN-inducing potential were used ( Fig . 5 ) . It is well known that highly pathogenic influenza viruses are not controlled well by the IFN system because the virus-encoded NS1 protein counteracts efficient activation of IFN genes in infected cells [21] . NS1-deficient influenza viruses which are very potent IFN inducers are highly attenuated in wild-type mice but remain virulent in mice lacking STAT-1 [21] , a transcription factor centrally placed in the signaling pathways of all IFN types [28] . We found that mutants of the influenza virus strains SC35M and PR8 lacking NS1 were completely non-virulent in IFN-α/β receptor-deficient mice and failed to replicate efficiently in the lung of such mice ( Figs 5A and 5B ) , which should not be the case if IFN-α/β was the only IFN subtype that confers resistance to influenza viruses . Our observation that double knockout mice lacking functional receptors for IFN-α/β and IFN-λ are highly susceptible to the NS1-deficient influenza virus mutants clearly demonstrates that IFN-λ provides the residual protection in IFN-α/β receptor-deficient mice . Some important conclusions can be drawn from our data regarding the role of different IFN types in antiviral immunity . First , the virus defense strategy of the lung is not exclusively based on the IFN-α/β system . Our data clearly demonstrate that the IFN-λ system also contributes to innate immunity against influenza A virus . The second important conclusion from our study is that the IFN-α/β system is dominant over the IFN-λ system . IFN-λ thus appears to be part of a secondary defense system which can fill gaps left by the IFN-α/β system . Future studies will help to distinguish between the possibility that IFN-λ is predominantly active against influenza viruses and the possibility that IFN-λ plays a broader role in the lung and improves innate immunity against other pathogenic viruses that infect the respiratory tract . Evidence in favor of the second possibility includes the observation that IFN-λ also restricted vaccinia virus replication in the lung of mice [29] . We further noted with interest that , reminiscent to the situation with NS1-deficient influenza virus , IFN-α/β receptor-deficient mice are able to restrict the growth of human respiratory syncytial virus in the lung far better than STAT-1-deficient mice [30] . This observation suggests that IFN-λ might also help controlling respiratory syncytial virus . Since receptors for IFN-λ are expressed on epithelial cells of many different organs including lung , stomach and intestine [13] , it is conceivable that the physiological role of this cytokine is to protect the host from viral infections via mucosal membranes at many different body sites . An important issue to be addressed in the future is whether IFN-λ might serve a similar role in humans .
All animals used were of C57BL/6 genetic background . Congenic B6 . A2G-Mx1 mice [31] carrying intact Mx1 alleles and B6 . A2G-Mx1-IFNAR10/0 mice lacking functional type I IFN receptors [19] were bred locally . C57BL/6 mice lacking functional type III IFN receptors ( IL28Rα0/0 ) [14] were crossed with B6 . A2G-Mx1 and B6 . A2G-Mx1-IFNAR10/0 mice to produce strains with intact Mx1 alleles and defective alleles for IL28Rα only , or IL28Rα and IFNAR1 in combination . Six- to eight-week-old animals were used for all infection experiments , which were performed in accordance with the guidelines of the local animal care committee . Animals were euthanized if severe symptoms developed or body weight loss approached 30% of the initial value . We used wild-type influenza A virus strains SC35M ( H7N7 ) and A/PR/8/34 ( H1N1 ) as well as mutants SC35M-ΔNS1 [22] and PR8-ΔNS1 [21] lacking the IFN-antagonistic factor NS1 . We further used wild-type Thogotovirus ( THOV ) or mutant THOV-ΔML lacking the IFN-antagonistic factor ML [23] , and the attenuated “clone 13” strain of Rift Valley fever virus ( RVFV ) lacking functional IFN-antagonistic factor NSs [25] . All these viruses are classified as BSL2 pathogens in Germany . Animals were anesthetized by intraperitoneal injection of a mixture of ketamine ( 100 µg per gram body weight ) and xylazine ( 5 µg per gram body weight ) before intranasal infection with the indicated doses of the various influenza A viruses in 50 µl PBS containing 0 . 3% BSA . For THOV and RVFV infections , 100 µl-samples of diluted virus stocks were applied intraperitoneally without anaesthesia . IFN-λ2 and IFN-λ3 were produced by transient transfection of 293T cells with appropriate expression plasmids [9] . The biological activity of IFN-λ2 and IFN-λ3 was determined as previously described [32] . Hybrid human IFN-αB/D that is highly active on mouse cells was used as positive control [17] , [33] . Samples containing the indicated amounts of IFN-λ2 or IFN-λ3 were either applied intranasally ( 50 µl ) to anesthetized animals or injected intraperitoneally ( 200–300 µl ) without anaesthesia . Lung homogenates were prepared by grinding the tissue using a mortar and sterile quartz sand . Homogenates were suspended in 1 ml of PBS , and tissue debris was removed by low speed centrifugation . Virus titers in supernatants were determined by performing plaque assays on MDCK II cells by serial 10-fold dilutions in PBS containing 0 . 3% BSA . Lung and liver were removed , and frozen immediately in liquid nitrogen . RNA was isolated from the organs using 1 ml of TriFast according to the protocol of the manufacturer ( peQLab ) . The RNA was further purified by using RNeasy mini kit columns ( Qiagen ) . One µg of each RNA preparation was reverse-transcribed using random-hexamer primers and reverse transcriptase . The reaction products were used to amplify the cDNAs by Taq polymerase for 30 cycles using the indicated primer pairs for mouse IFN-β ( accession no . NM_010510 , primers from positions 21–42 and 145–124 ) , the mouse IFN-α family ( accession no . NM_010504 , primers from positions 46–68 and 557–534 ) , mouse IFN-λ2 ( accession no . NM_001024673 , primers from positions 83–104 and 191–170 ) , and mouse β-actin ( accession no . X03672 , primers from positions 1374–1396 and 1585–1564 ) . RT-PCR products were separated by agarose electrophoresis , stained with ethidium bromide and visualized under UV light . Lung homogenates were prepared by grinding the tissue using a mortar and sterile quartz sand . Homogenates were lysed in buffer containing 50 mM Hepes ( pH 7 . 3 ) , 125 mM NaCl , 1% Nonidet P-40 , 1 mM EDTA , 0 . 5% sodium deoxycholate , 0 . 1% SDS , 1 mM DTT , 100 units/ml of benzonase , and protease inhibitors as recommended by the manufacturer ( Roche ) . Lysates were subjected to low speed centrifugation , and supernatants were diluted with concentrated gel loading buffer containing β-mercaptoethanol . Proteins were separated by SDS-polyacrylamide gel electrophoresis ( 10% gel ) and transferred onto polyvinyliden-fluoride membranes ( Millipore ) . The blots were probed with monoclonal mouse antibody specific for Mx1 [34] and monoclonal mouse antibody against actin ( Sigma ) . Horseradish peroxidase-labeled secondary antibodies and the chemoluminescence detection system ( Pierce ) were used to detect primary antibodies . Signal quantification was done with a ChemiDoc XRS equipment ( BioRad ) . | The contribution of IFN-λ to innate immunity against virus-induced diseases has remained unclear to date as appropriate mouse models were not available . We now present evidence that IFN-λ is involved in the antiviral defense . Mice lacking functional IFN-λ receptors were only slightly more susceptible to influenza virus than wild-type mice , but intranasal administration of IFN-λ efficiently protected IFN-α/β receptor-deficient mice from lethal influenza virus infection and induced the antiviral factor Mx1 in lungs . Mice lacking functional receptors for both IFN-α/β and IFN-λ were hypersensitive and failed to restrict even usually non-pathogenic influenza virus mutants lacking the IFN-antagonistic factor NS1 . By contrast , intraperitoneal application of IFN-λ failed to induce Mx1 in the liver of mice and did not protect against hepatotropic viruses . Furthermore , double-knockout mice were not more susceptible against hepatotropic viruses than IFN-α/β receptor-deficient mice , indicating that IFN-λ contributes to resistance against viral pathogens infecting the lung but not the liver . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"infectious",
"diseases/respiratory",
"infections",
"virology/host",
"antiviral",
"responses",
"infectious",
"diseases/viral",
"infections",
"immunology/innate",
"immunity"
] | 2008 | Interferon-λ Contributes to Innate Immunity of Mice against Influenza A Virus but Not against Hepatotropic Viruses |
Automated in situ hybridization enables the construction of comprehensive atlases of gene expression patterns in mammals . Such atlases can become Web-searchable digital expression maps of individual genes and thus offer an entryway to elucidate genetic interactions and signaling pathways . Towards this end , an atlas housing ∼1 , 000 spatial gene expression patterns of the midgestation mouse embryo was generated . Patterns were textually annotated using a controlled vocabulary comprising >90 anatomical features . Hierarchical clustering of annotations was carried out using distance scores calculated from the similarity between pairs of patterns across all anatomical structures . This process ordered hundreds of complex expression patterns into a matrix that reflects the embryonic architecture and the relatedness of patterns of expression . Clustering yielded 12 distinct groups of expression patterns . Because of the similarity of expression patterns within a group , members of each group may be components of regulatory cascades . We focused on the group containing Pax6 , an evolutionary conserved transcriptional master mediator of development . Seventeen of the 82 genes in this group showed a change of expression in the developing neocortex of Pax6-deficient embryos . Electromobility shift assays were used to test for the presence of Pax6-paired domain binding sites . This led to the identification of 12 genes not previously known as potential targets of Pax6 regulation . These findings suggest that cluster analysis of annotated gene expression patterns obtained by automated in situ hybridization is a novel approach for identifying components of signaling cascades .
Basic processes that cells go through—fate specification , proliferation , differentiation , migration , and programmed death—are driven by gene networks that are , for the most part , still poorly understood . In the past few years , several large-scale approaches have been launched to begin unraveling the regulatory pathways governing cell behavior . Novel strategies include RNA interference screens , interactome analysis , transcriptome mapping by microarrays , and ChIP-on-chip assays . Such cell-based analyses portray regulatory pathways as complex networks [1] . Analysis of gene expression by in situ hybridization ( ISH ) has proven to be a powerful means to validate pathways because ISH provides high-resolution information on gene expression in cells within the context of location in the organ or organism [2] . Recently , ISH on tissue sections has been automated , making it possible to determine gene expression patterns for thousands of genes , and thus enabling the construction of gene expression atlases ( e . g . , [3–5]; reviewed by [6] ) . In view of past success using ISH data in the validation of single regulatory interactions , it is likely that transcriptome-scale gene expression atlases will facilitate large-scale validation of regulatory networks [7] and , more importantly , contribute to the discovery of network components . To provide a proof of concept for this idea , we first applied ISH to generate a gene expression atlas of the midgestation mouse embryo ( E14 . 5 ) populated with approximately 1 , 000 genes , including known developmental genes as well as many genes whose expression and function in development has not previously been examined . We then textually annotated expression patterns of this atlas and utilized hierarchical clustering to mine for genes involved in the development of the cerebral neocortex , a brain region that has undergone dramatic structural and functional enhancement during mammalian evolution . Thereafter , we validated candidate genes by a combination of biochemical assays and ISH analyses on mutant embryos ( Figure 1 ) . The present study focuses on genes regulated by Pax6 , a transcription factor with both a paired domain and a homeodomain . Pax6 is an evolutionary highly conserved master mediator of development [8] and plays important roles in the formation of the mammalian cerebral cortex [9] , eye [10] , and pancreas [11 , 12] . The cortex of the midgestation mouse embryo consists of the ventricular , subventricular , and intermediate zones ( VZ , SVZ , and IZ , respectively ) , a subplate , a cortical plate , and the MZ ( marginal zone ) [9] . Pax6 is expressed in the VZ and SVZ where the stem cells of the cortex reside . A naturally occurring Pax6 null mutation termed Small eye ( Pax6sey or sey ) [13] is characterized by an enlargement of VZ and SVZ and an IZ that is populated by neurons normally found in the SVZ [14 , 15] . At the molecular level , Pax6 and the nuclear orphan receptor gene Nr2e1 ( Tlx ) appear to synergistically regulate the formation of the SVZ in which progenitors of the outer cortical layers arise ( reviewed by [16] ) . Hence in the absence of Pax6 and even more so in the absence of both Pax6 and Nr2e1 , the outer cortical layers are severely affected [17] . Our computational analysis of annotated expression patterns generated by large-scale ISH led to 12 new genes harboring a Pax6-paired domain binding site . More generally , this gene expression atlas of the midgestation embryo combined with robotic ISH offers a novel approach for finding putative components of genetic networks regulating critical aspects of mammalian development .
The spatial expression patterns of 1 , 030 genes ( “1K dataset” , Table S1 ) were determined by ISH on series of 24 equally spaced sagittal tissue slices from E14 . 5 embryos . This stack of sections provides a comprehensive representation of the tissues and organ primordia characteristic for this developmental stage . Digoxigenin-tagged riboprobes used for ISH were prepared by in vitro transcription from DNA templates that had been generated by PCR from cDNA using pairs of gene-specific primers ( for details see [18] ) . Following robotic ISH , high-resolution images of all sections were captured using a custom-made automated microscope [19] and resulting images were deposited on a public database , termed “genepaint . org” ( Figure 1A and [20] ) . Images are retrievable from this database using gene symbol , gene name , gene ID , or genepaint set ID [20] , which are provided in Table S1 . Images can be viewed using browser-integrated applications . A total of 887 ( 86% ) genes of the 1K dataset were at least weakly expressed at E14 . 5 while 144 ( 14% ) transcripts were not detected in any tissue . In parallel to the ISH analysis , a global transcriptome analysis of E14 . 5 embryos was carried out using Affymetrix 430 2 . 0 chips . These data have been made publicly available through the National Center for Biotechnology Information Gene Expression Omnibus ( NCBI GEO ) database ( see database accession number below ) . Using the absolute call attribute “present” as indication for expression , ∼13 , 000 genes were considered to be transcribed at E14 . 5 . Of the 887 genes that were found to be expressed by ISH , 757 could be unambiguously linked to one or several probe sets on the microarray . Of these , 480 ( 63% ) were scored as present by at least one probe set on each duplicate array . The discrepancy between ISH and microarray is partly due to transcripts that are restricted to a small number of cells ( see [18] for a discussion ) . Expert analysis of the ISH data ( see below ) showed that 74% ( 652 ) of the 887 transcribed genes were expressed in a regional manner , i . e . , transcripts for a particular gene were either distributed in a nonuniform manner within a specific tissue or found only in a subset of tissues and organs ( Figure 2A ) . Typical examples for regionally expressed genes directly downloaded from genepaint . org are shown in Figure 2B–2P . Note that robot-generated ISH data have sufficient quality and resolution to unambiguously tie expression to specific anatomical structures occasionally as minute as a single cell ( e . g . , Figure 2C inset ) . To enable searches for expression patterns in genepaint . org and to compare patterns across a large number of genes , we implemented an expert-based controlled vocabulary annotation of expression patterns . A total of 96 hierarchically organized structures ( Figure 3 ) were annotated , which collectively represent the majority of organ primordia , tissues , and tissue subregions that characterize the E14 . 5 embryo . Although the hierarchical tree is composed of 96 items , only 70 structures are unique “leaves” of the tree , i . e . , do not overlap with or fully encompass other structures . For each of the 96 structures , two attributes were allocated . The first attribute indicates the type of expression pattern characteristic for the structure in question ( regional , scattered , and ubiquitous; see Figure 2B–2P for examples ) . The second attribute is the intensity of expression ( not detected , low , medium , and strong expression; see also [20] for definition of attributes ) . In all cases , each annotation was determined by one annotator and confirmed independently by a second expert . In the rare cases of a discrepancy , a consensus decision was found . These textual annotations for all 96 structures for each of the 652 regionally expressed genes are available on genepaint . org , thus enabling user-initiated searches for genes that are expressed with a particular pattern and strength in a single or in multiple structures ( for instructions to carry out such a search using the “structure selection tool” see “Quickguide” of genepaint . org ) . For example , searching for genes that are regionally and strongly expressed in the neocortex of an E14 . 5 embryo will produce a list of ∼170 genes ( August 2007 ) . The validity of queries can readily be checked visually by using the image viewing applications integrated in genepaint . org . Annotation of expression patterns across anatomical structures enables hierarchical clustering of patterns with the prospect of revealing similarities between expression patterns . Within the set of 652 patterns , for every possible pair of pattern annotations , a distance score was calculated to measure the similarity between the two patterns across all anatomical structures . The scoring function at each structure was derived from both the strength of gene expression and the pattern . Genes were hierarchically clustered based upon these scores . Using an analogous procedure , the distances between all pairs of anatomical structures were calculated on the basis of the annotated patterns of expression across all genes . These latter scores were utilized to cluster the anatomical regions ( for details see Text S1 ) . Figure 4 shows the result of this cluster analysis , which is characterized by two main features . First , tissues that have similar embryonic origin and/or similar physiological functions group together ( Figure 4 , dendrogram on top ) . For example , all components of the central and peripheral nervous systems cluster together ( red branches in dendrogram ) . Likewise , endoderm-derived endocrine organs ( pancreas , thyroid , pituitary , and liver ) group together , and organs of the vascular system ( choroids plexus , atrium , ventricle , blood vessels , and meninges ) have a shared expression pattern and thus group . Organs equipped with sensory epithelia ( retina , cochlea , labyrinth , and olfactory epithelium ) as well as organs containing polarized epithelia ( lung , kidney , salivary glands , oesophagus , intestine , and stomach ) form groups . Thus , hierarchical clustering of expression patterns unravels relationships between tissues and to a significant extent the dendrogram recapitulates both the developmental lineage relationships and the shared physiological functions . The second feature of Figure 4 is the classification of the 652 regional expression patterns into 12 groups . Groups 1–5 , 9 , and 11 include genes expressed in a limited number of tissues , while group 10 genes are more broadly expressed . Groups 7 ( 82 genes ) and 8 ( 86 genes ) represent transcripts with a bias to the nervous system . Hierarchical clustering thus systematically compares hundreds of complex expression patterns to group related structures in a way reflecting the body plan of the organism . The emergence of previously established relationships in the embryonic architecture from a statistical analysis of controlled vocabulary annotation validates in part the second result of clustering—the presence of groups of similarly expressed genes . To better appreciate how the spatial expression pattern affects grouping of genes , in Figure 5 we show cluster 7 at higher magnification using the full spectrum of genepaint pattern and intensity annotation icons . The dominance of expression of cluster 7 genes in the tissues of neuroectodermal origin is evident , albeit the genes in the top portion of the tree ( Otx3 to Lrp5 ) are also significantly expressed in mesoderm-derived tissues . For example , the group consisting of Arhgap5 , Fzd2 , and Lrp5 are widely expressed in cells of mesodermal origin but still show expression in the VZ of the telencephalon ( Figure 6 , top red box ) . Figure 6 additionally illustrates the point that genes found on distinct short branches of the tree have nearly identical expression patterns ( see pictures of embryos enclosed in red boxes of Figure 6 ) . For example , a metabotropic glutamate receptor ( Grm3 ) , a zinc finger-containing transcription factor ( Zbtb20 ) , and a putative chloride channel ( Ttyh1 ) share a regional expression pattern in the ventral part of the telencephalon , in midbrain , hindbrain , spinal cord , and in dorsal root ganglia ( red box at the bottom of Figure 6 ) . It thus appears that the clustering was successful in arranging annotated patterns of expression in a meaningful way that is consistent with the image data . Genes with similar expression patterns may be part of a common signalling pathway . Many of the 82 genes of group 7 ( Figures 4–6 and first column in Table S2 ) encode proteins involved in signal transduction including Pax6 , which plays a key role in multiple developmental processes of the nervous system ( see Introduction and Discussion ) . Theoretically , among the 82 genes , several could be up- or downstream of Pax6 . Because the clustering matrix is derived from gene expression in the entire spectrum of annotated E14 . 5 embryonic tissues , we do not expect all 82 genes of group 7 to be coexpressed with Pax6 in the VZ of the developing cortex . To clarify this point , expression of all 82 genes in the germinal zone was examined in the appropriate E14 . 5 datasets of genepaint . org ( Figure 7 ) revealing that 40% ( n = 30 ) were coexpressed with Pax6 and are thus candidates for either regulating or being regulated by Pax6 ( column 2 of Table S2 ) . To determine which of the 30 genes coexpressed with Pax6 in the VZ of the E14 . 5 cortex are potentially part of a Pax6 regulatory network , we passed the candidates through two additional “filters” ( Figure 1B ) . First , we determined expression patterns in Pax6-deficient embryos ( Pax6sey/sey ) , as it is expected that the pattern of expression of Pax6 downstream genes would be changed in mutant tissue . Second , we searched for and subsequently experimentally validated Pax6-paired domain binding sites in those genes whose expression pattern was augmented or decreased in Pax6sey/sey cortex . To examine the expression pattern of the 30 candidate genes , we applied robotic ISH to sections of E15 . 5 mutant and wild-type embryos ( Figure 1B ) . At this developmental stage , cortical layering becomes very distinct ( Figure 8A ) and hence the Pax6sey/sey cortical phenotype is clearly noticeable . Comparing ISH results of wild-type and mutant cortices ( Figure 8 ) yielded a total of 16 genes that have an altered expression pattern in the developing cortex of Pax6sey/sey mice ( Table S2 , third column ) . These genes are Arx , Bcan , D930015E06Rik , Eomes , Igfbpl1 , Fzd2 , Lrp5 , Neurod1 , Neurod4 , Neurod6 , Neurog1 , Neurog2 , Odz2 , Pde1b , Sst , and Trim9 . Several ISH results shown in Figure 8 were validated by quantitative real-time PCR with the result that ISH and quantitative PCR ( qPCR ) analyses are consistent ( Figure S1 ) . Nevertheless , ISH provides a more detailed portrait than qPCR of how the absence of Pax6 protein affects gene expression . For example , while qPCR data indicate that Neurod1 is significantly downregulated in mutant cortex , ISH shows that this downregulation occurs predominantly in the IZ . In the case of Trim9 , qPCR indicates a 2-fold reduction in overall expression level . The ISH data attribute this reduction to the loss of a band of expression in the SVZ . The upregulation of Arx predicted by qPCR is focused to the SVZ . Inspection of the cortical expression patterns of the 16 candidates allowed us to classify them into two groups . Some genes show qualitative changes of expression pattern in the mutant cortex , while others show a mostly quantitative change over the entire area of expression . Genes belonging to the first group are Sst ( Figure 8B1 and 8B2 ) , Arx ( Figure 8D1 and 8D2 ) , Neurod1 ( Figure 8E1 and 8E2 ) , Neurod6 ( Figure 8F1 and 8F2 ) , and Trim9 ( Figure 8G1 and 8G2 ) . In this group , the most subtle changes are shown by Sst ( Figures 8B1 and 8B2 ) . In wild type , neocortical neurons expressing Sst are localized to the marginal zone ( MZ ) and the subplate ( SPL , open arrows in Figure 8B1 ) as well as to the IZ ( black arrows in Figure 8B1 ) . In the mutant , however , Sst-expressing neurons can be found only in the MZ ( open arrow in Figure 8B2 ) . Transcription factor Arx is expressed in scattered neurons at every level of the developing wild-type cortex , and particularly in the MZ ( Figure 8D1 ) . qPCR showed an overall increase in signal in the mutant neocortex ( Figure S1 ) , and ISH indicated that this increase does not occur in the scattered Arx-expressing cells of the IZ and MZ , but takes the form of a novel Arx-expressing domain presumably coinciding with the SVZ of the mutant cortex ( open arrow in Figure 8D2 ) . It is possible that this novel band of Arx-expressing neurons contains basal ganglia neurons that in the Pax6sey/sey cortex could have migrated through the pallio-subpallial barrier and invaded the cortex [21] . In the cases of Neurod1 , Neurod6 , and Trim9 ( Figure 8E–8G ) the mutant cortex did not show novel expression domains but rather a loss of very specific expression regions ( open arrows in Figure 8E1-8G1 ) . Neurog1 , Neurog2 , and Pde1b represent genes with relatively simple spatial expression patterns in the wild-type neocortex ( Figure 8I1-8K1 ) . Accordingly , changes in pattern seen in Pax6sey/sey are simple , taking the form of a disappearance of the single expression domain ( Figure 8I2-8K2 ) . Finally , D930015E06Rik , Lrp5 , and Fzd2 are genes showing quite widespread expression in wild-type neocortex ( Figure 8L1-8N1 ) . ISH indicates a global reduction in expression intensity in Pax6sey/sey cortex ( Figure 8L2-8N2 ) . Of note , even in these cases , ISH suggests a relative upregulation of expression in the subplate of Pax6sey/sey cortex ( open arrows in Figure 8M2 and 8N2 ) . In summary , the expression of half of the 30 candidate Pax6-regulated genes is changed in the cortex of Pax6sey/sey , indicating that a combination of robotic ISH and hierarchical clustering of annotations can be used for prioritizing candidate genes for a next round of analysis . Next , we investigated whether cluster 7 showed an enrichment of Pax6-regulated genes relative to other clusters . Cluster 10 contains 215 genes , many of which are expressed in the neocortex ( Figure 4 ) . A sample of 41 genes ( Table S4 ) was selected that colocalized with Pax6 in the E14 . 5 neocortex in a manner similar to that described for the 30 genes of cluster 7 ( Figure 7 ) . Applying the same criteria as were used for cluster 7 , we find that only ten ( 24% ) of the 41 cluster 10 genes exhibit an altered expression pattern in the E15 . 5 neocortex of Pax6sey/sey mice ( bold marked genes in Table S4 ) . Compared to 16 of 30 differentially expressed genes ( 53% ) in cluster 7 , hierarchical clustering resulted in significant enrichment of putative Pax6 targets among cortically expressed genes in cluster 7 ( p < 0 . 02 , Fisher's exact test , one-tailed ) . Differential gene expression in Pax6sey/sey cortex ( Figure 8 ) raised the possibility that the genes in question harbor Pax6 binding sites . In the absence of Pax6 , these binding sites would no longer be occupied , and this could directly affect gene expression . Alternatively , some of the 16 genes could be downstream of a Pax6-regulated gene thus implying an indirect role of Pax6 . We searched for sites that are conserved between human and mouse [8] and subsequently examined whether the predicted sites were capable of binding to a fusion protein composed of a Pax6-paired domain ( PD ) and a glutathione-S-transferase ( GST ) tag ( Pax6-PD-GST ) . The DNA binding affinities of either the full-length Pax6 protein or of an Escherichia coli made Pax6-PD-GST are similar [22] , prompting us to use the fusion protein for binding site analysis . While such DNA binding experiments would not prove functionality of Pax6 binding in the embryo , they provide a rational procedure to identify genes that share a regulatory element and , at least in part , validate the selection approach described in the previous sections . The extent of sequence conservation between human/mouse in the loci of the 16 differentially expressed genes was determined using the University of California Santa Cruz ( UCSC ) genome browser [23 , 24] ( Figure 9 ) . Human and mouse genomic sequences delineated by this browser were first aligned using the zPicture tool of the Dcode . org Comparative Genomics Center ( http://zpicture . dcode . org ) [25] followed by a definition of Pax6 binding sites in the conserved regions with the help of TRANSFAC professional ( V10 . 2 ) [26] . Previous work had already demonstrated Pax6 binding sites in Sst [27] and Neurog2 [28] . Overall , we identified in 14 of the 16 differentially expressed genes a total of 27 Pax6-conserved binding sites ( Figure 9 open and filled triangles; see Table S3 for mouse-predicted binding site–DNA sequences ) . All predicted sites were examined by electromobility shift assays ( EMSAs ) . Before examining the 27 sites , we carried out a series of validation experiments using previously characterized Pax6-paired domain binding sites . These sites were ACATTCACGCATGACTGACT derived from the Pax6-binding consensus sequence ( ANNTTCACGCWTSANTKMNY ) [22] , a Pax6 binding site identified in the Sst gene ( Table S3 ) [27] , and two sites in the Neurog2 gene ( termed E3 . 2 and E1 . 1 [28]; Table S3 ) . As expected , the consensus sequence-derived positive control yielded a very robust signal . The fragment was markedly shifted , could be competed with a 50-fold excess of nonlabeled DNA , and was supershifted using an anti-GST antibody ( Figure 10 , panel 1 ) . Next , we analyzed the Pax6 binding sites in the Sst and Neurog2 genes . The Sst site was shifted by Pax6-PD-GST and so were E1 . 1 and E3 . 2 ( Figure 10 , panel 2 ) . Of note , E1 . 1 had previously been functionally validated by transgenesis in vivo [27] . Relative to the consensus sequence , the amount of shifted complex , albeit significant , was low . Mutated E1 . 1 was not able to bind to Pax6-PD-GST fusion protein ( Figure 10 , panel 2 , last lane ) . Quantification of the shifted band showed that relative to the positive control 2 . 8 and 23% of the E1 . 1 and E3 . 2 oligonucleotides were protein bound ( Figure 11 ) . All 27 Pax6 binding sites predicted by bioinformatics were subjected to the same analysis as described for the controls . For 11 sites , we could not detect any binding under our conditions ( open triangles in Figure 8 , for DNA sequences see Table S3 ) . For the remaining 16 sites , we observed binding to a variable degree ( Figures 10 and 11 ) . For example , in the case of Arx , the percent of radioactive site bound was 24% relative to the consensus sequence ( Figure 11 ) . This band could be competed away with 50-fold excess of nonlabeled binding site DNA and be supershifted , albeit weakly , with an anti-GST antibody ( Figure 10 , panel 3 ) . Quantitatively similar shifts were observed with Neurod1 ( site 2 ) , Neurod6 , Pde1b ( site 1 ) , and Trim9 . Weak binding was seen for Bcan , D930015E06Rik , Fzd2 , and Neurod4 . Taken together , 12 new genes harboring Pax6-PD binding sequences were identified , in addition to those previously described in Neurog2 and Sst . Thus , of the 80 genes that hierarchical clustering grouped with Pax6 , 14 have experimentally verified Pax6-PD binding sites .
Our annotated and searchable spatial gene expression atlas of the midgestation mouse embryo is a useful resource allowing scientists to search and view expression patterns of individual genes . However , an annotated atlas also provides an entryway to questions that reach far beyond merely viewing expression patterns of individual genes . We demonstrated that hierarchical clustering of the annotation of expression patterns can lead to dendrograms grouping tissues and genes ( Figures 4–6 ) . In the case of organs , the outcome of clustering is striking in that tissues of common origin and function cosegregate , as is the case for the various constituents of the central nervous system . Such convergence not only reflects the fact that central nervous system tissues are chiefly composed of neurons that share a characteristic physiology , but that all neurons derive from neuroectoderm ( for a discussion of this issue , see also [5] ) . Subregions of the developing brain known to be closely related by lineage and function ( e . g . , hippocampus , neocortex , and olfactory bulb ) form neighboring branches of the dendrogram of organs . Tissues that owe their architecture and function to mesenchymal epithelial interactions ( e . g . , lung , kidney , salivary gland , teeth , and whisker follicles ) tend to be neighbors in the organ dendrogram . Hence clustering of annotations of patterns recreates an authentic body plan of the organism , a result reminiscent of that reported for Drosophila [30] . Because clustering is successful in recapitulation of the body plan , it is tempting to assume it also assembles genes in meaningful groups . To test this possibility we asked whether genes bunched in one of the 12 groups ( Figure 4 ) belong to a regulatory cascade . We selected group 7 , one of the larger and hence representative groups containing genes predominantly expressed in the central nervous system . A total of 23 group 7 genes encode transcription factors , including Pax6 , which is one of the most strongly conserved master mediators of eye and brain development in metazoans ( see Introduction ) . Clustering encompassed all annotated organs , but because regulatory networks are likely to exhibit some level of organ specificity , we focused on genes coexpressed with Pax6 during the development of the cortex , the seat of higher cognitive abilities . We found that 30 group 7 genes were coexpressed with Pax6 , 14 of which not only showed altered cortical expression in Pax6sey/seyembryos , but also had experimentally verifiable Pax6-paired domain binding sites . Holm et al . [32] have used microarrays to identify genes that are differentially expressed in the telencephalon of wild-type and Pax6sey/seyembryos at E15 . These data can be compared with our ISH analyses of cortical changes in gene expression at E15 . 5 ( Figure 8 ) . Among the ∼100 transcripts identified by [32] , three of them—Neurog1 , Neurog2 , and Arx—are also in our list . qPCR analysis suggests that Neurog1 and Neurog2 are downregulated in Pax6sey/sey embryos by more than an order of magnitude ( Figure S1 ) . By contrast , these data indicate that Arx is upregulated in the mutant cortex by a factor of 4 . Microarray results show qualitatively similar results , although the change in expression is less pronounced [32] . The same authors [32] found differential regulation of several genes that are contained in the 1K dataset , but did not cluster with Pax6 . Examples are Sema5a and Cxcl12 , which are expressed in the VZ of the neocortex and are downregulated in Pax6sey/sey embryos ( M . Warnecke , J . Oldekamp , G . Alvarez-Bolado , and G . Eichele , unpublished data ) . Both , Sema5a and Cxcl12 reside in cluster 10 , a cluster that includes genes expressed in nervous system and in mesoderm-derived structures ( Figure 4 ) . One way to avoid escaping of such genes is to restrict clustering to a subset of tissue types . Our strategy for discovering components of networks is not restricted to transcription factors; e . g . , kinases and their substrates could be identified by hierarchical clustering , which could be followed up by biochemistry using kinase assays . In this particular case , expression data could be combined with information from the protein interactome that identifies enzyme-substrate interactions . A prerequisite for a successful application of our strategy is a significant degree of regional expression of regulators and their targets . Because of the severity of its cortical , pancreatic , and ocular phenotypes , the Pax6sey/sey mutant has become emblematic for the molecular genetic approach to development . The cellular processes in which Pax6 has a key regulatory role include cell proliferation , adhesion , and migration [33] . Pax6 targets relevant to eye and pancreas development have been identified [11 , 33 , 34] , but in the cortex the Pax6 network has proved most difficult to unravel; Neurog2 is the only known direct target of Pax6 in this tissue [28] . Although the expression of adhesion-related proteins Cdh4 ( R-Cadherin ) and Fut9 is dramatically decreased in the Pax6sey/sey cortex [35–37] , it is not known whether these genes are direct Pax6 targets . Among the 12 new cortically expressed genes containing a Pax6-paired domain binding site , six encode transcription factor proteins ( Arx , Neurod1 , Neurod4 , Neurod6 , Neurog1 , and Neurog2 ) . This advocates for a multilayered activation cascade , i . e . , a network of considerable complexity . Except for Arx , a homeodomain transcription factor , the other transcription factors uncovered by this study belong to the basic helix-loop-helix ( bHLH ) family . These genes , also termed proneural genes , are essential regulators of neurogenesis , coordinating the acquisition of the neuronal fate—the specific neuronal subtype identity appropriate for birth date and location of a neuron [38] . Arx is required for proper forebrain development in humans [39] and mouse [40] . Expression of Arx characterizes a group of neurons that migrate tangentially from the basal ganglia into the cortex . Thus , augmented Arx expression in the IZ of Pax6sey/sey cortex ( Figure 8D1 and 8D2 ) could originate from increased migration of Arx-positive cells . Because Arx is also expressed in cells resident in the cortex , the Arx upregulation observed in the Pax6sey/sey cortex could be caused by a direct de-repression of Arx , a scenario supported by our finding that Pax6 binds to Arx regulatory regions . Previous studies have indicated that Arx positively regulates cell proliferation in the VZ [40] and the upregulation of this gene in the cortex of Pax6sey/sey may account for the thickening of the germinative zone in this mutant [41] . Pde1b is a calcium- and calmodulin-dependent phosphodiesterase and inhibits cyclic nucleotide signaling . Although its role in the neuroepithelium is unknown , this gene could represent a class of effectors in the network by which Pax6 , through binding to the Pde1b promoter , would regulate signaling . Wnt proteins are components of potent signaling cascades with major roles in the development of the brain [42–46] . Genes in the Wnt pathway ( Wnt7b and Tcf7l2-Tcf4 ) have been implicated in Pax6 function in the diencephalon [47 , 48] . Cluster 7 contains both of these genes and additionally Wnt1 , Wnt2b , Wnt3 , Wnt3a , Wnt7a , Fzd2 , and Lrp5 . Fzd2 and Lrp5 are Wnt coreceptors that are coexpressed with Pax6 in the cortex , and the corresponding genes have Pax6-paired domain binding sites . The Wnt pathway is an essential regulator of telencephalic “dorsalization , ” a process that confers cortex-forming capabilities to the dorsal half of the anterior neural tube [49 , 50] . Our study would thus place the Pax6 cascade at the intersection with the Wnt pathway . It is rational for a pathway regulating the subdivision of the cortical neuroepithelium into neocortex and hippocampus to intersect with a pathway involved in activation of proneural genes ( Neurod and Neurog families ) that confer region-specific neuronal subtype traits . Bcan ( Brevican or Cspg7 ) codes for a brain-specific chondroitin sulphate proteoglycan abundant in the extracellular matrix and having a function in the development of axonal tracts [51 , 52] . Intriguingly , the expression pattern of two other chondroitin sulphate proteoglycans ( Cspg3-neurocan and Ptprz1-phosphacan ) are altered in the Emx2 mutant [53] , which is another major transcription factor mutant with a cortical phenotype . Emx2 is thought to work in balance with Pax6 to partition the cortex into specialized functional areas [54] . These data suggest that extracellular matrix proteoglycans are important effectors in the networks responsible for the specification of cortical subdivisions and underline the central role of adhesion events in this process [33] . Trim9 encodes a brain-specific member of the “tripartite motif” protein family , which binds to the cytoskeletal microtubules [55 , 56] and could possibly represent an effector molecule by which Pax6 regulates cytoskeletal function during cell migration or polarization . Trim9 expression marks two sharply delimited bands in the developing cortex , the cortical plate and the SVZ ( Figure 8G1 ) . The latter disappears in the Pax6 mutant ( Figure 8G2 ) indicating a lack of proper differentiation in the SVZ , or perhaps the complete absence of this essential germinal compartment , consistent with previous observations [57] . The expression patterns of genes such as Eomes and Igfbnl1 are affected in the cortex of Pax6sey/seyembryos but no Pax6-paired domain binding sites were found within a conserved region . The present study identifies several genes expressed in the developing cortex that have Pax6-paired domain binding sites . DNA binding as assessed by EMSA shows that binding affinities to the various DNA fragments derived from human/mouse sequence conservation varies considerably . It can readily be seen from Figures 10 and 11 that strength of DNA binding of the Pax6-paired domain to particular DNA sequences is often only a fraction of that of the consensus sequence . It should be noted , however , that the in vivo validated site E1 . 1 of the Neurog2 gene [28] showed binding similar to that seen in e . g . , Neurog1 . Additional experiments using transgenic mice will be required to establish functional relevance of the candidate binding sites identified in the present study . Experimental and computational methods employed in the present work provide a partial list of components of the cortical Pax6 network . A next logical step will be to apply the approach used in the present study to mice mutated in the putative Pax6 target genes identified here . This will eventually provide a portrait of the undoubtedly very complex Pax6 regulatory network . The question is raised of how many components could the Pax6 network be composed ? The present study is based on expression patterns of approximately 8% of genes expressed in the midgestation mouse embryo and has led to 14 genes regulated by Pax6 in the developing cortex . Hence , upon completion of the E14 . 5 transcriptome atlas , the number of Pax6-regulated genes may reach a figure of ∼200 , not including those controlled by transcription factors downstream of Pax6 .
For automated ISH , mouse embryos were embedded in Tissue-Tek O . C . T . ( Sakura ) , quick frozen , and sectioned at 25-μm thickness . Hybridization was performed on a Tecan ISH robot . Nonradioactive , digoxigenin-tagged riboprobes were detected by a two-step chromogenic catalyzed reporter deposition protocol . Riboprobes were generated by standard methods and were usually 700–1 , 000 nucleotides long [18] . Template sequences are available at www . genepaint . org [20] . E14 . 5 mouse embryos belonged to either the NMRI or C57/BL6 strains . The strain for each expression pattern is given in genepaint . org and can be found in the info box on the set viewer page of the gene in question . The E15 . 5 wild-type and Pax6sey/sey embryos were littermates . Small eye is a spontaneous mutation [58] kept in the genetic background C57BL/6J × DBA/2J . Datasets were analyzed with the DAVID GoChart tool [59] . Molecular functions in Figure 2A correspond to GoChart level 2 . For real-time ( RT ) -qPCR , cortex tissue was dissected from E15 . 5 Pax6sey/sey embryos and wild-type littermates . Total RNA was prepared from pooled wild-type or Pax6sey/sey tissue samples . Reverse transcription was accomplished via standard methods . Real-time PCR was carried out with an iCycler ( BioRad ) using a SYBR-green quantification protocol [60] . Primers specific for candidate genes were used to determine expression levels in wild type and Pax6sey/sey . Expression levels were normalized to the house-keeping gene EF1α . Error bars in Figure S1 are standard deviations of replicates . Expression differences were in all cases statistically significant ( p < 0 . 001 ) as determined by Student's t-test . Microarray analysis was performed with RNA extracted from E14 . 5 whole embryos and Mouse Genome 430 2 . 0 microarrays ( Affymetrix ) according to standard methods . Analysis was completed in duplicate , and genes were considered expressed if both replicates had a “present” absolute call ( detection value p < 0 . 05 ) . Putative binding sites in human-mouse conserved regions of the candidate genes were identified with the help of the TRANSFAC ( binding sites ) and UCSC Genome Browser [23 , 24] ( conserved regions ) databases . Dendrograms for genes and anatomical regions were produced with R Project for Statistical Computing software ( http://www . r-project . org/ ) using Ward's clustering [61] . Distances between genes and between anatomical terms were calculated using Python programming language ( http://www . python . org/ ) . The distance metric between each pair of genes was defined on the basis of the strength and pattern of expression at each anatomical region . In a similar fashion , the distance between each anatomical region was defined on the basis of the strength and pattern of expression for every gene in that region ( see Tables S1–S3 in Supplementary Material for details ) . EMSA reactions were carried out according to [22] . The 2 . 48-kb Pax6 cDNA was obtained from Luc St . Onge and Peter Gruss ( Max Planck Institute of Biophysical Chemistry , Göttingen , Germany ) . A 562-bp ( nts 135–697 ) paired domain-containing BamHI-XmaI fragment was cloned in frame with the GST tag into the “pGEM 5×1” expression vector ( Promega ) . The Pax6-PD-GST fusion protein was expressed in BL21 bacteria and extracted according to standard protocols . It was found that 1 μl of bacterial lysate contained a total of 9 . 4 μg of total protein ( Bradford method , using bovine serum as standard ) . To every binding site oligonucleotide ( Table S3 ) two additional sequences were added: on the 5'end , CGC was added as a spacer , and to the 3'end , the 20-mer GGA TCA AGA GCT ACC AAC TC was added allowing primer extension with Klenow DNA polymerase . Radiolabeling with α32P-dATP was carried out with the Klenow DNA polymerase using 10 pmol of binding site oligonucleotide and oligonucleotide primer complementary to the 20-mer added to the 3' end . Labeled DNA was purified with G-50 columns . One microliter of labeled DNA ( about 105 cpm ) was incubated on ice for 30 min with 1 μl of bacterial lysate containing Pax6-PD-GST . For the competition assay , 50-fold excess of unlabeled DNA was incubated with 1 μl of bacterial lysate first , then 1 μl of DNA was added , and the mixture was incubated on ice for another 30 min . For the supershift assay the labeled binding site was incubated with bacterial lysate containing Pax6-PD-GST fusion protein on ice for 30 min , then anti-GST antibody was added and incubated for another 30 min . Finally , 5 μl of each product were loaded on a precooled 5% polyacrylamide gel and electrophoresed at 4 °C and 150 V for 2 . 5 h . Autoradiographs and gels were aligned , shifted bands and free probe bands were cut out , radioactivity contained therein was measured in a scintillation counter , and the ratio of shifted to total radioactivity was calculated for each experiment . This ratio was normalized to that obtained from binding of Pax6-PD-GST to the Pax6 consensus sequence ( “positive control” in Table S3 ) , which was included in all experiments . Also included in each analysis was a negative control , binding of Pax6-PD-GST to mt-E1 . 1 of Neurog2 ( Table S3 ) .
The accession number mentioned in this paper from the National Center for Biotechnology Information Gene Expression Omnibus ( NCBI GEO ) ( http://www . ncbi . nlm . nih . gov/geo ) is GSE6081 . | Signaling pathways drive biological processes with high specificity . Reductionist approaches such as mutagenesis provide one strategy to identity components of pathways . We used high throughput in situ hybridization to systematically map the spatiotemporal expression pattern of ∼1 , 000 developmental genes in the mouse embryo . The rich information collectively contained in these patterns was captured in annotation tables that were systematically mined using hierarchical clustering , resulting in 12 groups of genes with related expression patterns . We show that this process generates biologically meaningful , high-content information . The expression pattern of developmental master regulator Pax6 is found in a cluster together with that of 81 other genes . The paired DNA binding domain of Pax6 can bind to regulatory sequences in 14 of the 81 genes . We also found that the expression pattern of all these 14 genes is up- or downregulated in Pax6 mutant mice . These results emphasize that determining the expression pattern of many genes in a systematic way followed by an application of integrative tools leads to the identification of novel candidate components of signaling pathways . More generally , when complemented with appropriate data-mining strategies , transcriptome-scale in situ hybridization can be turned into a powerful instrument for systems biology . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods",
"Supporting",
"Information"
] | [
"developmental",
"biology",
"mammals",
"mus",
"(mouse)",
"neuroscience",
"animals",
"genetics",
"and",
"genomics"
] | 2007 | Regulatory Pathway Analysis by High-Throughput In Situ Hybridization |
Following Schistosoma japonicum ( S . japonicum ) infection , granulomatous responses are induced by parasite eggs trapped in host organs , particular in the liver , during the acute stage of disease . While excessive liver granulomatous responses can lead to more severe fibrosis and circulatory impairment in chronically infected host . However , the exact mechanism of hepatic granuloma formation has remained obscure . In this study , we for the first time showed that follicular helper T ( Tfh ) cells are recruited to the liver to upregulate hepatic granuloma formation and liver injury in S . japonicum-infected mice , and identified a novel function of macrophages in Tfh cell induction . In addition , our results showed that the generation of Tfh cells driven by macrophages is dependent on cell–cell contact and the level of inducible costimulator ligand ( ICOSL ) on macrophages which is regulated by CD40–CD40L signaling . Our findings uncovered a previously unappreciated role for Tfh cells in liver pathology caused by S . japonicum infection in mice .
Schistosomiasis remains a major public health problem in many developing countries in tropical and subtropical regions , which affects approximately 200 million people worldwide [1] , [2] . Following Schistosoma japonicum and mansoni ( S . japonicum and mansoni ) infections , eggs are trapped in host liver and trigger the formation of granuloma to surround eggs . However , severer granulomatous response precipitates fibrosis in the liver more quickly and eventually cause extensive tissue scarring , which in turn causes severer circulatory impairment of the affected organs [3] . Thus , better understanding the mechanism of granuloma formation is crucial to prevent excessive granulomatous lesions in schistosome infection . Through an unclear mechanism , CD4+ T cell response induced by egg antigens orchestrates the development of granulomatous lesions around individual eggs in host liver [3] . Naïve CD4+ helper T ( Th ) cells recognize schistosome egg antigens presented by antigen-presenting cells ( APCs ) to differentiate into distinct effector subsets . These include Th1 , Th2 , and Th17 cells or regulatory T ( Treg ) cells , which differentiate under specific cytokine milieu . These populations express chemokine receptors for their homing to liver and produce distinct profiles of effector cytokines , which play roles in liver granuloma formation and regulation . For example , Th2 and Th17 cells were reported to upregulate hepatic granuloma formation by secreting IL-4 and IL-17 respectively [4]–[6] , while Th1 and Treg cells were reported to downregulate hepatic granuloma formation [5] , [7] . Follicular helper T ( Tfh ) cells are a further distinguishable subset of Th cells [8] , [9] , which are characterized by the expression of numerous molecules including the surface markers CXCR5 , PD-1 , and ICOS , and the transcription factor Bcl-6 [10] , [11] . The reported key function of Tfh cells is to provide help to B cells to support their activation , expansion , and differentiation , as well as the formation of the germinal center ( GC ) [12] , [13] . Recently , some evidence has emerged to support the role of Tfh cells in the development of autoimmune pathology by providing help to B cells to produce autoantibodies [14]–[16] . Except for providing help to B cells , it is currently unknown whether Tfh cells have other functions . For example , previous reports showed that Tfh cells are induced and are the predominant source of IL-4 in reactive lymph nodes during helminth infection [17] , [18] . Although Tfh cells were induced during schistosome infection [19] and most IL-4-producing CD4+ T cells in reactive lymph nodes produced in response to S . mansoni antigens are Tfh cells [20] , it is not yet clear whether Tfh cells are involved in the development of liver pathology during schistosome infection . A number of cellular interactions between antigen-presenting cells ( APC ) and naïve precursors underlie Tfh cell development . For example , B cells are important for the generation of Tfh cells [13] , [21]–[25] . Dendritic cells ( DCs ) have been shown that can also drive Tfh cell development even in the absence of T-B cell interactions [26] , [27] . In addition , late activator antigen-presenting cell [28] and plasma cells [29] are also reported to be involved in the generation of Tfh cells . However , little is known with regard to whether macrophages , one important subset of APCs and playing a key role in the liver granuloma formation in chronic schistosomiasis japonica [30] , [31] , are involved in the generation of Tfh cells . In this study , we identified a novel role for Tfh cells in liver pathology by using a S . japonicum-infected mouse model , and provided new insights into the promotion of Tfh cell generation by macrophages .
To assess whether Tfh cells are expanded in mice infected with S . japonicum , we analyzed the percentages of CXCR5highPD-1high CD4+ T cells ( Tfh cells ) [32] in the spleens , lymph nodes , and livers of mice . Result in Figure 1 showed that the percentages and absolute numbers of CXCR5highPD-1high CD4+ Tfh cells were significantly increased in the spleen , lymph nodes , and liver of S . japonicum infected mice ( Figure S1 , Figures 1A , 1B , and 1C ) . Tfh cells are also characterized by altered expression of other markers , such as the transcription factor Bcl6 and the costimulatory receptor ICOS [10] . Thus , to further confirm the above CXCR5highPD-1high CD4+ T cells are Tfh cells , their expression of Bcl6 and ICOS was examined . Result in Figure 1D showed that CXCR5highPD-1high CD4+ Tfh cells expressed high levels of Bcl6 and ICOS compared to non-Tfh cells in the spleen , lymph nodes , and liver of S . japonicum infected mice . The interaction between ICOS and ICOSL is required for Tfh cell differentiation , but not for Th2 cell differentiation and migration into nonlymphoid tissues , and ICOSL knockout ( KO ) mice were widely used as a Tfh cell deficiency model [17] , [33]–[35] . To investigate whether Tfh cells are required for liver granuloma formation , we infected ICOSL KO mice that lack Tfh cells [8] with S . japonicum . Results showed that the percentages ( Figures 2A and 2B ) and absolute numbers ( Figure 2C ) of CXCR5highPD-1high Tfh cells were significantly reduced in the spleen , lymph nodes , and liver of ICOSL KO mice compared to wild-type ( WT ) mice . Meanwhile , the area of granuloma and the severity of the fibrosis in livers of Tfh cell-deficient mice were significantly reduced than that in WT mice ( Figures 2D , 2E , 2G and 2H ) and liver injury was milder in ICOSL KO mice ( Figure 2F ) . Interestingly , our results showed that schistosome antigen specific IgG and IgG1 production was greatly reduced in ICOSL KO infected mice ( Figure S2 ) , which consistent with the published data that proved the critical role of Tfh cells in production of antibody . To investigate the impact of ICOSL deficiency on other CD4+ T cell subsets involved in granuloma formation , we detected the percentages of Th1 , Th2 , Th17 and Treg cells in spleen , lymph nodes and liver in mice infected with S . japonicum . We found that both Th2 and Th17 cells , which were reported to upregulate hepatic granuloma formation [4] , [5] , showed no significant difference between ICOS KO mice and their WT controls with or without infection ( Figures S3A and S3B ) . Although the significant differences of Th1 and Treg cells in spleen and/or lymph nodes between ICOS KO mice and their WT controls were observed before infection , both Th1 and Treg cells , which were reported to downregulate hepatic granuloma formation [5] , [7] , were less in infected ICOSL KO mice livers ( Figures S3C and S3D ) . These data suggest that significantly reduced Tfh cells may account for the amelioration of liver pathology in ICOSL KO mice infected with S . japonicum . To further elucidate the contribution of Tfh cells to liver granuloma formation in S . japonicum-infected ICOSL KO mice , Tfh cells , non-Tfh , or Th2 control cells were purified from S . japonicum-infected WT mice and adoptively transferred into ICOSL KO mice 5 weeks after S . japonicum infection . Result in Figure S4 showed that eGFP+ Tfh cells still expressed the molecular markers of CXCR5 and PD-1 three weeks post-transfer . Results showed that compared with phosphate buffered saline ( PBS ) group , granuloma size and the levels of serum ALT/AST were not statistically significantly increased in mice receiving non-Tfh cells ( comprised of pooled antigen-specific Th1/Th2/Th17/Treg cells ) , which suggests that pooled antigen-specific CD4+ T cells may not be sufficient to promote the granuloma formation and liver injury . Of note , the area of granuloma and severity of fibrosis were significantly exacerbated after adoptive transfer of Tfh or Th2 cells into infected KO mice ( Figures 2D , 2E , 2G and 2H ) . Moreover , the adoptive transfer of Tfh or Th2 cells resulted in a significant increase in the levels of serum ALT/AST ( Figure 2F ) , compared with control mice injected with PBS alone or non-Tfh control cells . These results suggest that Tfh cells play a pivotal role in promotion of the liver granuloma formation and liver injury , although we did not directly rule out the possibility that it might be partially resulted from more antigen-specific CD4+ T cells in “Tfh group” than that in the “non-Tfh control group” . Taken together , these data suggest that Tfh cells contribute to liver pathology in mice infected with S . japonicum . To determine whether Tfh cells can be recruited to the liver of infected mice , we adoptively transferred eGFP+ Tfh cells or eGFP+CD44+CXC5−PD-1−CD4+ T ( eGFP+ non-Tfh ) cells from eGFP mice into S . japonicum-infected ICOSL KO mice . Three days after transfer , by flow cytometry and fluorescence microscopy , we found that most eGFP+ Tfh cells were observed in the blood circulation in uninfected ICOSL KO mice . However , in S . japonicum-infected ICOSL KO mice , most eGFP+ Tfh cells appeared in the liver ( Figures 3A and 3B ) . The eGFP+ Tfh cells were limited to appear in the granulomas forming around newly deposited eggs within 3 days after transfer . Interesting , most eGFP+ non-Tfh cells , which include Th1/Th2/Th17/Tregs , also appeared mainly in the blood circulation in uninfected mice , while eGFP+ non-Tfh cells were mainly in spleen , lymph nodes and liver in S . japonicum-infected mice . In addition , more GFP+ Tfh cells were found in S . japonicum-infected mice liver than eGFP+ non-Tfh cells ( Figures 3A and 3B ) . These data suggest that Tfh cells , which may be induced by APCs probably including B cells and/or DCs in lymph nodes and spleen as previously reported , are recruited to the mice liver and promote the granuloma formation around eggs after infected with S . japonicum . Macrophages make up approximately 30% of total liver granuloma cells in S . japonicum-infected mice , and most of these macrophages ( 50–90% ) display Ia antigens , acting as professional APCs [36] . Considering macrophages co-locate in spleen and lymph nodes with other two important APCs B cells and DC , and play an important role as APC in induction of the differentiation of Th1 [37] , Th2 [38] , Th17 [39] , [40] , and Treg cells [41]–[43] , it was interesting to investigate whether macrophages had an effect on the generation of Tfh cells . To address this question , we cultured CD4+ T cells together with macrophages from normal or infected mice in the presence or the absence of soluble egg antigens ( SEA ) extracted from S . japonicum eggs . SEA is a mixture of antigens including numerous molecules of protein , glycoprotein , glycolipid , lipoprotein and saccharide . SEA provides polyclonal stimulations to immune cells including APC and CD4+ T cells . Results showed that CD4+ T cells not only increased the surface expression of CXCR5 and PD-1 ( Figures 4A and 4B ) but also upregulated transcripts of the Tfh cell “master regulator” Bcl-6 and ICOS ( Figures 4C and 4D ) when exposed to macrophages from the infected mice . B cells and DCs , the other two important professional APCs co-located in spleen and lymph nodes with macrophages , are reported to have the ability to induce Tfh-cell development . Result showed that compared with B cells ( Figures 4E and 4F ) and DCs ( Figures 4G and 4H ) from infected mice , macrophages from infected mice induced a higher frequency of CXCRhighPD-1highCD4+ T cells . These results prove again that macrophages from S . japonicum-infected mice have the similar ability as B cells and DCs to drive the generation of CD4+ T cells into Tfh cells . In parallel cultures , CD4+ T cells were separated from macrophages by a porous ( 0 . 4-µm ) membrane in otherwise identical conditions . Results showed that compared to the co-culture group , the separation of CD4+ T cells from macrophages increased only the expression of PD-1 , instead of CXCR5 ( Figures 5A and 5B ) . Moreover , a small fraction of CXCR5+CD4+ T cells sorted from S . japonicum-infected mice spleens also expressed the macrophage marker F4/80 , which suggested that these may represent stable macrophage-T cell conjugates ( Figure 5C ) . In addition , the mean FSC value of the cells which expressed both T cell and macrophage markers was approximately 500 , compared to T cells ( CD4+CXCR5+F4/80− ) and macrophages ( CD4−CXCR5−F4/80+ ) that were approximately 200 and 300 , respectively ( Figure 5D ) , which suggested again that they were macrophage-T cell conjugates . Data showed that almost all CXCR5+CD4+F4/80+ cells expressed the T cell antigen receptor marker CD3 , and further suggested that they were macrophage-T cell conjugates ( Figure 5D ) . Next , we sorted the CD3+CXCR5+CD4+F4/80+ cells to high purity and then treated the cells with EDTA to dissociate cell-cell contacts . The resulting single cells segregated into approximately equal numbers of CD4+ and F4/80+ single-positive populations ( Figure 5E ) , which further confirmed the stable conjugates . The macrophage-T cell conjugates in S . japonicum-infected mice livers were also detected ( Figure S5 ) . Taken together , these data suggest that direct cell-cell contact is required for Tfh cell development driven by macrophages , and also suggest a physiological role of these conjugates in vivo . We found that the expression of ICOSL on DCs from S . japonicum-infected mice was significantly increased ( Fig . S6 ) , which supported by the report that the expression of ICOSL on DCs is required for the first step of Tfh-cell induction , and is further required for T cell–B cell interactions for the maintenance of Bcl-6 expression [33] . However , it is unknown whether ICOSL is required for Tfh cell generation driven by macrophages . To address this issue , we detected the expression of ICOSL by macrophages and found that macrophages from S . japonicum-infected mice expressed higher amounts of ICOSL than macrophages from normal mice ( Figure 6A ) . We adoptively transferred WT macrophages , ICOSL KO macrophages , or WT B cells as control from S . japonicum-infected mice into S . japonicum-infected ICOSL KO mice . Consistent with published data [44]–[46] , our result showed that macrophages were still alive in recipient mice one week after transferring ( Figure S7A ) . Furthermore , our study demonstrated that some of the adoptively transferred-macrophages migrated into spleen , lymph nodes and liver in S . japonicum-infected recipient mice one week post-transfer ( Figure S7B ) . Seven days after transfer , we detected the higher percentage and absolute number of Tfh cells in the spleens , lymph nodes , and livers of mice receiving WT macrophages compared to mice receiving ICOSL KO macrophages ( Figures 6B and 6C ) , indicating that macrophages from infected mice sufficiently promote the Tfh cell formation in vivo . Finally , we cultured normal mice derived CD4+ T cells together with macrophages from S . japonicum-infected WT or ICOSL KO mice , with or without antigen . The results showed that consistent with the result in Figure 4 , CD4+ T cells increased the surface expression of CXCR5 and PD-1 when exposed to WT macrophages but not ICOSL KO macrophages ( Figures 6D and 6E ) . Thus , these data indicate that ICOSL expressed by macrophages is essential for Tfh cell induction driven by macrophages . In addition to the ICOSL expressed on DCs , CD40L–CD40 interactions between CD4+ T cells and B cells or DCs is also required for Tfh cell generation [26] . Considering the level of ICOSL on B cells is regulated by the noncanonical NF-κB pathway , which can be triggered by signals from a subset of tumor necrosis factor receptor ( TNFR ) family members , including B-cell-activating factor and CD40 [47] , we hypothesized that CD40/CD40L signaling regulates Tfh cell generation by regulation of ICOSL expression on macrophages . To this end , we examined the expression of CD40 by macrophages and found that macrophages from infected mice had a higher level of CD40 than macrophages from normal mice ( Figure 7A ) , Incubation of macrophages with agonist anti-CD40 antibody led to the potent induction of ICOSL expression in vitro ( Figure 7B ) . In addition , agonist anti-CD40 antibody treatment had increased expression of surface CXCR5 and PD-1 on CD4+ T cells when exposed to macrophages from normal or infected mice in the presence of SEA . However , after treatment with anti-ICOSL antibody to block ICOS-ICOSL signaling in co-culture system , macrophages could not induce the expression of surface CXCR5 and PD-1 on CD4+ T cells ( Figure 7C ) . Taken together , the data in figures 6 and 7 suggest that CD40L–CD40 signaling regulates ICOSL expression on macrophages for Tfh-cell generation .
The appropriate granulomatous lesions formation favors the survival of both host and parasite after S . japonicum infection . Thus , rational treatment of patients with schistosomiasis requires a better understanding of the mechanism of granuloma formation to prevent excessive granulomatous lesions . In this study , we reported a novel role for Tfh cells in the development of liver pathology in S . japonicum-infected mice and uncovered an unappreciated function of macrophages in Tfh induction . Pathology induced by S . japonicum infection in host tissues , especially in the liver , is predominantly caused by the immune responses in response to schistosome eggs . Through an unclear mechanism , CD4+ T cell response induced by egg antigens orchestrates the development of granulomatous lesions around individual eggs trapped in host liver , which are composed of macrophages , eosinophils , and CD4+ T cells [3] . In this study , we found that the percentages and absolute numbers of Tfh cells in the spleen , lymph nodes , and liver were significantly increased 8 weeks after S . japonicum infection . However , whether Tfh cells , as the other CD4+ T cell populations ( Th1 , Th2 , Th17 and Treg cells ) do , contribute to the liver pathology in mice with schistosomiasis remains to be explored . Our result showed that Tfh cells had a profound effect on the formation of liver pathology in S . japonicum-infected mice . Using ICOSL KO mice , in which the generation of Tfh cells has been proven to be defective , we demonstrated for the first time that Tfh cells participate in the formation of hepatic granuloma in mice infected with S . japonicum , although we still can not fully rule out the possibility that ICOSL also plays a possible role in the generation of the other effector cells and impacts the formation of hepatic granuloma . In addition , we observed significantly lower levels of parasite antigen specific IgG and IgG1 antibodies in S . japonicum infected ICOSL KO mice . It is widely accepted that antibodies produced by B cells are dispensable for granuloma formation in mice infected with S . mansoni or S . japonicum [48]–[50] . Thus , the typical function of Tfh cells with regard to providing help to B cells to produce antibodies may not be involved in granuloma formation . Of note , IL-4-producing CD4+ T cells are generally considered to be essential for granuloma formation during schistosome infection [3] , while IL-21 could promote Th2 responses and upregulate granuloma inflammation [51] , [52] . Previous reports in mice showed that Tfh cells were significantly induced during S . mansoni infection [19] and that most IL-4-producing CD4+ T cells in reactive lymph nodes produced in response to S . mansoni antigens are Tfh cells [20] . Studies also reported that Tfh cells produce IL-21 [9] , [53] . These data indicate a potential role and mechanism for Tfh cells in liver granuloma formation in schistosome infection via the production of IL-4 and IL-21 . In addition , our recent unpublished data suggests that production of CXCL12 by Tfh cells may also involve in the upregulation of the granuloma formation . We have found that most Tfh cells during S . japonicum infection produced a high level of CXCL12 ( unpublished data ) , which is the ligand for CXCR4 on eosinophils , the most predominant cell types within granulomas , and is required for eosinophil migration [54] . Consistent with this notion , eosinophils were dramatically decreased in the liver of ICOSL KO mice ( unpublished data ) , suggesting that Tfh cells may also regulate granuloma formation by recruiting eosinophils to the liver in mice infected with S . japonicum . However , the exact mechanism for the involvement of Tfh cells in the formation of granuloma during S . japonicum infection requires further clarification . It is widely accepted that the generation of Tfh cells from naïve precursors typically involves interactions with antigen-presenting cells such as DCs within lymphoid tissues including spleen and the draining lymph nodes [32] . In addition , studies demonstrate that Tfh cells , which express CXCR5 , can migrate in response to CXCL13 and relocate to the follicles of lymphoid tissues . Thus , a high percentage of Tfh cells have been detected in the spleen and lymph nodes in various mice models [8] , [17] , [55] , [56] . However , Tfh cells have also been detected in nonlymphoid tissues , such as peripheral blood , in patients with immune-active chronic hepatitis B or systemic lupus erythematosus [57]–[59] , in the liver of hepatitis C virus ( HCV ) -infected patients [60] and mouse models [61] . In addition to in the spleen and lymph node , Tfh cells were also found in the liver from S . japonicum-infected mice in our study . Although our results obtained from Tfh cells transfer experiment proved that Tfh cells in peripheral were recruited to the liver and promoted the formation of liver granulomas , the mechanism of their cellular origin and migration still needs further illustration . Macrophages play an important role in host immune responses against S . japonicum infection . In addition to their role as phagocytes , most of these macrophages ( 50–90% ) display Ia antigens , and express MHC and costimulatory molecules , acting as professional APCs by processing and presenting SEA to CD4+ T cells [36] . A large number of macrophages , DCs and B cells co-locate in the marginal zone of spleen or lymph nodes , and it is well known that after activation , the entry of APCs to the white pulp , in particular to the T-cell zone , is an important step in the initiation of CD4+ T cell response [62] . Although the roles of B cells and DCs in the generation of Tfh cells [21] , [26] , and the role of macrophages in the induction of Th1 [37] , Th2 [38] , Th17 [39] , [40] , and Treg cells [41]–[43] have been documented in previous reports , no data have been published to address whether macrophages are also involved in the generation of Tfh cells . In addition to the reported ability of macrophages to induce generation of Th1 , Th2 , Th17 and Treg cells [63] , [64] , our observations show that macrophages also have ability to aid the generation of Tfh cells during S . japonicum infection , which is at least partially supported by our in vitro co-culture experiments and the macrophages transfer experiment . However , our in vitro experiment can not rule out the proliferation of the Tfh cells stimulated by macrophages also contribute to the increase of the Tfh cells . Although increasing evidence supports the concept that the liver is a secondary lymphoid organ , acting as a site of T cell activation and differentiation [65] , whether Tfh cells can be induced by macrophages in liver is worthy of further research . Our results suggest that CD4+ T cell-macrophage conjugates are necessary for Tfh cell generation , and establish a physiological relevance for these conjugates in vivo . Thus , our data are consistent with a model in which B cell–T cell conjugates regulate Tfh-cell generation [17] , although the exact mechanism of T cell-macrophage conjugates remains to be explored . Unexpectedly , the separation of CD4+ T cells from macrophages considerably increased the expression of PD-1 , which has commonly been used as a marker for T cell exhaustion in both mouse and human infections [66]–[68] , suggesting that macrophage-CD4+ T cell conjugates may contribute to prevent the exhaustion of CD4+ T cells . However , the characterization of the potential CD4+ T cell exhaustion needs to be investigated . Up to now , the mechanisms of the induction of CXCR5 and PD-1 in CD4 T cells are still unclear . Previous studies have shown that ICOS–ICOSL engagement between B cells and CD4+ T cells activates PI3 kinase ( PI3K ) signaling and provides signals to CD4+ T cells for the initiation and maintenance of Tfh differentiation and expression of CXCR5 and PD-1 [33] , [56] , [69] . Our data consistently showed that Tfh cell generation driven by macrophages is also dependent on ICOSL signaling . Intriguingly , our results showed that the level of ICOSL expression on macrophages from normal mice was much lower than that from infected mice , providing a possible explanation as to why macrophages from infected mice , rather than those from normal mice , significantly induced Tfh cell generation both in vitro and in adoptive transfer experiment . Thus , macrophages may represent a novel subset of APCs that prime Tfh generation in mice infected with S . japonicum . CD40L signaling in CD4+ T cells is critical for T cell priming and maintenance in most in vivo contexts [70] , [71] . CD40L–CD40 interactions , with either B cells or DCs , are required for Tfh cell development [26] . However , the exact mechanism of CD40–CD40L signaling in the generation of Tfh cells has remained unclear . For the first time , our results suggest that the activation of CD40–CD40L signaling upregulates ICOSL expression on macrophages , and subsequently facilitates Tfh cell generation . In summary , except for Th1 , Th2 , Th17 , and Treg cells , we have demonstrated a novel role of Tfh cells in the granulomatous pathology in mice infected with S . japonicum , which challenges the existing paradigm that Tfh cells are specialized for providing B cell help . We have also shown that macrophages are able to promote the generation of Tfh cells in a cell-cell contact manner and regulated by CD40–CD40L signaling . In addition , our data indicate a therapeutic potential to target a Tfh cell induction or migration axis in liver pathology caused by S . japonicum infection .
Animal experiments were performed in strict accordance with the Regulations for the Administration of Affairs Concerning Experimental Animals ( 1988 . 11 . 1 ) , and all efforts were made to minimize suffering . All animal procedures were approved by the Institutional Animal Care and Use Committee ( IACUC ) of Nanjing Medical University for the use of laboratory animals ( Permit Number: NJMU 09-0163 ) . Eight-week-old male C57BL/6J mice and eGFP C57BL/6J mice were purchased from the SLAC Laboratory ( Shanghai , China ) and Model Animal Research Center of Nanjing University ( Nanjing , China ) , respectively . Eight-week-old male ICOSL−/− C57BL/6J mice were obtained from Soochow University ( Suzhou , China ) . Animals were kept under specific pathogen-free conditions and were used at 8–12 weeks of age . In the infection experiments , mice were infected percutaneously with 12 S . japonicum cercariae ( Chinese mainland strain ) obtained from infected Oncomelania hupensis snails purchased from the Jiangsu Institute of Parasitic Diseases ( Wuxi , China ) . SEA was obtained from purified and homogenized S . japonicum eggs . The protein concentration of SEA was determined using a bicinchoninic acid ( BCA ) Protein Assay kit ( Bio-rad , Richmond , CA ) . Macrophage-T cell doublets were sorted with a FACSAria cell sorter ( BD Biosciences ) . Conjugates were dissociated with 2 mM EDTA and vigorous vortexing . Single cell suspensions were prepared by teasing spleens , inguinal and mesenteric lymph nodes ( LN ) and blood in PBS containing 1% EDTA followed by red blood cell ( RBC ) lysis . Hepatic lymphocytes were prepared as described previously with some modifications [72]–[75] . In brief , for preparation of single cell suspensions of hepatic lymphocytes , infected or control mice livers were perfused via the portal vein with PBS . The excised liver was cut into small pieces and incubated in 10 ml of digestion buffer ( collagenase IV/dispase mix , Invitrogen Life Technologies , Carlsbad , CA ) for 30 min at 37°C . The digested liver tissue was then homogenized using a Medimachine with 50-µm Medicons ( Becton Dickinson , San Jose , CA ) according to the manufacturer's instructions . The liver suspension was resuspended in 5 ml PBS and then placed on a lympholyte M ( Cedarlane , Ontario , Canada ) overlay in a 1∶1 ratio . Cells were spun at 2 , 200 rpm for 20 minutes , collected from PBS/Lympholyte M interface , washed and suspended in PBS containing 1% EDTA . For surface staining , 2×106 cells per 100 µl were incubated for 30 min at 4°C with the following fluorescently labeled monoclonal antibodies: CD3e-Percp-cy5 . 5 ( eBioscience , San Diego , CA ) , CD3e-PE ( eBioscience ) , CD4-FITC ( eBioscience ) , CD4-PE-Cy7 ( eBioscience ) , CXCR5-APC ( BD Pharmingen , San Diego , CA ) , PD-1-PE ( eBioscience ) , ICOS-PE-Cy5 ( eBioscience ) , F4/80-FITC ( eBioscience ) , F4/80-PE ( eBioscience ) , CD40-PE ( eBioscience ) , CD44-Percp-cy5 . 5 ( eBioscience ) , ST2-APC ( eBioscience ) , and ICOSL-PE ( eBioscience ) . After staining of surface markers , the cells were permeabilized with cold Fix/Perm Buffer , and Fc receptors of cells were blocked with anti-mouse CD16/32 ( eBioscience ) for 15 min . The Bcl6-Alexa Fluor 488 ( BD Pharmingen ) was then added and incubated for 30 min at 4°C . The cells were then washed twice in wash buffer before analysis . The details of measurement of Th1 , Th2 , Th17 , and Treg cells by flow cytometry are provided in Text S1 . Peritoneal macrophages ( PM ) were prepared as described previously with some modifications [76] . Briefly , mice were sacrificed by cervical dislocation , and 7 ml of ice-cold PBS containing 1% fatal bovine serum ( FBS ) and 50 µg/ml penicillin-streptomycin ( Sigma-Aldrich , St . Louis , MO ) was injected into the abdominal cavity . The medium containing peritoneal exudates cells ( PEC ) was recollected and transferred to sterile plastic tubes . The suspended cells were centrifuged at 1500 rpm for 5 min at 4°C , and the cells were resuspended in PBS with 1% FBS . The cells ( 2×106 cells/well ) were then seeded into 12-well culture plates ( Costar , Cambridge , MA ) and were allowed to adhere for 2–3 h at 37°C with 5% CO2 . Non-adherent cells were removed by washing the wells with sterile PBS six times , and the remaining monolayers were all PM . Cell viability was ≥90% in all experiments . PM were either directly subjected to flow cytometry analysis or stimulated in vitro by anti-mouse CD40 antibody ( 500 ng/ml; Biolegend , San Diego , CA ) for 3 days . CD4+ T cells , B cells , or DCs were MACS purified from the spleens of normal mice or infected mice using a CD4+ T cell negative-isolation kit ( Miltenyi Biotec , Auburn , CA ) , CD45R ( B220 ) MicroBeads ( Miltenyi Biotec ) , or CD11c MicroBeads ( Miltenyi Biotec ) , respectively . Purified CD4+ T cells ( 2×106 cells/well ) from normal mice were incubated in triplicate wells with PM ( 2×105 cells/well ) , B cells ( 2×105cells/well ) , or DCs ( 2×105cells/well ) from normal or S . japonicum-infected mice for 3 days with or without SEA ( 20 µ/ml ) in the presence or absence of agonistic anti-CD40 antibody ( 500 ng/ml; Biolegend ) . Anti-ICOSL antibody ( 5 µ/ml; eBioscience ) or isotype-matched control antibody ( 5 µ/ml; eBioscience ) were used to block ICOS-ICOSL signaling in co-culture system . Then , CD4+ T cells were collected and examined using surface staining . All experiments were repeated once . Spleen and mesenteric LN cells from eGFP C57BL/6J or WT mice 8 weeks after infection with S . japonicum were pooled , and CD4+ T cells were presorted by using a CD4+ T cell negative-isolation kit ( Miltenyi Biotec ) . CD4+ T cells were stained with CXCR5-APC and PD-1-PE antibodies , or with CD44-Percp-cy5 . 5 and ST2-PE antibodies . CXCR5highPD-1high Tfh cells , Tfh depleted ( non-Tfh control ) cells or CD44+ST2+CD4+ T cells ( Th2 ) [77] , [78] were FACS purified by using a FACSAria cell sorter ( BD Biosciences ) to investigate the role of Tfh cells in liver pathology . In addition , eGFP+CD4+ T cells were stained with CXCR5-APC , PD-1-PE , and CD44-Percp-cy5 . 5 antibodies . CXCR5highPD-1high Tfh cells and CD44+CXCR5lowPD-1low ( non-Tfh control ) cells were FACS purified by using a FACSAria cell sorter to investigate the recruitment of Tfh cells into liver granuloma . Sorted cells were more than 97% pure . FACS-sorted Tfh , non-Tfh control cells or Th2 cells were resuspended in PBS and injected intraperitoneally ( ip ) into the ICOSL−/− mice 5 weeks after S . japonicum infection ( 3×106 cells/mouse ) . A group of mice that did not receive T cells but PBS was used as an additional control ( mock transfer ) . Mice were sacrificed 3 days after transfer to investigate the recruitment of Tfh cells into liver granuloma , or 3 weeks after transfer to investigate the role of Tfh cells in liver pathology , respectively . The details of measurement of Tfh-cell plasticity are provided in Text S1 . PMs were prepared as described above and B cells were sorted using MACS ( Miltenyi Biotec ) . PMs or B cells were resuspended in PBS and injected ip into the mice 8 weeks after S . japonicum infection ( 4×106 cells/mouse ) . Mice were sacrificed 7 days after transfer to investigate the generation of Tfh cells by macrophages in vivo . The details of measurement of survival and migration of macrophages are provided in Text S1 . Livers were fixed in 10% neutral buffered formalin . Paraffin-embedded sections ( 4 µm ) were dewaxed and stained with hematoxylin and eosin ( H&E ) for granulomas analysis or sirius red ( Sigma ) for fibrosis analysis . For each mouse , the sizes of 30 granulomas around single eggs were quantified using AxioVision Rel 4 . 7 ( Carl Zeiss GmbH , Jena , Germany ) . Data are expressed in area units . All images were captured at 100× magnification using an Axiovert 200M microscope ( Carl Zeiss ) , and granulomas were analyzed using Axiovision software ( Carl Zeiss ) . Moreover , fibrosis was determined histologically by measuring the intensity of fibrosis in six random ( 100× ) digital images captured from collagen-specific sirius red-stained slides of each mouse using Image-Pro Plus software as previously described . The mean optical density of collagen was determined by dividing integral optical density by the image area [79] . To determine hepatocyte damage , levels of serum alanine transaminase ( ALT ) and aspartate aminotransferase ( AST ) were assayed using an Olympus AU2700 Chemical Analyzer ( Olympus , Tokyo , Japan ) . Liver was harvested from infected animals and immediately frozen in optimal cutting temperature ( OCT ) embedding compound over liquid nitrogen . Frozen livers were cut into 8-µm sections on a Leica cryostat and fixed in a mixture of ice-cold 75% acetone/25% ethanol for 5 min . All images were captured using an Axiovert 200M microscope ( Carl Zeiss , Inc . ) and analyzed using Axiovision software . Total RNA was extracted from cells using an RNeasy Mini Kit . Complementary DNA ( cDNA ) synthesis was performed using SuperScript RT II and oligo ( dT ) . SYBR Green-based RT-PCR was performed with FastStart Universal SYBR Green Master ( Rox ) reagents and an ABI PRISM 7300 . Relative expression was calculated using the 2−ΔΔCt method normalized to hypoxanthine-guanine phosphoribosyltransferase ( HPRT ) . The following primers were used: GAPDH: 5′-ggtgaaggtcggtgtgaacg-3′ and 5′-accatgtagttgaggtcaatgaagg-3′; Bcl-6: 5′-agacgcacagtgacaaacca-3′ and 5′-cgctccacaaatgttacagc-3′ . Data were analyzed by a two-tailed Student's t-test using SPSS 11 . 0 software ( IBM ) . The significance of the difference between the treatment groups was identified using a t-test . P-values of less than 0 . 05 were considered as statistically significant . | Schistosomiasis is a chronic helminthic disease that affects approximately 200 million people . After S . japonicum infection , parasite eggs are trapped in host liver and granulomas are induced to form around eggs . Severe granuloma subsequently results in serious liver fibrosis and circulatory impairment chronically . It is important to fully elucidate the mechanism of the granuloma formation . Here , we show that Tfh cells play a novel role of promoting the hepatic granuloma formation and liver injury , and identified a novel function of macrophages in Tfh cells induction in S . japonicum-infected mouse model . In addition , we show that the generation of Tfh cells driven by macrophages is cell–cell contact dependent and regulated by CD40-CD40L signaling . Our findings revealed a novel role and mechanism of macrophages in Tfh cell generation and the liver pathogenesis in S . japonicum-infected mouse model . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"infectious",
"diseases",
"medicine",
"and",
"health",
"sciences",
"veterinary",
"diseases",
"zoonoses",
"neglected",
"tropical",
"diseases",
"biology",
"and",
"life",
"sciences",
"tropical",
"diseases",
"parasitic",
"diseases",
"veterinary",
"science"
] | 2014 | Follicular Helper T Cells Promote Liver Pathology in Mice during Schistosoma japonicum Infection |
In chronic infection , Mycobacterium tuberculosis bacilli are thought to enter a metabolic program that provides sufficient energy for maintenance of the protonmotive force , but is insufficient to meet the demands of cellular growth . We sought to understand this metabolic downshift genetically by targeting succinate dehydrogenase , the enzyme which couples the growth processes controlled by the TCA cycle with the energy production resulting from the electron transport chain . M . tuberculosis contains two operons which are predicted to encode succinate dehydrogenase enzymes ( sdh-1 and sdh-2 ) ; we found that deletion of Sdh1 contributes to an inability to survive long term stationary phase . Stable isotope labeling and mass spectrometry revealed that Sdh1 functions as a succinate dehydrogenase during aerobic growth , and that Sdh2 is dispensable for this catalysis , but partially overlapping activities ensure that the loss of one enzyme can incompletely compensate for loss of the other . Deletion of Sdh1 disturbs the rate of respiration via the mycobacterial electron transport chain , resulting in an increased proportion of reduced electron carrier ( menaquinol ) which leads to increased oxygen consumption . The loss of respiratory control leads to an inability to recover from stationary phase . We propose a model in which succinate dehydrogenase is a governor of cellular respiration in the adaptation to low oxygen environments .
The World Health Organization has estimated the prevalence of Tuberculosis ( TB ) in the human population to be nearly two billion people . Although only a fraction of those individuals will ever display symptoms , TB is still a significant cause of worldwide mortality and was responsible for 1 . 3 million deaths in 2012 [1] . The organism responsible for this disease , Mycobacterium tuberculosis , owes its unqualified success as a pathogen to the ability to survive and persist in a human host , where it can evade immune surveillance and establish a sub-clinical infection . These latently infecting bacilli have the potential for reactivation in certain circumstances , as is commonly seen in HIV-induced immunosuppression [2] . In addition to immune evasion mechanisms found in some other chronic pathogens , M . tuberculosis appears to evade immunity by adopting a metabolically active but quiescent state during which cell division is limited [3] . In fact , the current antibiotic therapy regimen recommended by the WHO is multiphasic and is modeled around the presence of tolerant persister cells that are not cleared in the initial two months of treatment . The reliable occurrence of this subpopulation in clinical investigations has led to the addition of a continuation phase to the antibiotic course , which can last four months or more . Currently , the physiological adaptation which enables this organism to persist remains an area of active research , but targeting persisters should considerably improve the outcome of therapeutic efforts . The inability to physically isolate a persister subpopulation without perturbing its labile state has prompted the adoption of a number of approaches to gain insight into the basis of the phenomenon . These models , which recapitulate a slowly- or non-dividing state in vitro , have revealed a number of interesting clues to persister physiology . It is important to note that M . tuberculosis is widely considered to be an obligate aerobe with the important stipulation that even though division is limited in anaerobic conditions , bacterial cultures can remain viable for decades [4] . A model developed by Wayne was instrumental in delineating the oxygen set points which result in cessation of division below 1% dissolved oxygen ( DO ) , and a decline in survival below 0 . 06% DO , thus providing a framework to understand dormancy [5] . More recently , Gengenbacher et al . found that when quiescence is initiated through nutrient starvation the bioenergetic remodeling results in a decrease of ATP to one-fifth its log phase level , a concentration which apparently is reflective of maintenance of the protonmotive force [6] . Watanabe et al . subsequently verified these results and further noted that the depletion of ATP correlated with an apparent dearth of NAD+ , at very low dilution rates in continuous culture [7] . The information gleaned from these studies directly informs the mechanistic descriptions of new TB drugs , including the diarylquinolone , Bedaquiline , a newly approved ATP-synthase inhibitor which is effective against dormant mycobacteria [8] . Although a number of studies have examined the transcriptional response of dormant cells , direct genetic evidence of metabolic genes essential for growth rate transitions was reported from studies of the abundance of specific mutants in transposon insertion libraries following alteration of the dilution rate in continuous culture . Among enzymes with a bioenergetic function , genes involved in energy metabolism ( a putative succinate dehydrogenase ) , and a number of oxidoreductases were found to be important for this transition suggesting that the resumption of growth requires the benefits of oxidative phosphorylation [9] . It is difficult to point to a specific physiological adaptation which would be responsible for survival without knowledge of which of the diverse in vivo microenvironments might harbor persistent mycobacteria , but some groups have tried screening approaches aimed at addressing these questions in specific tissues [10] , [11] . In terms of bioenergetic capacity , these studies revealed that one member of an operon containing a putative succinate dehydrogenase appeared to be essential for in vivo mycobacterial survival in a mouse model during the chronic phase of infection , a finding that was subsequently repeated using an analogous transposon-based screen [10] , [12] . Oxidative flux through the TCA cycle is directly coupled to the electron transport chain via the oxidation of succinate and the corresponding reduction of membrane-localized quinones . Disruption of this activity would be a good strategy for control of growth in the energy limiting conditions that M . tuberculosis is thought to encounter in vivo [13] . This is an important consideration because ATP generation by oxidative phosphorylation is energetically much more efficient than ATP generation by substrate-level phosphorylation . M . tuberculosis has two operons annotated as succinate: quinone oxidoreductase - as well as a putative quinol:fumarate oxidoreductase which should be capable of succinate oxidation ( see Annotation in Text S1 ) . To date , the functional activity of these complexes has not been investigated , so we sought to understand their role in the transition from aerobic growth to persistence in M . tuberculosis . To this end , we targeted the two operons with homology to succinate dehydrogenase , which are encoded by Rv0247c-Rv0249c and Rv3316–Rv3319 ( Figure 1 ) ( or according to the convention of Berney et . al . as sdh1 and sdh2 , respectively ) [14] , for further study . In this work , we employed a combination of genetic , physiological and biochemical approaches to dissect the roles of Sdh1 and Sdh2 in the metabolic shiftdown of M . tuberculosis during adaptation to hypoxia . We report that Sdh1 ( and not Sdh2 ) is the primary aerobic succinate dehydrogenase of M . tuberculosis . Deletion of this enzyme resulted in a number of bioenergetic deficiencies such as a major deficit in viability during stationary phase or during the chronic phase of infection in C3HeB/FeJ mice . The cause of this energetic insolvency was a peculiar mismanagement of oxygen consumption due to an imbalance in the redox state of the menaquinone pool . The Δsdh1 mutant consumed oxygen with close to perfect uncoupled kinetics , whereas wild type ( wt ) M . tuberculosis enacted an oxygen conservation strategy . The respiratory rate was dependent on the redox state of the menaquinone pool and respiration could be stimulated in non-respiring cells by adding exogenous reductant .
To determine the role of each enzyme complex , we prepared strains with null deletions of each in attenuated ( mc26230 ) and virulent ( H37Rv ) strains of M . tuberculosis using specialized transduction ( Table S1 and Complementation in Text S1 ) [15] , [16] . For safety reasons , we relied on null mutants of attenuated strains for all assays in which virulence was not a primary focus . The resulting mutant strains displayed no observable differences in growth rate in media containing glucose or glycerol as a primary carbon source ( Figure 2A , B ) , however we observed a growth defect for Δsdh1 when succinate was the sole available carbon source compared to the parent or Δsdh2 ( Figure 2C ) . These results were consistent for virulent and attenuated strains . In addition , we observed a stationary phase exit defect in which Δsdh1 was unable to be rescued from two-month old cultures , and the sdh2 mutant grew poorly after a similar period ( Figure 2D ) . The parental cultures or complemented strains exhibited no comparable decrease in growth rate or saturation even after eight months of stationary phase , indicating that these operons do not have perfectly redundant catalytic activities in vitro . Succinate dehydrogenase catalyzes the two-electron oxidation of succinate to fumarate with a corresponding reduction of quinone to quinol , but physiologically , the succinate oxidation:fumarate reduction catalytic ratios are dependent on substrate concentrations and are critically dependent on the redox potential [17] , [18] . Absolute pool sizes of metabolic intermediates are highly dynamic in living cells as a function of growth stage , pH , gas mixture , and temperature . As a result , the predominant direction of catalysis for each enzyme at any time cannot be inferred by annotation alone . In fact , the SDH reaction in mycobacteria should have an unfavorable free energy because the redox potential of menaquinone is lower than that of the succinate to fumarate reaction [19] . We evaluated gene function of the two sdh operons in a physiologically relevant context using a targeted metabolomic approach by analyzing differences in pool sizes of central carbon metabolites for cells in aerobic growth and in an anaerobic model [20] . Comparison of the mutant strains to the parental strain during aerobic growth revealed a significant 4-fold increase in intracellular succinate in Δsdh1 but no difference in Δsdh2 . This was accompanied by a 0 . 5-fold decrease of malate concentration in Δsdh1 compared to the parental strain , whereas the Δsdh2 strain showed no difference; these data suggest a loss of succinate dehydrogenase activity in the Δsdh1 strain ( Figure 3 ) . Consistent with observations made by others [21] , we detected an accumulation in the total intracellular succinate concentration of the parental strain of M . tuberculosis of 8-fold after 10 days of anaerobiosis , while the concentration in Δsdh1 increases only 1 . 5-fold during this span . Conversely , total malate concentration rises slightly in the wt strain ( 1 . 7-fold ) , while the Δsdh1 mutant shows a 7-fold increase . The accumulation of intracellular succinate is suggestive of an inability of this strain to perform succinate oxidation , but since total concentrations of α-ketoglutarate decrease , and glyoxylate , oxaloacetate and malate increase in hypoxia , a portion of this succinate is likely to be from the reported activity of isocitrate lyase [21] . Consistent with this , during hypoxia we observed significantly less accumulated succinate in the Δsdh1 mutant relative to the parent ( whereas Δsdh2 had an intracellular succinate concentration higher than the parent ) and malate concentrations were 2 . 2 ( for Δsdh1 ) and 1 . 8-fold ( for Δsdh2 ) increased , though these differences were not significant . We next verified that the aerobic accumulation of succinate in the attenuated M . tuberculosis mutants was reflective of the condition in the virulent strain using the same method . During aerobic batch culture , H37RvΔsdh1 and H37RvΔsdh2 accumulated succinate in excess of the parental H37Rv strain , and this accumulation was corrected for in the complementing strain ( mc27292 ) which constitutively expresses sdh1 . This behavior is consistent with the complementation in the attenuated strains ( see Figure S1 , and Complementation in Text S1 ) . Based on these differences in metabolite pools , we analyzed the predominant direction of catalysis in the same aerobic and anaerobic models using stable isotope labeling ( see Metabolomics in Text S1 ) . Cells were grown in 7H9 medium supplemented with 10% OADC and labeled with [1 , 4-13C] aspartate in both four days of aerobic log phase growth and after twelve days in hypoxia using methods similar to those previously described [7] . We traced the fate of isotopically labeled carbon in TCA intermediates during aerobic growth ( Figure S2A ) and in hypoxia ( Figure S2B ) and determined the proportion of each labeled metabolite with respect to all isotopologues for each intermediate . The stable isotope labeling supported the classification of Sdh1 as an aerobic succinate dehydrogenase , but little difference in metabolite ratios was observed in strains lacking Sdh2 in these conditions . We conclude from the metabolomic data that a functional reassignment should be considered for the operon encoded by Rv0247c-Rv0249c . We propose that Rv0247c-Rv0249c ( Sdh1 ) encodes the primary succinate dehydrogenase of M . tuberculosis and the operon encoded by sdhCDAB ( Sdh2 ) performs catalysis in an as yet undefined condition . To seek further support of this proposed classification , we analyzed gene expression of mutant strains in aerobic and hypoxic conditions ( Table S2 , and Methods in Text S1 ) . Although no significant upregulation by the opposing sdh gene cluster was observed during aerobic growth , sdh1 is significantly upregulated in Δsdh2 during anaerobiosis . Genes in the sdh2 operon were not upregulated in Δsdh1 at either oxygen tension . This scheme is consistent with transcriptional data from oxygen-limited M . smegmatis that shows a 2-fold increase in sdh2 transcripts but a 30-fold decrease of sdh1 transcripts [14] . As preservation of a proton motive force ( PMF ) is an important component of anaerobic survival , we monitored CFUs of sdh mutant strains in aerobic and anaerobic conditions in the presence of sub-lethal concentrations of the protononophore carbonyl cyanide m-chlorophenyl hydrazone ( CCCP ) . Whereas 10 µM CCCP had a bacteriostatic effect on normoxic cultures ( Figure S3A ) , the same concentration of CCCP resulted in a loss of viability of greater than 3-logs at 35 days of treatment ( Figure S3B ) . Both mutant strains were more susceptible to PMF inhibition than the parental strain , but we were unable to recover colonies from Δsdh2 cultures after 21 days . This data supports the conclusion that Sdh2 is the generator of the PMF in hypoxia , as we have previously observed in M . smegmatis [22] . Next , we assessed the contribution of each sdh mutant to aerobic respiration in bioreactors operating in batch mode and in continuous culture . Cells were inoculated into a bioreactor system in which DO concentration , optical density , midpoint redox potential and pH could be measured simultaneously and were monitored throughout the growth curve as oxygen was depleted by the organism . Surprisingly , the parental strain initiated down-modulation of its respiration rate at ∼40% DO , while Δsdh1 continued to respire unabated until the DO was entirely depleted ( Figure 4A ) . Conversely , cells harboring a deletion of sdh2 consumed oxygen at a reduced rate and were able to modulate respiration as DO was depleted to ∼6% . These experiments revealed that constitutive overexpression of the complemented strain using the hsp60 promoter overcompensated the respiratory phenotype . In subsequent experiments , we found that the oxygen consumption curve could be complemented in a strain expressing sdh1 with a novel integrated tet-responsive promoter ( Ptet2 ) ; at low levels of anhydrotetracycline inducer ( 25 ng/mL – see Figure S4A , Complementation in Text S1 ) . We concluded that these induction levels reflect the concentration of active enzyme in each condition; therefore , perfect complementation would require levels of expression which closely match wt levels throughout the growth curve . The increased oxygen consumption of Δsdh1 should result in an increased membrane potential and an increased growth rate . However , data collected during batch culture experiments revealed that the initial growth rate for Δsdh1 is actually slightly slower in aerobic conditions ( Table 1 ) ; this rate decreases considerably once oxygen is depleted , yet the parental strain maintains faster population doubling times than either sdh mutant ( Table 1 , Figure S5 ) . This apparent uncoupling of respiration from growth was further analyzed in a separate chemostat experiment . When cultures were grown with a 24 hr doubling time , Δsdh1 ( but not the parental strain ) was unable to maintain the growth rate at DO levels of 25% , 5% , or 1% , and consequently washed out of continuous culture ( Figure 4B ) . Because the membrane potential ( ΔΨ ) is the major component of the PMF at neutral pH values , we assessed the membrane potential by measuring uptake of the lipophilic cation tetraphenylphosphonium ( TPP+ ) [23] . In aerobiosis the ΔΨ was comparable between the wt and Δsdh mutants ( 53–66 mV ) ( Table 2 ) . In hypoxia , the ΔΨ was considerably higher ( 30 vs . 18 mV ) for the Δsdh1 mutant compared to the wt and Δsdh2 strains ( Table 2 ) . Mycobacteria use menaquinone as their main electron carrier in the electron transport chain . Coupling succinate oxidation ( E°′ ∼ +30 mV ) to menaquinone reduction ( E°′ ∼ −80 mV at pH 7 ) is an energetic challenge , because this reaction is endergonic [24] . A model to explain this conundrum posits that reversed electron transport across the cytoplasmic membrane can provide the energy required to drive the oxidation of succinate using the PMF [25] . This suggests that the increased respiration rate in the Δsdh1 strain is due to the absence of reverse electron flow and consequently an altered redox state of the quinone pool . To confirm the hypothesis that the respiratory rate is a function of the redox balance of the menaquinone/menaquinol pool we sought corroborating evidence using M . tuberculosis strains harboring deletions of the type II NADH dehydrogenases ndh and ndhA . These enzymes are thought to be the primary means of electron input in Mycobacteria [26] during aerobic growth . A down-modulation of oxygen consumption by these strains occurred at ∼50% and ∼10% , respectively ( see Figure 4A ) . Complementation of ΔndhA was similar to that of the sdh enzymes with overcomplementation of oxygen consumption when ndhA was expressed episomally using a constitutive promoter ( Figure S4B ) , further illustrating the necessity of “fine tuning” respiratory enzyme levels to achieve maximal growth . This finding supports the paradigm that enzyme activities that facilitate rapid reduction of the quinone pool serve to increase the respiratory rate in the wt strain ( since their deletion reduces oxygen consumption ) , and fumarate reduction functions as a respiratory brake during aerobiosis by an opposing oxidation of the pool . The unexpected disparity in DO-sensitive modulation of respiration by the type II NADH dehydrogenases suggests a wider strategy to indirectly sense oxygen concentrations in the immediate environment and spend reducing equivalents accordingly before taking the drastic step of uncoupling biomass production from respiration . The apparent diminution of succinate oxidation in Δsdh1 during aerobiosis , and its uncontrolled respiratory phenotype alluded to an imbalance in the redox state of the menaquinone pool . We sought to confirm this biochemically by extracting menaquinones ( MK-9 ) from cells growing aerobically , and at 1% DO in bioreactors ( see Methods ) . Ratios of menaquinol:menaquinone of the parental strain were balanced when grown aerobically , but heavily skewed toward the oxidized state at low DO , conversely Δsdh1 had higher concentrations of menaquinol ( reduced form ) , which was sufficient to drive respiration even at low oxygen levels ( Figure 5A ) . In aerobically growing cells , we found the quinone pool to be balanced ( ratioMK-9red/MK-9oxid = 0 . 87 ) , indicating equilibrium between respiratory rate and carbon flux . Because the balance of quinone reduction can shift rapidly , we sought further confirmation by monitoring data from a probe for midpoint redox potential . Cultures were grown in a bioreactor running in batch mode as described above but flowed compressed air into the bioreactor at 1L/hr . Using this measure , M . tuberculosis can be seen to utilize available oxygen then switch off respiration until oxygen builds up to a threshold concentration before switching on aerobic respiration again . Importantly , increases in the redox potential precede the onset of oxygen consumption by several minutes during which pH does not change ( Figure S6 and S7 ) ; supporting the hypothesis that oxygen consumption is managed by quinone redox balance . Δsdh2 behaves in a manner similar to wt , but Δsdh1 appears to maintain a negative midpoint redox potential and respires all available dissolved oxygen without allowing it to build up in the vessel ( Figure S8 ) . The above behavior is consistent with previous reports that respiratory rate can be directly controlled with first-order kinetics by the degree of reduction of the quinone pool in membrane vesicles and mitochondria [27] , [28] . We took advantage of the relatively low midpoint redox potential of menaquinone [29] , and sought evidence that the respiratory rate of intact mycobacterial cells could be stimulated using the membrane permeable reducing agent dithiothreitol ( DTT ) . We hypothesized that cells which have entered the phase of respiratory downshift brought on by low oxygen concentrations should be stimulated to respire if menaquinol can be replenished by an exogenously applied reducing agent . To test this , M . tuberculosis strain mc26230 was grown to early stationary phase and oxygen consumption was monitored in a Clark-type oxygen electrode ( see supplementary methods ) with and without the addition of DTT ( Figure 5B ) . Stimulation of oxygen consumption was observed up to concentrations of 40 mM reductant , after which little increase was observed . Importantly , no stimulation of oxygen consumption was noted in media alone or in preparations of heat-killed cells from the same culture . No synergistic increase in oxygen consumption was observed in log phase cells in similar experimental conditions when the starting DO of culture media was greater than 50% that of aerated media , i . e . cells that are already respiring at maximal rates are not induced to respire faster by the addition of reductant . To assess the effect of disregulated respiratory activity on pathogenesis and persistence , we tested the ability of Mtb Δsdh1 and Δsdh2 to cause disease in several established murine models . Previous experiments utilizing a high-throughput genetic screen have revealed subunits of sdh1 ( but not sdh2 ) to be underrepresented in the lungs of C57Bl/6J mice during chronic infection [10] , [30] . It is not clear if the fitness defect observed in those screens is the result of reduced virulence or an inability of the mutants to maintain their numbers during chronic infection , but we were unable to recreate this phenotype with null deletion strains using the C57Bl/6J mice ( Figure S9 ) . To assess virulence , we infected immunodeficient Rag-1−/− mice via low-dose aerosol; these mice produce no mature T or B cells and are thus unable to control mycobacterial infection [31] . Whereas immunodeficient mice infected with H37Rv had a median survival time of 26 days , Δsdh1 infected mice had a slightly longer median survival time of 29 days . Interestingly , the Δsdh2 strain displayed a hypervirulent phenotype; these mice had a median survival time of only 22 days ( Figure 6A ) . The overexpressing complemented strains for both of these deletions were less lethal than either the mutants or the parental strain . Given the predictive constraints of the mouse model in TB infection , particularly the inability of the murine immune system to form fibrous caseous granulomas [32] , we think that any impact of these mutations on survival ( or strains harboring deletions in respiratory enzymes ) could be lessened because oxygen levels are likely always sufficient for growth in the murine lung . A murine model was developed to address this limitation; the C3HeB/FeJ mouse is an inbred strain that develops fibrous encapsulated lung lesions post-aerosol infection which appear to contain hypoxic centers [33] . We reasoned that the respiratory mismanagement of Δsdh1 would lead to a survival deficit in the lesions of mice containing hypoxic lesions . To test this hypothesis , we infected C3HeB/FeJ mice via aerosol and monitored burden over time ( Figure 6B ) . By twenty weeks of infection , Δsdh1 had tenfold fewer cells per lung than the H37Rv parent ( 5 . 79 log10 CFU vs . 6 . 67 log10 CFU ) and Δsdh2 was similar to the wt . It is important to note that after nine weeks the Δsdh1 burden dropped slightly , while wt cells continued dividing until week twenty . This suggests that deletion of Sdh1 leads to an inability to maintain bacterial numbers in the host , however , the difference in bacterial burden between wt and Δsdh1 was not as dramatic as we would have expected based on our in vitro results . This might be explained by the fact that gross pathology of upper lungs at twenty weeks did not reveal encapsulated granulomatous ) lesions ( Figure S10 ) , thus oxygen was likely not restricted in the lungs of these mice .
The bioenergetic program that sustains M . tuberculosis during latency and in models that recapitulate persistence is of great interest because this survival is likely due to inhibition of growth that stems from an idle metabolic state [3] , [34] . A mechanistic understanding of quiescence is of crucial importance to the planning of new antitubercular compound screens , which can be designed to directly target this population . To this end , we sought to understand the function of the enzyme responsible for the direct coupling of anabolism via the TCA cycle and the electron transport chain - succinate dehydrogenase . Prior to this work , the individual roles of the two predicted succinate dehydrogenases of M . tuberculosis had not yet been experimentally determined , and no obvious phenotype was reported in M . tuberculosis H37Rv containing a null deletion of the hypoxia-upregulated fumarate reductase , frdABCD [7] . Genetic manipulation of M . tuberculosis followed by an intracellular metabolomic approach allowed us to probe the functions of the two annotated Sdh enzymes and their role in cell physiology . Importantly , these enzymes were found to strongly influence aerobic respiration , and deletion of sdh1 resulted in an increased rate of respiration , but did not result in faster cell growth . The work presented here validates the predicted role of sdh1 as the primary succinate dehydrogenase during aerobiosis It has been almost eighty years since Loebel and colleagues formally noted the capacity of M . tuberculosis for curtailing its oxygen consumption under anaerobic or starvation conditions [35] , but a mechanism for this phenomenon is absent from the literature . Two distinct phases of adaptation to decreasing oxygen tension have been described; NRP ( non-replicating persistence ) stage 1 - marked by the cessation of cell division at ∼1% oxygen , and NRP stage 2 – a quiescent state occurring below 0 . 06% oxygen in which biomass production ceases [36] . Our data imply that M . tuberculosis employs an orchestrated respiratory slowdown as oxygen levels fall; this program is initiated while oxygen is still plentiful . The respiratory rate is fine-tuned by the opposing activity of the succinate dehydrogenase and fumarate reductase activities to maintain an optimal growth rate . This suggests that this tuning is controlled by balancing substrate concentrations , as has been suggested in electrochemical analysis of isolated enzymes [37] , post-translationally [38] , and via catabolite repression [39] . Management of respiration has important consequences for the proclivity for survival of M . tuberculosis amid a range of pathological niches in which oxygen tension can vary significantly , because ATP generation is much more efficient when electrons are committed to oxidative phosphorylation than through substrate-level phosphorylation alone . We favor a simple mechanistic explanation for the controlled respiratory slowdown that is consistent with structural studies of the terminal cytochrome c oxidase complex and the progression of the Q-cycle ( Figure 7 ) [40]–[42] . Organisms will respire at optimal rates with a balanced quinone pool in which quinol ( reduced ) is present in sufficient concentration to immediately occupy the center P of the cytochrome oxidase complex; but when quinol is limiting - in an oxidatively skewed pool - respiration will progress at a less-than optimal rate . Figure 5A shows that whereas the wt strain has a largely oxidized quinone pool at 1% DO , the Δsdh1 mutant maintains a balanced pool , resulting in unchecked oxygen consumption . These data support a mechanism for respiratory downshift in wt M . tuberculosis that works as follows: as oxygen concentration drops below 40–30% , succinate oxidation also decreases leading to its buildup ( hypoxic cells accumulate a sevenfold increase in intracellular concentration ) . This ‘unrespired’ succinate does not contribute to the reduction of membrane menaquinones , and as the ratio of menaquinol:menaquinone decreases from the activity of other electron donors , the cytochrome oxidoreductase is deprived of its substrate , thus decreasing the rate of oxygen consumption . However , the loss of the primary means of succinate oxidation in Δsdh1 results in a premature accumulation of succinate during aerobiosis that is partially relieved by Sdh2 , which can function as a succinate dehydrogenase when cytosolic substrate concentrations favor this reaction . Published measurements of ubiquinone pools in E . coli show a similar trend [43] in quinone poise as oxygen decreases; but in that facultative anaerobe , reduced quinone increases as cells approach anoxia . The observation that a strong reductant can stimulate respiration of poorly respiring cells ( Figure 5B ) provides additional evidence of a menaquinol-limiting respiratory scheme . To our knowledge , this is the first time that any group has shown that oxygen consumption can be stimulated in living organisms that have shut off respiration by provision of exogenous electrons . The oxygen consumption profiles of the two sdh mutants revealed another interesting aspect of mycobacterial physiology , a downshift of respiratory activity was initiated by the parental strain in the range of 40–30% DO . The decline in the rate of respiration indicates that the organism switches to a less thermodynamically efficient mechanism for ATP production as oxygen levels drop – but are still sufficient for growth . Current understanding of the mechanics of the decline in respiratory activity exhibited by M . tuberculosis upon adaptation to anaerobic conditions has been guided by analysis of the transcriptome of cells as they pass into hypoxia in various models [14] , [44] , [45] . Here we report that M . tuberculosis accomplishes gross control of aerobic respiration by depriving the cytochrome c oxidoreductase of menaquinol via a slow electron flux through Sdh1 and demonstrate how carbon passing through the TCA cycle is subject to this mechanism that couples growth and electron transport ( Figure 7 ) . Importantly , this modulation in the rate of oxygen consumption occurs long before oxygen becomes limiting for growth [26] , and is absent any exogenously-provided inhibitor of respiration . This respiratory management scheme should have direct in vivo relevance considering that physiological oxygen levels are only a fraction of those commonly used in in vitro culture models , and reflect the point at which we observe a downmodulation of oxygen consumption [46] . This might explain the pathological preference of M . tuberculosis for the upper lobes of the lung [47] where mycobacterial cellular respiration can function more efficiently . As cells are carried into tissues farther from the lung epithelia , oxygen becomes scarce and cells are forced into a less efficient bioenergetic program which could lead to decreasing ATP production and more reliance on glycolysis , β-oxidation , or storage compounds to meet energy demand . Aerosol infection of C3HeB/FeJ mice using the Δsdh1 strain led to a tenfold reduction in CFUs in the lungs ( Figure 6B ) . Given the limitations of the murine TB model [48] , [49] , and the lack of encapsulated hypoxic lesions we observed in lung sections , we believe that the consequences of the phenotype reported here would be even more pronounced in models that more closely resemble human pathology , such as the rabbit or guinea pig . The niche in which latent M . tuberculosis survives , avoiding immune surveillance and maintaining undetectable cell numbers , is presently unknown . Several hypotheses have been suggested including the necrotic centers of granulomas [50] , [51] , adipocytes [52] , and recently mesenchymal stem cells [53] . This latter work is especially interesting in light of the conclusions presented here . It implies that the oxidative burst experienced when invading M . tuberculosis is engulfed by an alveolar macrophage would serve to inhibit respiration by shifting redox balance – toward an oxidized quinone pool in which quinol becomes limiting for oxygen reduction . Since cells in NRP-2 maintain an energized membrane , and are notably tolerant to single antibiotics but retain sensitivity to some combinations [54] , we think it is plausible that persistence is a function of the oxidative state of the milieu , and is the result of reduced respiratory flux . The presence of adequate oxygen alone is not sufficient to stimulate respiration; quinone redox homeostasis must be restored before respiration can reach optimal levels and the cell can take advantage of the energetic benefit of oxygen as its terminal electron acceptor . The necessity for members of the M . tuberculosis complex to maintain two possible frd enzymes ( sdhCDAB & frdABCD ) may be an indication of a metabolic plasticity which enables them to simultaneously utilize multiple carbon sources with different oxidation states and divert this carbon either into biomass production or storage molecules during growth , or into energy production for maintenance of PMF and repair during non-growth states [55] , [56] . The previously observed rerouting of a portion of carbon flux into the reductive C4 arm of the TCA cycle [7] suggests the involvement of fumarate reductase activity in hypoxia , indicating that other pathways contribute to anaerobic survival to some extent . Redundancy in Frd catalysis remains a possible explanation , and further genetic analysis will need to be performed to establish this , but thus far we have been unable to delete both sdh1 and sdh2 ( or sdh2 & frdABCD ) sequentially to address this hypothesis . Interestingly , Baek and Sassetti found that transposon mutagenesis of sdh2 led to an inability to shut down growth in hypoxia , thus if the primary means of succinate oxidation is through Sdh1 , oxygen ( but not carbon ) limitation does not result in cessation of growth , and succinate dehydrogenase activity continues to push carbon through the TCA cycle to continue biomass production [57] . There is now widespread acknowledgement of the fact that a reduction in the duration of TB chemotherapy could be achieved by finding ways to target non-replicating M . tuberculosis . The recent FDA approval of Bedaquiline lends credence to the idea that non-replicating cells still remain susceptible to inhibitors targeting maintenance bioenergetics , albeit at a reduced rate compared to current effective drugs [8] , [58]–[60] . In this communication , we propose that the removal of a metabolic block on M . tuberculosis respiration imposed by the contending action of the aerobic succinate dehydrogenase and fumarate reductase activities would prevent the orderly metabolic shift to quiescence . Compounds that serve to reduce quinones in non-dividing organisms would exhibit the pleiotropic effects garnered by increasing respiration , including enhancing membrane potential-driven uptake and decreasing fitness . Thus , progress toward the goal of shortening chemotherapy might be better served by searching for enhancers of respiration , which may reduce the numbers of organisms which are shifted to a persistent state .
Attenuated strains of M . tuberculosis were constructed by allelic exchange via specialized transduction [15] from the parental strain H37Rv . Null mutants in M . tuberculosis strains H37Rv , mc27000/mc26230 ( ΔpanCD , ΔRD-1 ) [61] , show identical growth characteristics in standard atmosphere as the parental strain ( unpublished results ) . T-Coffee [62] was used to assess homology between enzyme subunits ( Figure 1 ) and scores are presented as alignments of individual subunits corresponding to sdh2 . For a full list of strains used in this work , see ( Table S1 ) . For CFU experiments , mycobacteria were grown to OD600 0 . 5 and subcultured into media containing antibiotic and incubated at 37°C in a shaking incubator , or shifted to an anaerobic chamber ( <1 ppm O2 ) in bottles with vented caps and incubated shaking at 37°C . For growth experiments using single carbon sources , 7H9 media was supplemented with NaCl and BSA and individual carbon sources ( see Supplementary Methods for more detail ) . Analysis was performed using an Acquity UPLC system ( Waters , Manchester , UK ) coupled with a Synapt G2 quadrupole–time of flight hybrid mass spectrometer . Column eluents were delivered via Electrospray Ionization . UPLC was performed in HILIC mode gradient elution using an Acquity amide column 1 . 7 µm ( 2 . 1×150 mm ) using a method previously described [63] . The flow rate is 0 . 5 mL/min with mobile phase A ( 100% acetonitrile ) and mobile phase B ( 100% water ) both containing 0 . 1% formic acid . The gradient in both positive and negative mode is 0 min , 99% A; 1 min , 99% A; 16 min , 30% A; 17 min , 30% A; 19 min 99% A; 20 min 99% A . The mass spectrometer was operated in V mode for high sensitivity using a capillary voltage of 2 kV and a cone voltage of 17 V . The desolvation gas flow rate is 500L/h , and the source and desolvation gas temperature are 120 and 325°C . MS spectra were acquired in centroid mode from m/z 50 to 1 , 000 using a scan time of 0 . 5 s . Leucine enkephalin ( 2 ng/µL ) was used as lock mass ( m/z 556 . 2771 and 554 . 2615 in positive and negative experiments , respectively ) . For further details , see Metabolomics in Text S1 . Measurement of oxygen consumption rate in M . tuberculosis was performed using a Clark-type oxygen electrode ( Rank Brothers Cambridge , UK ) with data collected using an ADC-24 data logger ( Pico Technology , Cambridgeshire , UK ) . Cells were prepared in 490 cm2 roller bottles ( HSR = 26 ) , ( Corning , NY ) . For culture densities below OD600 4 . 0 , cultures were centrifuged for 5 minutes at 4 , 000 rpm and resuspended in fresh 7H9 media from which catalase was omitted . To detect induction of oxygen consumption by reductants , 5 mL early stationary phase cells ( OD600 5 . 0 ) were added to the incubation chamber and basal oxygen consumption was monitored for 100–200 seconds , at which point compound was added . After 200 seconds , maximal uncoupled oxygen consumption rate was determined by the addition of 20 µM CCCP for 100 seconds . We grew mycobacterial strains as described above in media containing OADC and the appropriate antibiotic for two passages before a single passage in media in which antibiotic was omitted immediately prior to animal infection . Female C57BL/6 mice , Rag-1−/− , and C3HeB/FeJ mice were obtained from Jackson Laboratory . Rag-1−/− mice were infected with ∼1×106 CFU of virulent mycobacteria via high volume tail vein injection . C57BL/6 mice and C3HeB/FeJ mice were infected via aerosol from a suspension of bacterial culture in PBS containing 0 . 05% Tween 80 and 0 . 004% antifoam , which yielded ≈100 or ≈50 cfu per lung . Four mice from each infection group were killed 24 h post-exposure , and lung homogenates were plated on 7H9-agar plates to determine the efficiency of aerosolization . We determined bacterial loads in lungs and spleen by plating for CFU at the indicated times from four mice per infection group . Five mice from each group were also used to determine survival times of infected mice . Pathological analysis and histological staining of organ sections were done on tissues fixed in buffered 10% formalin . Mouse protocols used in this work were approved by the Institutional Animal Care and Use Committee of Albert Einstein College of Medicine . Mouse studies were performed in accordance to National Institutes of Health guidelines using recommendations in the Guide for the Care and Use of Laboratory Animals . The protocols used in this study were approved by the Institutional Animal Care and Use Committee of Albert Einstein College of Medicine ( Protocol #20120114 ) NP_214761 ( Rv0247c ) , AFN48101 ( Rv0248c ) , CCP42978 ( Rv0249c ) , CCP46136 ( SdhC ) , CCP46137 ( SdhD ) , CCP46138 ( SdhA ) , CCP46139 ( SdhB ) , NP_216370 ( Ndh ) , CCP43122 ( NdhA ) | This work establishes the principle that Mycobacterium tuberculosis undergoes a metabolic remodeling as oxygen concentrations fall that serves to decrease its rate of oxygen consumption and therefore oxidative phosphorylation . Furthermore , cells can be stimulated to respire , even in low oxygen conditions , by providing reducing equivalents to the respiratory chain by either genetic manipulation ( deletion of succinate dehydrogenase ) or by exogenous addition of reducing agents such as DTT . Thus , activation of persister cells may be accomplished by increasing their respiration rate in low oxygen conditions . These findings will inform the design of novel drug screens which should seek enhancers of cellular respiration to find compounds which will serve to shorten the duration of TB chemotherapy . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"biochemistry",
"oxygen",
"metabolism",
"medical",
"microbiology",
"electron",
"transport",
"chain",
"microbial",
"pathogens",
"biology",
"and",
"life",
"sciences",
"bioenergetics",
"microbiology",
"metabolism"
] | 2014 | Succinate Dehydrogenase is the Regulator of Respiration in Mycobacterium tuberculosis |
A compelling body of literature , based on next generation chromatin immunoprecipitation and RNA sequencing of reward brain regions indicates that the regulation of the epigenetic landscape likely underlies chronic drug abuse and addiction . It is now critical to develop highly innovative computational strategies to reveal the relevant regulatory transcriptional mechanisms that may underlie neuropsychiatric disease . We have analyzed chromatin regulation of alternative splicing , which is implicated in cocaine exposure in mice . Recent literature has described chromatin-regulated alternative splicing , suggesting a novel function for drug-induced neuroepigenetic remodeling . However , the extent of the genome-wide association between particular histone modifications and alternative splicing remains unexplored . To address this , we have developed novel computational approaches to model the association between alternative splicing and histone posttranslational modifications in the nucleus accumbens ( NAc ) , a brain reward region . Using classical statistical methods and machine learning to combine ChIP-Seq and RNA-Seq data , we found that specific histone modifications are strongly associated with various aspects of differential splicing . H3K36me3 and H3K4me1 have the strongest association with splicing indicating they play a significant role in alternative splicing in brain reward tissue .
Classic experiments in cell culture systems have defined a ‘histone code’ , whereby distinct combinations of histone posttranslational modifications ( histone marks ) are associated with transcriptional repression or activation and play important roles in many other biological processes [1 , 2] . Recent evidence indicates that some histone marks , such as H3K36me3 and H3K4me3 are functionally coupled to alternative splicing [3–5] , however , a globally systematic analysis to investigate this relationship is lacking . Genome-wide mapping of histone modifications in many organisms has revealed their non-random distribution around exons , with some types of histone marks enriched in exonic regions compared to intronic regions [6 , 7] . Specifically , several studies on human-derived datasets find that histone modifications are associated with exon inclusion or exclusion , indicating the role of histone modifications in the regulation of alternative splicing . For example , a global analysis of the relationship between alternative splicing and histone modifications in two human cell lines found that histone modifications are globally associated with exon inclusion/exclusion patterns and that the change of histone modification patterns corresponds to cell-type specific exon usage [8] . Additional studies have demonstrated a functional correlation between specific histone marks and splicing outcomes at the human FgfII locus [4] and a predictive function for specific histone marks in alternative exon expression in human somatic cells [9] . Table 1 contains a detailed summary of published data on the link between histone modifications and alternative splicing . Despite these recent advances , the genome-wide association between histone modifications and alternative splicing , as elucidated by the analysis of global exon inclusion/exclusion level splicing patterns remains unclear . Furthermore , published approaches are not directly applicable to unannotated alternatively spliced exons because the models rely on defined exon and splice isoform annotation . Thus , the goals of our analysis are to define the association of specific alternative splicing patterns and histone modifications and to determine which marks are likely to play a dominant role in the regulation of alternative pre-mRNA splicing , as assessed by the statistically significant associations of these phenomena . In particular , we focused on sequencing from the nucleus accumbens ( NAc ) , a brain-reward region , given evidence that drugs of abuse as well as natural rewards regulate expression of the myriad of enzymes that catalyze and metabolize histone modifications as well as their genome-wide deposition . Indeed , major efforts in next generation chromatin immunoprecipitation sequencing ( ChIP-Seq ) and RNA-Sequencing ( RNA-Seq ) of reward brain regions have demonstrated that regulation of the epigenetic landscape likely underlies chronic drug abuse and addiction [16 , 17] . Furthermore , studies have implicated dysregulation of alternative splicing in human neurological disease [18] and cocaine exposure in mice . Prior studies in mice , including our own , intriguingly revealed that in addition to global regulation of histone modifications [15 , 16 , 19] , cocaine drives differential alternative splicing to a far greater extent than had been previously appreciated [15] . In addition , our previous study found that cocaine drives enrichment of several histone modifications , including H3K4me3 , H3K36me3 , H3K9me3 and H3K27me3 , on different types of alternatively spliced exons . To expand upon this previous finding , we developed a systematic approach to test for the global association between histone modifications and specific types of alternative splicing . Given mounting evidence for a role of histone modifications in alternative isoform expression , we developed two independent analyses , which we termed , “Exon Alternative Splicing Type” and “Exon Splicing Complexity” , both of which require quantification of ChIP signal at splice junctions . Briefly , exon splicing type is based on the classification of exon behavior according to transcript level annotation , as described in [15] . Specifically , each exonic region is classified into six different types: promoter , constitutive ( non-alternatively spliced exons ) , variant , alternative acceptor ( altAcceptor ) , alternative donor ( altDonor ) and polyA . The ChIP-Seq signal distributions around the splice sites are then compared between alternatively spliced exons and constitutive exons to uncover significant associations . Alternatively , exon splicing complexity is based on defining exon complexity as the total number of distinct exons to which the test exon is connected ( spliced reads ) . We used classical statistical models and permutation methods to quantify the association between splice-site localized ChIP-Seq signal and exon complexity . The two approaches differ in that the first analysis depends on a fixed set of transcript annotations , while the second analysis depends on the convergence of ChIP-Seq and RNA-Seq data and is independent of transcript annotation . Finally , our approach required the incorporation of controls for confounding factors , such as gene expression level , gene size , exon location , and other factors which may be independently associated with histone modification patterns and alternative splicing . Our findings indicate specific histone marks are associated with exon type and splicing complexity in brain reward region . The enrichment of each histone mark varies with respect to exon type , with H3K36me3 showing the greatest enrichment at alternative isoforms relative to other marks . Random forest and permutation test show specific histone marks , such as H3K36me3 and H3K4me1 , play a significant role in alternatively splicing . The computational methods developed in this study can be applied to other model organisms .
We analyzed RNA-Seq and ChIP-Seq data derived from the nucleus accumbens of mice treated with cocaine ( 20 mg/kg i . p . ) or saline for 7 days [15] . The sequences were aligned to mouse genome ( mm9 ) . The ChIP-Seq signal distributions were derived from the +/- 200bp flanking regions surrounding the acceptor and donor splice sites ( Fig 1A ) , respectively . We then plotted the mean ChIP-Seq signal distributions for four types of histone H3 modifications under cocaine and saline ( Fig 2 ) treatment . These include H3 lysine 36 trimethylation ( H3K36me3 ) , H3 lysine 27 trimethylation ( H3K27me3 ) , H3 lysine 9 dimethylation ( H3K9me2 ) and H3 lysine 4 monomethylation ( H3K4me1 ) . All of these marks are differentially regulated genome-wide by cocaine administration [15 , 20–22] . Exon type was defined using the same criteria described in [15] based on ensemble annotation ( see methods ) . In total there are six exon types: promoter , canonical , alternative acceptor , alternative donor , variant and polyA ( Fig 1B ) . As detailed below , we find that the enrichment of each histone mark varies with respect to exon type . Notably , H3K36me3 shows the greatest enrichment at alternative isoforms relative to other histone marks , with the ChIP signal being very strongly associated with alternative promoter usage ( p-value < 2 . 2e-16 ) , as compared to the other alternative splice types . In addition , H3K36me3 and H3K4me1 show the clearest separation of signal distribution patterns for different exon types , compared to the other histone marks analyzed . The flanking regions of exons were tested for a significant difference in ChIP-Seq signal between the different non-constitutive exon types and constitutive exons , using T-tests . The difference is considered statistically significant if the adjusted p-value is smaller than 0 . 05 . H3K36me3 is significant for alternative promoter and alternative polyA exon types , relative to constitutive exons , but only associated with some regions of altDonor , altAcceptor and variant types ( Fig 2A , Table 2 ) . Alternatively , we find that , relative to constitutive exon enrichment , the level of H3K4me1 is significantly higher in all alternatively spliced exon types except altDonor and altAcceptor at 3’ downstream ( Fig 2D , Table 2 ) , while AltAcceptor at 3’ downstream is specifically enriched for H3K27me3 and H3K9me2 ( Fig 2B , Table 2 ) . H3K27me3 also has a weak association with altAcceptor exons at 5’ and 3’ upstream ( Fig 2B , Table 2 ) . Finally , we find that the level of H3K9me2 is associated with alternative promoter and polyA exons and that there is a association with altDonor and altAcceptor exons ( Fig 2C , Table 2 ) . To further explore the histone modification patterns for different types of alternatively spliced exons , we computed the difference in ChIP signal between every exon and its corresponding constitutive exon , pooled by exon type ( Fig 3 ) . We found a significant difference in ChIP signal for each exon type analyzed ( p-value < 2 . 2e-16 ) . Moreover , H3K36me3 shows the greatest variation among different exon types while there is minimal variation for H3K9me2 and H3K27me3 . In summary , ChIP signal is considerably associated with alternative exon type , with H3K36me3 and H3K4me1 most strongly associated among the four histone marks studied . While our first analysis found that ChIP signal distribution is different for different exon types , indicating that ChIP signal can be used to identify exon types . We use Random Forests [24] to determine if the ChIP signal distributions are specific for each exon type . Random forest is an ensemble method that combines the results of multiple regression trees [24] and generally shows a better predictive performance over individual algorithms [25] . We regarded the exon type as the response variable of the model and the ChIP signal in the upstream and downstream +/- 200 bp flanking regions from exons of different exon types as the explanatory variables . This was done independently for the two treatments cocaine and saline . Table 3 shows the performance of the models based on cocaine and saline treatment do not differ significantly . The model has very good accuracy and macro-averaged precision for both treatments , with accuracy much higher than random ( ~0 . 58 ) ( Table 3 ) . These results demonstrate that the model has excellent power to distinguish different exon types based on the ChIP signal , indicating that the histone modifications patterns are associated with alternative splicing . In addition , the model performs equally well under cocaine and saline treatment , suggesting that chromatin-directed alternative splicing is a basal transcriptional mechanism . To elucidate which histone mark is most strongly associated with the different exon type , we built the model based on the variables from each mark and further applied a process similar to the model selection approach in which each explanatory variable was progressively added into the current model to search for the best model . The model performance was calculated based on the component combination of histone marks . As illustrated in Table 4 , the accuracy of models varies from 0 . 59 to 0 . 72 . H3K36me3 and HK9me2 have the largest and smallest accuracies , respectively . The model performance increases with each mark added sequentially until the accuracy of model with H3K36me3+H3K4me1+H3K27me3 which is comparable to the full model—adding H3K9me2 does not further contribute to the model performance . Additionally , we consider the importance score of each variable from the full model ( Fig 4 ) . Consistent with what we observed above , the ChIP-Seq signal of H3K36me3 in the flanking region 5’ splice site downstream is the most informative for classifying the exon type , followed by H3K4me1 at the 3’ site . The 3’ downstream of H3K27me3 at donor splice sites also shows a greater contribution than the other markers , while H3K9me2 and H3K27me2 in most of the regions are the least informative . Therefore we conclude that the H3K36me3 and H3K4me1 play a dominant role on the regulation of alternative splicing . Unlike the above approach , which depends on a fixed set of transcript annotations and ChIP-Seq data , the second analysis is based on defining the exon splicing complexity as the total number of distinct exons that the exon in question is connected to by spliced reads . We test for significant association between the continuous ChIP-Seq variable and the discrete complexity variable . In testing for this association , particular attention must be made to control for the possibility of confounding factors ( Fig 5 ) . ChIP signal and complexity may independently be associated with expression level , which could result in a significant association due entirely to this confounding factor . To control for this an ANCOVA model is used to regress out the expression level factor . Additionally , the association may also be confounded with the number of exons of the gene , the size of the gene , the location of the exon in the gene , the ChIP signal window size , and which samples were paired . To avoid these confounding factors , subsets of data points were chosen so as to hold all of these variables constant . We did not hold expression level constant because that would not allow for enough data points to perform the analysis , which is why ANOCVA was used to control for that factor . Since there is no clear systematic choice for the remaining parameters , we took a global approach as follows . For a fixed choice of parameters ( e . g . genes with five exons , focusing on the 3’ junction of exon 2 , with ChIP window size of 250 bases , pairing ChIP sample 2 with RNA-Seq sample 3 ) we compute the ANCOVA p-value for association between the ChIP signal and complexity . We plot the distribution of the ANCOVA p-values over the entire parameter space resulting in a distribution of p-values . As control , for each choice of parameters we permute the ChIP signal randomly among the genes and compute the ANCOVA p-value for the permuted data . We plot the distribution of these permuted ANCOVA p-values over the entire parameter space for comparison to the distribution of unpermuted p-values . A significant separation of these distributions , particularly near the small values of p , indicates a significant association between ChIP signal and splicing complexity . This approach of looking for a separation of real and permuted distributions over the entire parameter space avoids making an arbitrary choice of parameters . We see a clear separation of distributions for two of the histone marks , H3K36me3 , and H3K4me1 ( Fig 6 ) . This is consistent with prior molecular biological approaches demonstrating that these two marks are functionally involved in alternative splicing ( see Discussion ) . This finding is also consistent with that derived from the exon type analysis above , indicating that these two histone marks are specifically associated with alternative pre-mRNA splicing in in the brain .
Alternative isoform expression presents a compelling mechanism by which neurons mount a stable response to environmental stimuli , functionally analogous to stable isoform selection by differentiating neurons during development . In the latter context , alternative splicing confers neuronal identity and is maintained throughout life [26] . We thus sought to explore the epigenetic regulation of alternative splicing through an investigation of the genome-wide association of specific histone modifications and alternatively expressed exons . To uncover the genome-wide association between histone modifications and alternative splicing in mammalian brain , we analyzed global , exon-level splicing patterns in neurons . We designed two approaches to test for the association of specific alternative splicing patterns and histone modifications , in order to determine which play a significant role in the regulation of alternative pre-mRNA splicing . We first developed a method to demonstrate the association of particular histone modifications with different types of alternatively spliced exons in brain . This led to the finding that enrichment of each analyzed histone modification varies with respect to each type of spliced exon , with H3K36me3 showing the greatest enrichment at alternative isoforms relative to other histone post-translational modifications . This result is well founded in the context of prior research in non-neuronal systems . First , genome-wide analyses of nucleosome-positioning data sets from humans , flies and worms show that exons have increased nucleosome-occupancy levels with respect to introns [6 , 27] , and that H3K36me3 is found consistently to be preferentially enriched in exons . Second , the correlation of exon inclusion levels and nucleosome distribution patterns suggests that nucleosome positioning defines exons at the chromatin level , indicating that DNA-coded splicing signals mediate the observed differences in the chromatin landscape of exons and introns . Finally , beyond nucleosome occupancy , exons are differentiated from introns by specific histone modifications [7 , 28] , which may play a key role in exon recognition during co-transcriptional splicing . We took an unbiased approach to examine the behavior of histone modifications at alternative splice sites , and discovered that H3K36me3 and H3K4me1 , but not H3K9me2 , show a clear separation of distribution patterns for different types of alternatively spliced exons . Specifically , we find that H3K36me3 is highly significantly associated with alternative promoter and alternative polyA splice types . Furthermore , the ChIP-Seq signal of H3K36me3 in the flanking region of the acceptor splice site is the most informative for classifying alternative exon types using random forest classifiers ( Fig 7 ) . These results are especially promising given prior data using a de novo pattern-finding algorithm which indicates that enrichment of H3K36me3 correlates with increased exon usage in alternatively spliced genes [14] . This finding was expanded in a study in C . elegans , which further confirmed that the exon enrichment of H3K36me3 is globally conserved in human and mouse genomes [29] . While ours is the only genome-wide computational analysis to associate H3K36me3 with alternative splicing in a mouse brain reward region , several groups have reported promising experimental data to support this mechanism at specific genes . For example , using a β-globin gene reporter system , Kim et . al . demonstrated that splice-site mutations , which correlated with enhanced retention of a U5 snRNP subunit on transcription complexes downstream of the gene , affected H3K36 methylation , while a polyA site mutation did not [5] . Further , global inhibition of splicing by spliceostatin A caused a rapid repositioning of H3K36me3 away from 5’ ends in favor of 3’ ends , indicating a direct relationship between splicing mechanisms and H3K36 methylation status . Finally , a landmark paper by Luco et . al . demonstrated a direct interaction between H3K36me3 and the spliceosome machinery , specifically PTB and MRG15 , at the FGFR2 , TPM2 , TPM1 , and PKM2 loci in human cell lines [4] . Together with the results of our global analysis , these studies strongly implicate H3K36me3 in mediating alternative splicing in vivo . As described above , the role of specific histone modifications in marking exons is well-documented , yet there is limited data on this mechanism specifically in the brain , despite the critical role of alternative splicing in neuronal identity and survival [18 , 26 , 30] . Our prior study on the role of histone modifications in the nucleus accumbens found that , compared to saline , cocaine drives differential isoform expression to a markedly greater extent [15] . Further , particular histone modifications were differentially enriched by cocaine treatment , and this enrichment could be further distinguished on the basis of the associated exon alternative splice type . To expand upon this previous finding , we focused on an exploratory analysis of the relationship between histone modifications and specific types of alternative splicing and developed a systematic approach to test for the global association between them . Specifically , while the exon type annotation is similarly derived , the exon type approach differs from our prior analysis in that we model the association based on the distribution , but not enrichment between treatments , of histone marks in flanking regions . The association we derived is therefore a global effect and independent of specific target clusters within treatment groups ( see methods ) . Despite these distinct computational approaches , both studies identify an important role of H3K36me3 . With respect to differential enrichment , Feng et . al . found that this mark is differentially enriched in cocaine-treated NAc tissue at alternative donor , alternative acceptor and variant spliced exon types , all of which we find here to be significantly differentially enriched for H3K36e3 relative to constitutive exons . Similarities in the two datasets also emerge with the analysis of H3K4me1 and me3 , which are found to be differentially enriched in cocaine-treated samples at alternative acceptor and variant exon types [15] . While our previous analysis identified a clear role for H3K4me3 in the association with alternative exon expression , our current method was unable to analyze this particular mark due to insufficient read coverage of ChIP-Seq samples necessary for the complexity analysis . Finally , H3K4me1 and H3K4me3 enrichment has also been implicated in chromatin-directed alternative splicing in cell culture [4] , indicating a functional role for both H3K4me1 , H3K4me3 and H3K36me3 in alternative splicing . We have taken an unbiased approach to investigate chromatin-directed alternative splicing in brain , having developed an innovative computational model to test the association between alternative exon expression and specific histone modifications . Using this method we have applied a single statistical test to the association of ChIP-Seq and RNA-Seq data within a brain-derived dataset to find that there are highly significant associations between alternative splicing and the specific histone marks H3K36me3 and H3K4me1 . This association is found in both treated and un-treated neuronal tissue , indicating the fundamental nature of this global mechanism . Future studies will be needed to discern the role of cocaine , as well as other forms of neuronal activation , in the regulation of chromatin-directed alternative splicing , both globally and at specific genes .
High-throughput ChIP-Seq and RNA-Seq datasets from [15] were downloaded from GEO ( https://www . ncbi . nlm . nih . gov/geo/ , GSE42811 ) . Please refer to the published work for details on animal treatment and sample preparation [15] . For each treatment , four histone modifications were assayed: H3K36me3 , H3K4me1 , H3K27me3 and H3K9me2 using ChIP-Seq . ChIP-Seq reads were aligned to mouse reference genome ( NCBI37/mm9 ) using Bowtie2 ( Version 2 . 1 . 0 ) with default parameters [31] . The RNA-Seq reads were aligned using STAR ( Version 2 . 4 . 1d ) [32] with index built with RefSeq annotation . Data were normalized and quantified by the PORT pipeline ( https://github . com/itmat/Normalization ) and further normalized for gene length by the FPK method ( fragments per kilobase of gene length ) . The exon-level expression values were also normalized for read depth by PORT . The definition of exon type depends on a given set of gene annotations and is defined using the criteria described in [15] . Different exon types were classified by pairwise comparison of the boundaries of exons across various isoforms from the same gene . There are six exon types in total: promoter , constitutive , alternative acceptor , alternative donor , variant and polyA . Briefly , each gene’s exons were sorted from the 5’ to the 3’ end according to their genomic coordinates; then each exon is compared across isoforms from the same gene . If an alternative left or right boundary is found , it is classified as “alternative acceptor/donor . ” If an exon overlaps with an intron , it is classified as “variant” and if there is an alternative boundary found in the first/last exon , it is classified as “promoter/polyA . ” For simplicity , exons that belong to multiple types were removed from analysis . Thus each exon type represents a unique combination of exon-intron boundaries . Histone modification signals on the flanking regions of exons were calculated in two steps: First , for each exon , flanking regions were defined as the 400bp centered at the acceptor and donor splice sites respectively . If the length of an exon is less than 400bp , the two flanking regions were truncated so they do not overlap . Second , for a given exon , we kept only the ChIP-Seq reads that overlap at least one of the two flanking regions ( Fig 1A ) . The total number of overlapping reads was then equalized across all samples by random resampling , to make them comparable . The resampling approach results in uniform ( null hypothesis ) signal distributions across all samples , while scaling approaches result in uniform means but heterogeneous distributions , which is not desirable . The normalized reads were then quantified for each flanking region . To visualize the signal distribution for each exon type , we computed the average ChIP-Seq signal across the flanking regions , averaged over all exons of the same type ( Fig 1A and Fig 2 ) . We further divided the regions surrounding each splice site into the exonic and intronic regions with 200bp length each ( except when truncated due to a small exon length ) . For each exon this gives ChIP-Seq signal for four regions: the intronic region at the acceptor splice site ( 5’ upstream ) , the exonic region at the acceptor splice site ( 5’ downstream ) , the intronic region at the donor splice site ( 3’ downstream ) and the exonic region at the donor splice site ( 3’ upstream ) . One of the ways that we demonstrate that there is an association between ChIP signals and exon type is by showing that ChIP-Seq signal is predictive of exon type . Therefore , we consider this as a multi-class classification problem , for which a variety of classification algorithms can be applied . In this study , we used the Random Forest classification algorithm because of the wide consensus on its performance [25] . Additionally , Random Forest classification provides a ranking of the different histone modifications by their predictive power . The performance of the model was evaluated by 5-fold cross validation . The dataset was randomly partitioned into five equal sized subsamples . Among those five subsamples , one subsample was used as test data to evaluate the model performance and the remaining four subsamples were used as training data to construct the model . The model was further tuned based on training data to achieve the best parameters by calculating the model performance under different combination of parameters . The model with lowest error rate was then selected . Then Accuracy , macro-averaged precision and macro-averaged recall were calculated based on test data to measure the model performance . These values were calculated according to the confusion matrix as shown below: For each exon type i , i = 1…6 , the confusion matrix is calculated as: Accuracy=∑i=16tpi+tnitpi+fni+fpi+tni6 Errorrate=∑i=16fpi+fnitpi+fni+fpi+tni6 Macro−averagedprecision=∑i=16tpitpi+fpi6 Macro−averagedrecall=∑i=16tpitpi+fni6 Five-fold validation means this whole process was repeated five times , with each of the five subsamples used as the test data in turn . Splicing complexity is an integer associated to the 5’ or 3’ end of an exon . In contrast to exon type , complexity depends on a particular RNA-Seq data set . In particular for a given exon boundary b , the complexity c ( b ) is the number of distinct locations that are connected to location b by a spliced read . For a fixed location x , there may be many reads which splice from b to x , however the complexity only counts distinct locations , not distinct reads , so each distinct genome location x increments the complexity by one if there are any reads at all spliced from b to x; otherwise it increments by zero . The splice junction indicated by the read may or may not be annotated; all reads which splice from location b are counted . For example , if a gene has only one expressed splice form , then the complexity of all of its exon boundaries equals one , except at the two boundaries consisting of the start and end of transcription , for which the complexity equals zero . With few rare exceptions , complexity typically varies from between zero and ten . In this analysis a statistical association between splicing complexity and ChIP-Seq signal is tested for . We expect splicing complexity to increase with the expression level of the gene , which could then be confounded with the ChIP signal , if that signal is also associated with expression level . An ANCOVA model is employed to control for this possibility . The ANCOVA model can be formulated as follows: yij=μ+αi+βxij+ϵij Where yij is the ChIP-Seq ( histone modification ) signal of exon j with splicing complexity i , xij is the normalized FPK ( expression level ) of exon j with splicing complexity i , μ is the reference level , αi is the effect of splicing complexity i ( i = 1… . n ) , and β is the regression slope that quantifies the ( linear ) relationship between the FPK and the ChIP-Seq signal . Type III SS is used for testing the significance of each splicing complexity level ( Ho: α1 = α2 = α3 = … = αn = 0 ) and the linear relationship between the ChIP-Seq signal and FPK ( Ho: β = 0 ) . If the test is significant , it indicates that after controlling for expression level , there is still a significant difference for the ChIP-Seq signal among different exon complexity levels .
High-throughput ChIP-Seq and RNA-Seq datasets from [15] were downloaded from GEO ( https://www . ncbi . nlm . nih . gov/geo/ ) , accession number GSE42811 . | The mammalian brain responds to environmental stimuli through changes in gene expression . Over the past decade a robust body of bioinformatics data has shown that in neuronal tissue such gene expression is regulated by changes in the epigenetic landscape , including modifications to chromatin . Further , a small but compelling body of literature has recently described chromatin-regulated alternative splicing , suggesting a novel function for neuroepigenetic remodeling in alternative isoform expression . However , the extent of the genome-wide association between particular histone modifications and alternative splicing remains unclear , in part due to limitation in methods to model the convergence of chromatin modifications and gene expression changes . We report here our innovative computational approach to model the association between alternative splicing and histone marks in the nucleus accumbens , a brain reward region . We found , remarkably , that specific histone marks are associated with and predict both alternative splicing exon type and alternative splicing exon complexity in the brain , with particular histone marks showing the significant enrichment at alternative exons . This approach is the first to model chromatin-mediated alternative splicing globally , and our findings are consistent with recent data on the molecular biology of this mechanism , homing in on a subset of chromatin modifications that are functionally relevant to alternative splicing in vivo . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods",
"&",
"models",
"Data",
"access"
] | [
"alkaloids",
"medicine",
"and",
"health",
"sciences",
"chemical",
"compounds",
"nucleus",
"accumbens",
"dna-binding",
"proteins",
"brain",
"invertebrate",
"genomics",
"histone",
"modification",
"alternative",
"splicing",
"behavioral",
"pharmacology",
"epigenetics",
"cocaine... | 2017 | Histone posttranslational modifications predict specific alternative exon subtypes in mammalian brain |
Vector-borne transmission of Chagas disease has become an urban problem in the city of Arequipa , Peru , yet the debilitating symptoms that can occur in the chronic stage of the disease are rarely seen in hospitals in the city . The lack of obvious clinical disease in Arequipa has led to speculation that the local strain of the etiologic agent , Trypanosoma cruzi , has low chronic pathogenicity . The long asymptomatic period of Chagas disease leads us to an alternative hypothesis for the absence of clinical cases in Arequipa: transmission in the city may be so recent that most infected individuals have yet to progress to late stage disease . Here we describe a new method , epicenter regression , that allows us to infer the spatial and temporal history of disease transmission from a snapshot of a population's infection status . We show that in a community of Arequipa , transmission of T . cruzi by the insect vector Triatoma infestans occurred as a series of focal micro-epidemics , the oldest of which began only around 20 years ago . These micro-epidemics infected nearly 5% of the community before transmission of the parasite was disrupted through insecticide application in 2004 . Most extant human infections in our study community arose over a brief period of time immediately prior to vector control . According to our findings , the symptoms of chronic Chagas disease are expected to be absent , even if the strain is pathogenic in the chronic phase of disease , given the long asymptomatic period of the disease and short history of intense transmission . Traducción al español disponible en Alternative Language Text S1/A Spanish translation of this article is available in Alternative Language Text S1
Chagas disease , responsible for more deaths in the Americas than any other parasitic disease [1] , has become an urban problem in the city of Arequipa , Peru [2] , [3] . Nevertheless the debilitating symptoms of chronic Chagas disease , common across southern South America , are rarely seen in hospitals in the city ( see Supp . Info Text S1 ) . The lack of obvious disease in hospitals and the general population in Arequipa has led to speculation among local physicians that the local strain of the etiologic agent , Trypanosoma cruzi , has low chronic pathogenicity ( personal observation ) , even though it has caused fatal acute infections in infants [4] and animal models[5] , [6] . The vast majority of the 8–10 million individuals infected with T . cruzi [7] have the indeterminate form of Chagas disease . These individuals exhibit no symptoms or signs of their infection , but 20% to 30% are expected to progress to cardiac or digestive forms of chronic Chagas disease , which are difficult to treat and potentially fatal [8]–[10] . Progression to clinically evident cardiac disease is a slow process[8] , [10] . For vector-borne transmission , decades may pass from the time of initial infection through contact with the feces of an infected triatomine bug until the onset of cardiac symptoms . Treatment with existing antitrypanosomal drugs , benznidazole or nifurtimox , appears to slow or prevent disease progression [9] , but treatment is thought to be more effective when administered early in the course of infection [11] . The long asymptomatic period of Chagas disease leads us to an alternative hypothesis for the absence of clinical cases in Arequipa: transmission in the city may be so recent that most infected individuals have yet to progress to late stage disease . In order to evaluate this hypothesis it is necessary to elucidate the timing of T . cruzi infection in the population . The date of infection with T . cruzi is rarely known for individuals with Chagas disease . No existing assay of parasite or host factors yields reliable clues to the duration of infection . The dynamic process of T . cruzi transmission , however , creates clear spatial and temporal patterns of infection in insect vectors [12]-[14] and human hosts [15]–[20] . Traditionally , analyses of infectious diseases have aimed either to describe risk factors for infection at a static moment in transmission , using statistical methods to smooth heterogeneities in exposure between individuals created by the agent's spread [21]–[22] , or to model the agent's spread through time and space , assuming a homogenous population [23] . Recent work has taken a more unified approach , estimating parameters of spread and local risk factors together [24]–[25] . Outside of a handful notable exceptions , especially [26]-[27] , most unified analysis of disease spread and local risk factors address situations in which the time and place of the introduction of the disease agent is known . Here we develop a regression model , ‘epicenter regression , ’ to infer the temporal and spatial spread of a disease agent for the more common situation in which the site or sites of introduction of a disease agent is unknown . Trypanosoma cruzi is a slow-moving parasite . When it is transmitted in an epidemic phase , the period of time during which each individual is exposed to infection varies greatly based on how far away he or she lives from the site of introduction of the parasite . The heterogeneity in exposure time that results from the slow spread of the parasite may result in observable spatial clustering of infections . Observed spatial clustering , however , may also arise in endemic transmission if local risk factors for infection are concentrated in certain areas . Epicenter regression explicitly models the duration of an individual's exposure as a function of the distance of their household to an ( unknown ) site of introduction of a disease , then , given the exposure time of each individual's household , estimates the effect of risk factors measured in the house on the probability of infection for each individual . We fit models to patterns of T . cruzi infection in insect vectors and humans hosts in the peri-urban community of Guadalupe , Arequipa , Peru [3] , [15] . We use Bayesian methods and Monte Carlo Markov Chains ( MCMCs ) , an approach which allows us to make inference on parameters that reflect uncertainty about unknown factors , such as the location of cases in a dynamic infectious process [26]–[28] . We consider models with a single site of introduction of T . cruzi as well as models with multiple sites and times of introductions leading to multiple micro-epidemics . By ‘micro-epidemic’ we mean a focus of transmission , seeded from the same introduction of the parasite , that is discrete and discernable from the larger pattern of transfomission in the community . We also compare these models to an endemic model , in which each individual is assumed to have been exposed to the parasite since birth . Under the endemic model clustering of infection is explained mainly by clustering of household risk factors . Using the estimates from these models , we calculate the expected prevalence of infection among the population for each calendar year up to disruption of transmission through insecticide application in 2004 . We discuss how a more precise understanding of the history of transmission might explain the absence of late-stage Chagas disease in Arequipa , and potentially inform clinical management of individuals with indeterminate Chagas disease .
We conducted cross-sectional entomologic and serologic surveys in one recent settlement ( pueblo joven ) , Guadalupe , on the southwestern margin of the city of Arequipa , Peru [14] . Vector-borne transmission of Chagas disease in Guadalupe was disrupted through application of deltamethrin insecticide in November of 2004 . Concurrent to insecticide application , vector presence and density were determined through one person-hour timed search by trained professionals . Live and moribund fifth instar and adult triatomines were examined for T . cruzi consecutively for each site until 1 positive insect was found , 10 negative insects had been examined , or all available insects had been examined , whichever came first [14] . In August of 2005 , all residents of the community were invited to participate in a serological survey for Chagas disease [15] . The positions of all households in Guadalupe were determined with a handheld global positioning system ( Garmin Corporation , Olathe , KS , USA ) . Sera were screened using a commercial enzyme-linked immunosorbent assay ( ELISA ) kit with an epimastigote lysate antigen ( Chagatek , Biomerieux , Buenos Aires , Argentina ) following the manufacturer's instructions . Positive ELISAs were confirmed by an immunofluorescent antibody assay ( IFA ) following standard methods; a titer of 1∶32 was considered positive . Statistical analysis of data from Guadalupe was approved by the Institutional Review Board of the University of Pennsylvania . Epicenter regression makes inference on where and when a disease agent was introduced into a community , as well as the effect of household-level covariates on the risk of infection given exposure . We begin with a simple model [29] in which an individual has a constant risk of being infected once exposed . The probability that an individual is infected is equal to 1 minus the probability of the individual escaping infection: 1 - e -Risk given exposure Duration of exposure . This equation is known as the catalytic model , because it also describes the probability of a change in state of molecules exposed to a constant bombardment of a catalyst [30]–[31] . We expand this framework into a biologically plausible model by allowing ‘risk given exposure’ and ‘duration of exposure’ to vary among individuals depending on observed covariates , and estimate the effect of the covariates from the observed data . We estimate the risk of exposed individual , i , due to the covariates measured in their household , j , using a traditional method to estimate risk , a log-linear model: Riski = . Here , Xi represents a vector of covariates measured in each individual , and the β parameter describes how those covariates increase or decrease the log of the risk to each individual living in the exposed household . We assume that the effect of each covariate is constant , and varies neither year-to-year nor location-to-location . The intercept term , α , denotes the baseline risk of exposed individuals when all covariates are zero . We examined covariates previously shown to be associated with T . cruzi infection in children in the community . These included the presence of animals , almost exclusively dogs and cats , sleeping in the domestic area of the household and the number of vectors in the domestic portion of the household during timed entomologic search [15] . For each individual we estimated the period of time over which he or she had been exposed to T . cruzi as the lesser of two more-readily observable or estimable correlates of exposure: the individual's age , and the duration of exposure of their household . We assume the time of exposure of a household is a function of three unobserved parameters: T , the length of time since the introduction of T . cruzi in the community; d , the distance of the household from the household into which the parasite was first introduced , and , r , the rate of spread of the parasite . In the present analysis , we assume that a household becomes exposed at a time proportional to its distance from the site of introduction of the parasite . This assumption is a common simplification of mathematical analysis of invasive organisms based on diffusion equations , which concludes that , when individuals are assumed to move according to random walks , the wave front of the population will advance at a constant rate away from the site of introduction [23] . The time after the introduction until the parasite reaches the household , during which the residents of the household are not exposed , is: djk/r , and the total duration of exposure is T- djk/r , where djk is the distance of household j from the location of introduction k . The probability of infection of individual i under the model is: for all individuals born before their household was exposed to vectors carrying the parasite; and: for all individuals born after their household was exposed to vectors carrying the parasite . A steep hilltop separated households in the study area . We calculated the distance between the households going around this hilltop . As in the related proportional hazards model [32] , time of exposure in epicenter regression must be limited to positive values . In the above model if d/r is greater than T , then time of exposure is set to zero , and the probability of infection is equal to zero . We used data on the presence of T . cruzi in insects to further refine our estimates of parameters T , d and r . For parameter regimes in which the model predicted no exposure for households in which we had observed T . cruzi in vectors , we set the likelihood to zero . We expanded our epicenter regression framework to consider the possibility that T . cruzi was introduced into Guadalupe on multiple occasions . Each introduction would occur in a different household , k , at a different time , Tk . We calculate the duration of exposure of individuals in household j due to introduction into household k as: Tk – ( djk/r ) . We assume that once transmission is established in a household it remains established . We also assume that the introduction of additional parasites into a household in which transmission is already established does not incrementally increase the risk of infection among those living in the house . The duration of exposure of each individual is therefore the maximum of their exposure to each introduction , or their age if their household was exposed prior to their birth . The probability of infection of each individual is otherwise as calculated above . For comparison purposes , we fit an endemic model , in which we assume that each individual's time of exposure was equivalent to their age . The infection probability of individual i living in household j under the endemic model is: . Bayesian analysis of epicenter regression begins with what is known about the epidemic before testing people in the community , the “priors” of the model [33] . A key assumption of the model is that the parasite was introduced into a household and is spreading from the site of introduction . Before observing the data we have no information about into which household ( or households ) the parasite was first introduced , and therefore set a uniform prior distribution on the probability that each household was a site of introduction . We set non-informative prior distrifbutions ( Gaussian with a mean of zero and a standard deviation of 103 ) on the effect of household-level covariates ( the βs ) . The community of Guadalupe was 40 years old at the time of data collection . We therefore set a uniform prior , bounded at 1 and 40 years , on the time since introduction of the parasite , for each introduction , Tk . Providing an explicit prior on the speed of spread of the pathogen allows us to better tailor our analysis to the particulars of T . cruzi transmission by T . infestans . We based our prior on the speed of spread of the wavefront of T . cruzi on longitudinal data from Villa la Joya , a community similar to Guadalupe in terms of household density , history , and animal husbandry practices . In Villa la Joya we surveyed 30% of households in January of 2008 , uncovering 6 sites of T . cruzi infection among triatomine bugs . We conducted an exhaustive survey in conjunction with insecticide application by the ministry of health , following the methodology used in Guadalupe , in November and December of 2008 . We identified forty-four additional foci of T . cruzi in vectors , in well-defined clusters around pre-existing sites of parasite presence , during the exhaustive survey . Assuming that the parasite had spread from the center of each cluster , it would have traveled approximately 18 meters over the course of 10–12 month interval between the preliminary and exhaustive surveys in order to reach each of the 44 foci . In no case had the parasite spread farther than 58 meters from a pre-existing site . Based on these findings , we assigned a normal distribution , with a mean of 20 meters and a variance of 100 meters , to describe the prior probability on the yearly speed of the spread of the parasite in Guadalupe . A spread rate of 20 meters/year is akin to a single parasitic household infecting on average between 6 and 7 additional households a year when introduced into a susceptible community . We fit epicenter regression models using Bayesian methods and Monte Carlo Markov Chains ( MCMCs ) . We updated MCMCs using the Metropolis and Metropolis-Hastings algorithms [33] ( see annotated code in technical appendix: Text S3 , Dataset S1 , Text S4 , Text S5 ) . For the endemic model and models with 1 to 10 introductions , we ran 50 replicate MCMC chains , each of a length of 1 million estimates . We discarded the first 100 , 000 and retained every 10th estimate in the remainder of a chain to diminish autocorrelation among the estimates . For each pair of models compared , we estimated the Bayes factor by the average , over the 50 pairs of chains , of the ratio of harmonic means of the posterior likelihood for the models [34] . We assured convergence of the chains using Geweke's test [35] as well as the Gelman-Rubin statistic [36] . We considered models with 1 through 10 introductions of parasite into the population ( see movies in supplemental materials: Text S2 , Video S1 , Video S2 , Video S3 , Video S4 ) . We estimated the prevalence for each calendar year under alternative models by integrating the risk of infection per unit of time over the time period that each individual was exposed . Finally , we quantified the expected number of cases of late-stage Chagas disease among individuals infected with T . cruzi at the time of the study . The time between infection and development of late stage disease is note fully known , but generally estimated at between 10 and 30 years [8] . We used three alternative models to describe the probability of onset of late-stage disease as a function of time: 1 . A Poisson distribution centered at 20 years; 2 . A Gaussian distribution with a mean of 20 years and 95% of the probability density between 10 and 30 years; and , 3 . A uniform distribution bounded at 10 and 30 years . For each of these we integrated over the predicted temporal dynamics and calculated the expected number of cases , and the probability of observing exactly zero cases . We used a conservatively high estimate of the proportion of individuals who will eventually develop late stage disease , 30% [8] .
Models describing T . cruzi transmission as epidemic fit the data collected in the community of Guadalupe much better than an alternative , endemic model . The odds in favor of the epidemic models compared to the endemic model increased more than 4 orders of magnitude after considering the observed data – i . e . the Bayes' factors comparing the epidemic models to the endemic model exceeded 104 ( Figure 1 ) . A Bayes' factor of greater than 10 is considered ‘strong evidence’ in favor of one model over another [34] . Models describing transmission of T . cruzi as a series of focal micro-epidemics were better supported than a model with a single epidemic stemming from one site of introduction . In particular , a model with four micro-epidemics had the greatest support from the observed data; the odds in favor of this model compared to the single-epidemic model increased 15-fold after considering the observed data ( Figure 1 ) . Under the four-epicenter model , the parasite was first introduced into Guadalupe about 20 years ago . When we tabulated the exposure time and risk of infection of individuals in the population in the four-epicenter model we found that around half of infections occurred in the 5 years previous to disruption of transmission through insecticide application , and 90% of infections occurred over a period of 12 years ( Figure 2 ) . These estimates were consistent across models with 2 to 10 epicenters ( Figure 2 ) . In contrast , under the endemic model , prevalence increased slowly over a much longer time frame . Spatially , the first introduction in the four-epicenter model occurred in the southwest of the community ( Figure 3 ) . A second micro-epidemic was then seeded in the northwest of the community , followed by a third in the southeast . Estimates on the position of the fourth micro-epidemic varied; some centered on a household with vectors carrying T . cruzi to the far east of Guadalupe , while others were nearer to a household with a human case in the far north of the community . Models with more than four introductions generally further subdivided these micro-epidemics; such divisions decreased the statistical support for these models . Temporally the four-epicenter model captured the relationship between age and prevalence observed in the data ( Figure 4 ) . Estimates of the time of the first introduction of T . cruzi into the community were remarkably similar across epidemic models ( 19 . 98 years ago in the one-epicenter model , 20 . 31in the four-epicenter model ) . The estimated effect of covariates on the risk of infection given exposure were also similar , with a 1–2% increase of risk per bug caught in the domestic area of the household and a 40–60% increase in risk among exposed individuals who allowed animals to sleep inside the domestic area of the household at night ( Table 1 ) . Posterior estimates on the rate of spread of T . cruzi , however , varied greatly . In the best-fit four-epicenter model a single exposed household would expose on average 4–5 additional households per year in a susceptible community . By contrast , in the one-epicenter model , the estimated rate of spread would expose an unrealistically large number of households ( on average 15 households/year in a susceptible community ) . The higher estimated spread rate was counterbalanced by a lower estimated risk of infection given exposure in the one-epicenter model compared to the four-epicenter model . When we combined our posterior model estimates of the timing of infection with T . cruzi with three alternative models of the distribution of waiting times between infection with the parasite and development of late stage Chagas disease , we found very high probabilities of observing less than one case of late stage disease in Guadalupe at the time of our survey ( Figure 5 ) . Under the four-epicenter model the probabilities of observing less than one case of late-stage disease were 0 . 696 , 0 . 937 , 0 . 967 under the Poisson , Gaussian and uniform models respectively . Given the posterior estimates of the timing of T . cruzi infection under the one-epicenter model , the probabilities of observing fewer than one case were 0 . 797 , 0 . 763 , 0 . 779 under the Poisson , Gaussian and uniform models .
Transmission of Trypanosoma cruzi in peri-urban Arequipa occurs in a series of spatially-focal micro-epidemics . The oldest of these micro-epidemics in the community of Guadalupe began only around 20 years ago . By the time vector-borne transmission of T . cruzi was disrupted through insecticide application in 2004 , prevalence of human infection had reached 5% and was rapidly climbing . The relatively high prevalence of infection seems to conflict with the paucity of patients with Chagas cardiomyopathy in local hospitals , leading some to conclude that the strain of parasite in the region has low chronic pathogenicity . However , we have shown that most infections in Guadalupe likely occurred over a brief period of time prior to insecticide application . Our results provide support for a different explanation for the lack of late-stage Chagas disease in the city: the damage done by the vector and parasite may be unobserved because most individuals have yet to pass from the long asymptomatic period to symptomatic Chagas disease . Our findings do not disprove the hypothesis that the parasites circulating in Arequipa are less pathogenic than other strains . Previous studies suggest that T . cruzi strains in Arequipa are of limited genetic diversity , possibly due to a founder effect [37]–[38]—a finding , which if true , would be consistent with the results presented here . We have typed fifteen isolates of T . cruzi from Arequipa following methods described in [39] . The majority of strains , though not all , are T . cruzi type I , including isolates from communities neighboring Guadalupe [Vitaliano Cama , personal communication] . It is absolutely possible that the particular strains of T . cruzi I that predominate in Arequipa cause less chronic pathology . However , we emphasize that , based on the results from our models , the lack of late-stage Chagas cardiomyopathy in peri-urban Arequipa cannot be taken as evidence of a weaker parasite , and that in the absence of such evidence , preparations should be made for an increasing burden of clinical disease in the region over the coming years . Clinically , our findings may contribute to a re-evaluation of treatment guidelines for indeterminate Chagas disease that is currently underway [40]–[41] . Drug treatment is generally thought to be more effective early in the course of T . cruzi infection [11] . Currently many countries , Peru included , do not routinely offer treatment to patients over 15 years of age [1] . One justification for this policy is based on the assumption that age is a good surrogate for exposure to the parasite , and a reasonable surrogate of duration of infection among cases . That is , in the absence of evidence to the contrary , older individuals are assumed to have been carrying the parasite for many years , and therefore to be less likely to benefit from drug treatment . This assumption may be correct when transmission follows an endemic pattern . However , when transmission occurs in an epidemic , or multiple micro-epidemics , the assumption is incorrect . In peri-urban Arequipa and other epidemic situations , age alone is not a valid surrogate for exposure . Temporal and spatial information taken together give a better picture of how long an individual has been exposed to T . cruzi . Geographically , our finding that T . cruzi was first introduced in the southwest of Guadalupe makes sense . Guadalupe is a peninsula , surrounded on three sides by fields that provide no habitat to triatomine vectors , while the southwest of the community borders a large hillside of similar settlements with documented Chagas disease transmission [3] . Once T . cruzi was introduced into the southwest of Guadalupe it is clear that it did not spread house to house through a simple diffusion process . Instead the parasite was transported to other sites in the community , and there seeded new micro-epidemics of transmission . We can only guess at the mechanism of transport of the parasite . Animals , especially guinea pigs , are commonly gifted or traded both within and between communities; infected guinea pigs could have sparked new micro-epidemics in Guadalupe . Infected insects can also fly to establish new foci of transmission; while the chance that any one insect establishes transmission is very small [42] , given a large number of dispersing insects , establishment might occur occasionally . Individual-level data on human migration histories might allow us to study whether any infected individual was likely to have brought the parasite to Guadalupe from elsewhere [43] . Our finding of a very recent history of infections was robust to the precise number of micro-epidemics , and it is unlikely to be affected if these micro-epidemics were initiated through one mechanism rather than another . Generally , our approach is applicable to any situation in which the expected observation of an organism at sampling locations at a certain time is a function of the ( unknown ) site or sites of introduction of the organism into the system . The functional relationship between the expectation at a sampling site to the site ( s ) of introduction can be a simple function of distance , as we have used here , or can include information about the habitat between the sampling and introduction sites . The method can be informative when some prior information , on the advance of the disease agent or the likely sites of introduction , is available . The method is not likely to be informative in the absence of both . Our application of epicenter regression to T . cruzi transmission simplifies what is , in reality , a complex transmission cycle , and it is limited by the cross-sectional nature of the data . Our prior information on the speed of spread of the parasite came from empirical observations from another part of Arequipa; nevertheless the inherently stochastic nature of biological dispersal brings into question our ability to extrapolate our observations in one community to another [44] . Measurement of the Bayes' factor is also limited by errors of estimation [34] . Our primary goal was to estimate the timing of T . cruzi infections , as opposed to estimating the precise number of introductions of the parasite in the community . Although the four micro-epidemic model was strongly supported over the single-epicenter model , and substantially supported over the two-epicenter model , there was little difference in fit among models with 3-10 micro-epidemics . Other methods , such as reversible jump MCMC [45] , might be more efficient when the number of introductions is of primary interest [28] . Our study focused on a single peri-urban community . Since the completion of our study we have observed similar patterns of micro-epidemics of T . cruzi transmission in entomologic data across the southern half of the city of Arequipa ( unpublished data ) . In more rural areas outside the city , T . cruzi transmission was disrupted in the mid 1990s [46] . There are additional anecdotal reports of a lack of late-stage disease among individuals with Chagas disease residing outside of the city . A lack of late-stage disease among individuals infected many decades ago might provide evidence of lower chronic pathogenicity . In contrast , a finding of higher prevalence of late-stage disease among such individuals would provide direct evidence against this hypothesis . The traditional , endemic patterns of transmission of Trypanosoma cruzi by Triatoma infestans have been largely disrupted across southern South America through a concerted vector control program known as the Southern Cone Initiative [1] . Currently the initiative is challenged by vectors and parasites returning to areas previously under insecticide control [13] , [40]–[42] , and by new foci of transmission in and around urban centers [14] , [15] . The micro-epidemics of Chagas disease transmission we observed in Arequipa may be typical following emergence or re-emergence of the vector and parasite , rather than an anomalous pattern . Distinguishing between epidemic and endemic transmission will improve understanding of the dynamic relationship between prevalence of T . cruzi infection and the burden of clinical Chagas disease . | Chagas disease has become an urban problem in the city of Arequipa , Peru , yet there are very few people exhibiting severe symptoms of the disease . Severe symptoms often do not appear until decades after infection . To determine why so few people were exhibiting severe symptoms , we used a new method , epicenter regression , to trace the history of Chagas disease transmission in a community of Arequipa , Peru . Our findings suggest that transmission in Arequipa occurred through a series of small epidemics , the oldest of which began only around 20 years ago . These micro-epidemics infected nearly 5% of the community before the insect that carries Chagas disease , Triatoma infestans , was eliminated by insecticide application . Most human infections in the study community arose over a brief period of time immediately prior to the insecticide application . According to our findings , the severe symptoms of Chagas disease are expected to be absent from the community because of the short duration of infection with the parasite , even among older individuals with Chagas disease . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"medicine",
"public",
"health",
"and",
"epidemiology",
"epidemiological",
"methods",
"ecology",
"epidemiology",
"infectious",
"disease",
"epidemiology",
"biology",
"population",
"ecology",
"spatial",
"epidemiology"
] | 2011 | Retracing Micro-Epidemics of Chagas Disease Using Epicenter Regression |
Successful maintenance of cellular lineages critically depends on the fate decision dynamics of stem cells ( SCs ) upon division . There are three possible strategies with respect to SC fate decision symmetry: ( a ) asymmetric mode , when each and every SC division produces one SC and one non-SC progeny; ( b ) symmetric mode , when 50% of all divisions produce two SCs and another 50%—two non-SC progeny; ( c ) mixed mode , when both the asymmetric and two types of symmetric SC divisions co-exist and are partitioned so that long-term net balance of the lineage output stays constant . Theoretically , either of these strategies can achieve lineage homeostasis . However , it remains unclear which strategy ( s ) are more advantageous and under what specific circumstances , and what minimal control mechanisms are required to operate them . Here we used stochastic modeling to analyze and quantify the ability of different types of divisions to maintain long-term lineage homeostasis , in the context of different control networks . Using the example of a two-component lineage , consisting of SCs and one type of non-SC progeny , we show that its tight homeostatic control is not necessarily associated with purely asymmetric divisions . Through stochastic analysis and simulations we show that asymmetric divisions can either stabilize or destabilize the lineage system , depending on the underlying control network . We further apply our computational model to biological observations in the context of a two-component lineage of mouse epidermis , where autonomous lineage control has been proposed and notable regional differences , in terms of symmetric division ratio , have been noted—higher in thickened epidermis of the paw skin as compared to ear and tail skin . By using our model we propose a possible explanation for the regional differences in epidermal lineage control strategies . We demonstrate how symmetric divisions can work to stabilize paw epidermis lineage , which experiences high level of micro-injuries and a lack of hair follicles as a back-up source of SCs .
All cells within the body organize into distinct phylogenetic lineages . At the end of each lineage are the non-dividing , terminally differentiated cells . Usually these cells , such as neurons , adipocytes or muscle fibers , are highly specialized and endow tissues with their respective functions . The origin of all differentiated cells can be traced back to their progenitors , the so-called stem cells ( SCs ) [1 , 2] . Typically , tissue-specific SCs are long lasting and self-renewing ( i . e . at least 50% of SC progeny remain as SCs ) [2 , 3] . They also maintain high proliferative potential and assure lifelong lineage survival both under physiological steady-state conditions , and upon lineage depletion after injury or disease . These SC properties are vividly demonstrated by the experiments when the entire tissues are restored from just one grafted SC . For instance , new functional prostate tissue can reform following transplantation of one prostate SC under kidney capsule [4] . Similarly , transplantation of a single hematopoietic SC can reconstitute the entire bone marrow in lethally irradiated mice that would otherwise die from the inability to make new blood [5–7] . Another example is scarring alopecia , the type of baldness caused by the autoimmune attack on hair SCs—once SCs are lost , hairs can never grow again [8 , 9] . Successful maintenance and repair of cellular lineages critically depends on the fate decision dynamics of SCs upon division . Long-term steady-state maintenance of lineages requires that only 50% of all SCs progenies remain as SCs , and even slight shift in fate outcomes over time can lead to lineage exhaustion or uncontrolled expansion . For example , in the hair follicle , melanocyte SCs are more susceptible to exhaustion compared to epithelial SCs; therefore , commonly hair graying occurs faster than hair loss [10–12] . On the other hand , uncontrolled lineage expansion occurs upon myelodysplastic syndrome , a type of blood malignancy when mutated hematopoietic SCs increase their self-renewal rate to more than 50% . Over time , mutated SCs outcompete normal SCs , and accumulation of defective progeny cells leads to the loss of blood function and results in acute myeloid leukemia , a life-threatening complication of the myelodysplastic syndrome [13–15] . From these examples it is evident that tight control of SC fate decision dynamics is of paramount importance . In principle , steady-state maintenance of SCs can be achieved with three strategies [16] , see Fig 1: Assuming that individual cell division decisions are stochastic , at the tissue level , modes ( b ) and ( c ) result in neutral clone competition phenomenon , when some SC clones expand , some contract , while others stay constant [16 , 25–28] . Theoretically , either one of the strategies ( a-c ) can achieve lineage homeostasis . However , it remains unclear which strategies are more advantageous and under what specific circumstances , and what minimal control mechanisms are required to operate them . From the control point of view , SC fate can be decided either intrinsically or extrinsically [16] . Intrinsic control implies that the fate of daughter cells is determined by the signals present within the mother SC . Generally , this strategy is used only in specialized circumstances , for example during initial stages of embryonic development in C . elegans or in Drosophila neuroblasts [29 , 30] . Extrinsic control strategy , when the fate of daughter cells is determined at the time of SC division by the instructive environmental signals is more prevalent in adult tissues and organs . Depending on the source of the signal , autonomous and non-autonomous control mechanisms are recognized [16] . Non-autonomous mode is common in complex , highly regenerative tissues , such as hair follicles and bone marrow , and requires presence of the so-called niche cells [31–35] . These , often specialized cells of mesenchymal origin , generate the complex signaling micro-environment that regulates multiple aspects of lineage behavior both in time and in space , including but not limited to: ( i ) SC quiescence vs . proliferative activation [36–38] , ( ii ) symmetric vs . asymmetric fate of daughter cells [21 , 23 , 39] , ( iii ) progressive specification and differentiation of non-SC progenies [40–42] , ( iv ) switch between homeostatic vs . damage-response lineage behaviors [10 , 43 , 44] . For instance , skin hair follicle SCs activate cyclically . During the rest phase of the cycle , specialized mesenchymal niche supplies hair SCs with strongly inhibitory signals , keeping them quiescent . However , at the onset of the new hair growth phase , the same niche cells start to supply SCs with strongly activating signals . Furthermore , upon skin wounding , hair follicle SCs are guided to enter an “emergency” mode and transiently change their default fate: rather than sending progenies toward hair root , they send them upwards into the newly forming wound epidermis [43 , 45–47] . Generally , anatomic and signaling complexity of SC niches allows for spatial-temporal separation of SC activities and a plethora of lineage behaviors depending on the specific tissue , organ , as well as organismal needs . At the same time , such regulatory complexity makes it difficult to study the key homeostatic property of SCs—the fate decision control upon division . In addition to non-autonomous , niche-based SC control , some tissues display largely autonomous regulation . The latter mechanism implies regulation of SCs by their own daughter cells , both of SC and non-SC kind . Autonomous strategy is commonly seen in tissues with relatively simple microanatomy , such as stratified epidermis in mammalian skin [25–28 , 48–50] . Skin epidermis is vertically stratified and is arranged in successive layers of epithelial cells . SCs occupy lowermost basal layer , while their non-SC progeny , including terminally differentiated cells , move into the upper suprabasal layers . Depending on the anatomical location , epidermal thickness can range from just 2–3 layers , like in the back skin of mice , to several dozens , like the in mouse’s paw skin [51] . Multiple genetic tools for lineage tracing and signaling perturbations make mouse epidermis a particularly attractive experimental system for studying mechanisms of autonomous lineage control . Furthermore , because epidermal SCs switch their proliferative behavior in ontogenesis and upon wound healing , it enables studying aspects of plasticity of lineage control networks . In the present paper we address the questions of SC division symmetry by means of mathematical modeling . Our approach is based on that developed in [52 , 53] , and it contributes to the large theoretical literature on SC dynamics , see e . g . theoretical work of [54–57] , and a review in [58] . Some of the important areas of mathematical modeling in the context of SCs include discrete and continuous models in the context of carcinogenesis [59–71]; modeling of SC in the hematopoietic system [72–76]; deterministic modeling of two- , three- , and multi-compartmental systems under various regulation functions [77–81]; stochastic modeling of SC systems and the analysis of fluctuations [82–87] . The focus of the present study is to investigate how different division types contribute to lineage homeostasis/turnover . We provide analysis that allows to quantify the ability of two types of divisions ( symmetric and asymmetric ) to maintain homeostasis . What SC division strategy is better at maintaining the nearly constant population size ? Quoting [16] , “Asymmetric divisions are a key mechanism to ensure tissue homeostasis . In normal stem and progenitor cells , asymmetric cell division balances proliferation and self-renewal with cell-cycle exit and differentiation . ” At the intuitive level , it appears that asymmetric SC divisions should be associated with a more robust homeostatic maintenance . It can be argued that purely asymmetric SC divisions do not change the total number of SCs and therefore ensure the maintenance of a constant cell population , see e . g . [18] . It turns out however that tight homeostatic maintenance of the lineage ( including differentiated cells ) is not necessarily associated with purely asymmetric divisions . In this paper we show that asymmetric divisions can either stabilize or destabilize the lineage system , depending on the underlying control network . Once we establish the relationship between symmetry of cellular fate and lineage stability , we apply our computational model to biological observations in the context of mouse epidermal SCs and autonomous lineage control . It has been observed [25–28] that the proportion of symmetric divisions is higher in mouse paw epidermis compared to that of the tail and ear . By using our model we propose an explanation for this phenomenon .
Let us suppose that the lineage consists of two types of cells ( two compartments ) , SCs and daughter ( differentiated ) cells . Let us denote by I and J the current number of stem and daughter cells , respectively . The processes of division ( including differentiation/proliferation decisions ) and death are dictated by probabilities and rates defined in Table 1 ( a ) . Next , we need to quantify the control loops that exist in a given system . We assume that LI , J = L ( ϵI , ϵJ ) , DI , J = D ( ϵI , ϵJ ) , etc , where ϵ measures the strength of dependence of the probabilities and rates on the cell population numbers . It is convenient to introduce the continuous variables x = ϵI , y = ϵJ . To define the control network , we consider the partial derivatives of the rates and probabilities with respect to x and y , evaluated at the equilibrium . We will use the subscripts x and y to denote such partial derivatives , see Table 1 ( b ) . A two-compartment system is characterized by the following four derivatives: px , py , qx , and qy , which we call controls . To clarify the biological meaning of these parameters , consider the quantity py . If it is nonzero , it means that the probability of SC differentiation is controlled by the differentiated cell population . Moreover , if py < 0 , this means that the control is negative ( the more differentiated cells in the system , the less likely the SCs are to differentiate ) ; py > 0 means the existence of a positive control loop . The other three quantities can be interpreted in a similar manner . It was shown in [52] that at least two of the four controls must be nonzero in order for the system to have a stable homeostatic equilibrium . Minimal control systems are defined as models with a restricted number of nonzero controls , and are presented in Fig 3 . In the schematic , round cells and star-like cells represent stem and differentiated cells respectively . The first horizontal arrow in each diagram indicates the division decision , and the second horizontal arrow the differentiation decision . Arch-like positive and negative arrows depict the dependence of the two decisions on each population . For example , if a negative arrow originates at SCs and points at the divisions decision , this means that the divisions are negatively controlled by the SC numbers , qx < 0 ( see diagram #1 in Fig 3 ) . It was shown in [52] that with two compartments , there are two distinct minimal control systems with two controls , and three systems with three controls ( see also S1 Text ) . The first two models ( #1 and #2 ) in Fig 3 are the only two systems that can be stable in the presence of no more than two controls . The other three models ( #3–5 in Fig 3 ) are the only three irreducible three-control systems , that is , they cannot be reduced to models #1 or #2 by setting one of the controls to zero . While from the point of view of stability , all five of the networks are possible , further biological considerations are required to identify which control network is relevant for a particular tissue . Some of those considerations may include the matching of various moments of compartment sizes with the observations , robust recovery dynamics , etc . In the particular case study considered in this paper ( mouse epidermis ) network #5 appears to be the most relevant , as explained below . Next we demonstrate how by varying the proportion of symmetric vs asymmetric SC divisions , one can change homeostatic properties of the system in the context of models #1–5 . We will focus on the analysis of variance of the cell populations . A relatively small variance indicates stable , robust homeostasis . A large variance increases the probability of extreme events , such as extinction or growing out of control . By using stochastic analysis ( see the Methods Section ) we can calculate the variance of the number of SCs , Var[I] , and the variance in the number of differentiated cells , Var[J] , as functions of the parameters . In particular , it is possible to determine how these quantities depend on the four controls ( Table 1 ( b ) ) and the frequency of symmetric SC divisions , S . It turns out that in two out of five control systems in Fig 3 , the variance increases with S . Namely , in systems #2 and #3 , Var[J] increases with S , and in addition , in #3 Var[I] also increases with S ( in #2 , the variance of SC numbers is independent of the symmetry ) , see Eqs ( 33 ) and ( 34 ) . Therefore , in these two control systems , purely asymmetric divisions are optimal from the viewpoint of minimizing fluctuations in cell numbers at homeostasis . The opposite result is observed for systems #1 , #4 , and #5 . There , purely symmetric divisions turn out to be the optimal choice . In those three systems , the variance of differentiated cell numbers is a decreasing function of S , and in addition , in #4 , the variance of SC numbers is also a decreasing function of S , see Eqs ( 32 ) , ( 35 ) and ( 36 ) . In these three qualitatively different control networks , symmetric divisions are associated with the most stable homeostatic state . Next , we demonstrate this theoretical finding by numerical simulations . The results reported in the previous section hold for any functional forms of controls . Here we illustrate these findings by considering two specific examples . Some technical details about the simulation setup are provided in S1 Text . Recall that ϵ measures the strength of control of the various processes by the cell population , and x = ϵI , y = ϵJ; we further denote Δ = qx py − qy px , and B = 2L* S* ( px − py ) − qy , where the partial derivatives with respect to x and y are defined in Table 1 ( b ) and the star indicates that the quantity is evaluated at the equilibrium . The quantities Δ and B appear in the expressions for the variances ( Methods ) . Throughout this section , we will assume SI , J takes some constant value c , where 0 < c ≤ 1 . Although SI , J is not necessarily constant , its derivatives do not enter the stability conditions or expressions for population variances ( as explained in the Methods section ) , and therefore we make the simplest assumption on this function . Below are two examples , where in order to illustrate the theory numerically , we chose some specific functional forms for the controls . Mammalian epidermis develops through several distinct stages . In mice , one cell layer-thick epidermis first appears at embryonic day E8 . 5 [88–90] . Over the course of next few days and till day E13 . 5 , primordial epidermal cells divide strictly symmetrically along the horizontal plane of the skin , and this contributes to the rapid expansion of epidermal surface in synchrony with rapid growth of the embryo body [89 , 91] . Starting from E13 . 5 , fully symmetric SC-generating strategy switches to a mixed mode , consisting of both symmetric and asymmetric divisions [48–50 , 88] . Asymmetric divisions that generate one basal SC and one suprabasal non-SC comprise approximately 70% of all divisions in day E15 . 5 mouse embryos [50] . The 30% symmetric divisions at that time are likely necessary for epidermis to add more SCs and to grow in absolute area as embryo continues to enlarge . What is the division strategy in adult epidermis that stopped expanding and reached its steady state ? Several recent studies demonstrate that adult mouse epidermis is maintained via a mixed division mode , with basal SCs undergoing all three division types: asymmetric and two types of symmetric divisions . The support for the mixed mode is provided by the low-dose induction lineage tracing experiment . In this experiment , an inducible genetic system is used to randomly and permanently label few scattered basal epidermal SCs and all of their progenies . Labeling of just a few basal SCs assures that most of the marked SCs will be far from one another to prevent fusion of their progeny populations . Because over time , the total number of labeled clones decreased , while some of the remaining clones expanded in size , this supports symmetric divisions: loss of clones results from divisions generating two non-SCs , while expansion of clone sizes results from divisions generating two SCs [27 , 28] . Interestingly , the exact ratio of division types appears to differ depending on the anatomical location . For example , in the mouse tail epidermis the ratio of asymmetric to SC + SC symmetric to non-SC + non-SC symmetric divisions is approximately 80%:10%:10% [26 , 28] and in the mouse ear it is 78%:11%:11% [27] . However , in the footpad epidermis , it appears to change in favor of symmetric divisions—60%:20%:20% [25] . We would like to find an explanation of this increased fraction of symmetric SC divisions in the footpad compared to the ear/tail epidermis . While little is known about the signaling aspects of SC fate determination in adult epidermis , the available published biological data point toward non-intrinsic mechanism . Mitotic spindle analysis indicates that , in contrast to embryonic epidermis , only 3% of basal SCs in adult mouse epidermis divide strictly vertically [28] . Moreover , daughter cell fate selection appears to depend on the dynamic signaling inputs generated in the basal layer: ( i ) short-range acting WNT ligands promote basal SC division , and ( ii ) long-range Dkk signal drives cell differentiation , i . e . non-SC identity [25 , 92] . This type of autocrine/paracrine signaling from SCs closely matches our minimal control system #5 , see Fig 3 . Indeed in system #5 , SCs exert positive control on their own division ( matching the role of WNT ligands ) as well as positive control on differentiation decision ( matching the role of Dkk1 ) . Also , control network #5 requires positive regulation of lineage differentiation by differentiated cells . Therefore , we speculate that other signaling events , in addition to WNT/Dkk1 , are likely involved in regulating epidermal lineage homeostasis . Indeed , multiple signaling pathways , including Notch , TGFβ , IKK , Ras/MAPK , PI3K and p63 have been implicated in regulation of epidermal proliferation and differentiation ( reviewed in [93] ) . Of these , Notch signaling is of a particular interest . Notch signaling is active predominantly in suprabasal epidermal cells , where it drives their differentiation [94–97] , matching third signaling event in the minimal control system #5 . What differences can account for the increase in symmetric divisions in the footpad epidermis ? While to date , this issue has not been studied experimentally , the three most notable distinctions of the footpads from other body sites are: We will use mathematical modeling to propose an answer to the following question: Why does footpad epidermis have a larger proportion of symmetric divisions ? We will explore the above three differences to see if any of them favors symmetric divisions , in the context of stable homeostatic tissue maintenance .
In this paper we studied the role of symmetric and asymmetric divisions in the maintenance of tissue homeostasis . We have designed a general stochastic model that can be solved analytically to quantify how the amount of variation in the population size depends on various system parameters . We have shown that depending on the precise “wiring” of the controls in a control network , symmetric divisions can either stabilize or destabilize the system . In particular , among 5 minimal control loops identified in a two-compartment system [52] , in two of them increasing the percentage of symmetric divisions will increase fluctuations , and in the remaining three it will decrease fluctuations , leading to an increased stability . In the context of our study , mouse epidermis is an ideal model system for examining principles and mechanism of autonomous lineage control: Using our model , we studied the role of divisions symmetry in mouse epidermis . In particular , we focussed on the data on the percentage of symmetric divisions in different anatomic regions of the skin . While in the ear and tail epidermis , 20–22% of all divisions are symmetric , in the footpad epidermis this percentage increases to 40% . We showed that in the control system that best characterizes the epidermal lineage regulation , increasing the percentage of symmetric divisions enables the cell population to respond to environmental changes associated with micro-injuries . Conversely , decreasing the percentage of symmetric divisions allows to minimize the relative fluctuation size in cell populations in the presence of an exogenous source of SCs . This is relevant for the specific case-study of mouse epidermis . On the one hand , the footpad is characterized by a higher level of micro-injuries compared to the ear and tail epidermis . Indeed , footpad skin is exposed to a variety of mechanical stresses , including friction from running , scratching , burrowing , fighting . All of these likely increase the probability of micro-injury to suprabasal epidermal compartment , and elevate the rate of cell loss from the lineage as compared to other , better protected anatomical areas , such as back skin , tail , ears . Our analysis suggests that the increased percentage of symmetric divisions in the footpad may be an adaptation to stabilize the tissue that faces the highest rate of micro-injuries from friction and abrasions . On the other hand , the footpad is characterized by the lack of hair follicles . The presence of hair follicles in the other regions such as ear and tail serves as an extra source of SCs . Our model shows that in the presence of such a source , asymmetric divisions are optimal from the point of view of keeping the size fluctuations low . Our work adds to the discussion of the role of symmetry in the maintenance and dynamics of SC lineages . In [3] , cellular strategies are considered in the context of homeostasis maintenance . It is stated that the balance between cell proliferation , differentiation and death can be achieved in two ways . On the one hand , it can be “obtained at the level of a single SC , which divides strictly into a new SC and a progenitor . ” On the other hand , this “balance can also be achieved at the level of the SC population . Some SCs might be lost due to differentiation or damage , whereas others divide symmetrically to fill this gap . ” The following question is raised in [3]: Why should mechanisms of tissue maintenance so often lean toward symmetric SC self-renewal ? One possible answer comes from the ability of all symmetrically-dividing SCs to efficiently respond to injury and correct for lineage depletion . It however could be argued that the symmetric divisions are turned on in response to a sudden significant loss of cells , while the asymmetric division strategy can be employed in the course of normal homeostasis . In [100] we addressed a possible role of symmetric SC divisions as a cancer prevention mechanism . It was argued that symmetric divisions may slow down the accumulation of double-hit mutants , thus delaying the onset of many cancers , which depend on the inactivation of several tumor suppressor genes . In the present paper , we study cell division patterns from a different prospective , by looking at the maintenance of healthy tissues at homeostasis . In general , each trend or strategy that has evolved in an organism , has been subject to a large number of selection pressures . In this paper we focus on only one type of selection pressure , namely , the pressure to keep the fluctuations down for an increased tissue stability ( in [100] we focused on the selection pressure to delay the generation of cancerous mutations ) . In the case of footpad , both of these favor symmetric divisions , at least in the context considered in the two studies . There are however many other aspects of the evolutionary process that are not taken into account here . One class of factors not included in the model is the true anatomical constraints of the epidermis . The basal compartment of the epidermis is limited in size and thus crowded , and the signaling mechanism regulating SC vs . differentiated cell decision-making is subject to physical constraints . Specifically , in real-life situations , the divisions of epidermal SCs can be truly symmetric if the mitotic axis is parallel to skin surface ( i . e . both daughter cells remain in the basal layer ) . However , published data show that in adult epidermis , mitotic axis is randomly determined ( horizontal , vertical and anywhere in-between ) and commonly as the result of this one daughter cell is forced into the suprabasal layer , where it immediately experiences low WNT and high Dkk1 signaling that promotes its differentiation . Thus , these types of SC divisions are “forced” to be asymmetric . This mechanism alone will likely considerably limit the number of truly symmetric divisions . This can explain why 100% of symmetric divisions shown in our model is not realistic , since the model does not account for the real-life anatomic constrains of the skin . The resulting solution found in the real-life epidermis is a trade-off between the anatomical and physical constraints , and possible evolutionary pressures , such as the ones described here . In the example worked out in this paper we show that symmetric divisions are more important for the footpad epidermis than they are for the ear and the tail . As an important future direction , a model with a more realistic 3D representation of cells in their niches , that describes the alignment and the geometry of SC divisions , could be created to combine the trends found here and anatomical considerations . A step in that direction has been made in [101] , where a bi-compartmental SC niche was considered . In such a niche , one compartment is at the interface with the differentiated progeny and the other compartment is spatially separated from the differentiated cells . Further complexity can be added by explicitly modeling a spatially distributed system . There are several other extensions of this work that are natural . While the mouse epidermis can be described as a two-compartmental lineage system , other tissues are characterized by a larger number of cell types of different degrees of differentiation ( prominently , hematopoietic lineage ) . An extension of the current formalism to multi-compartment systems can be done by using the methodology developed in [102] . Further , the current model can only handle near-equilibrium situations . A different approach is required to study significant injuries and wound healing .
A stochastic model of cell population renewal is considered ( see [52 , 53] ) . The cells are subject to the following changes in a Poisson process with an infinitesimally small time-increment , Δt: A deterministic model that captures these events can be expressed as the following system of ordinary differential equations: x ˙ = L S ( 1 - P ) - L S P = L S ( 1 - 2 P ) , ( 13 ) y ˙ = 2 L S P + L ( 1 - S ) - D , ( 14 ) where x and y refer to the numbers of stem and differentiated cells , and L , P , and S are all functions of x and y . The stochastic description in terms of the Kolmogorov forward equation is given by the following equation for the variable φI , J ( t ) , the probability to find the system in state ( I , J ) at time t: φ ˙ I , J = φ I + 1 , J - 2 L I + 1 , J - 2 S I + 1 , J - 2 P I + 1 , J - 2 + φ I - 1 , J L I - 1 , J S I - 1 , J ( 1 - P I - 1 , J ) + φ I , J - 1 L I , J - 1 ( 1 - S I , J - 1 ) + φ I , J + 1 D I , J + 1 - φ I , J ( L I , J + D I , J ) , ( 15 ) where the processes of the right hand side are presented in the same order as they appear in the list above . Note that system Eqs ( 13 and 14 ) is the “macroscopic law” obtained at the zeroth order of the “linear noise approximation” [52 , 53] . We are interested in deriving equations for the mean values of the cell populations and their variances . To do this , we first define the steady states of the system , ( i0 , j0 ) , by the following equations ( which are obtained by solving Eqs ( 13 ) and ( 14 ) ) : L i 0 , j 0 = D i 0 , j 0 = L * , P i 0 , j 0 = 1 2 , S * = S i 0 , j 0 . ( m i x e d d i v i s i o n s s t e a d y s t a t e ) ( 16 ) L i 0 , j 0 = D i 0 , j 0 = L * , S i 0 , j 0 = 0 , P * = P i 0 , j 0 . ( p u r e l y a s y m m e t r i c d i v i s i o n s s t e a d y s t a t e ) ( 17 ) Both equilibria are characterized by a balance between divisions and deaths ( the first equation in Eqs ( 16 ) and ( 17 ) ) . In the first ( mixed divisions ) equilibrium , the probability of differentiation events is equal to the probability of proliferation events , thus ensuring that the expected change in the number of SCs is zero . The first two equations in Eq ( 16 ) define the equilibrium population sizes i0 and j0 . The fraction of symmetric divisions , SI , J , does not influence the solution for i0 and j0 , but , as shown below , can affect its stability properties and the size of fluctuations in the system . The second ( purely asymmetric ) equilibrium is attained if the fraction of symmetric divisions can be made zero . The population sizes are determined by the first two equations in Eq ( 17 ) , and the probability of differentiations , formally defined by the last equation , becomes irrelevant at equilibrium . Below we focus on the mixed divisions steady state . Calculations pertaining to steady state Eq ( 17 ) can be found in S1 Text . The methodology presented here is based on the assumption of weak dependencies of the functions LI , J , DI , J , etc on their variables . It is developed in [52] and justified rigorously in [53] . Let us use the symbol ZI , J to denote any of the functions LI , J , PI , J , DI , J , and SI , J . Suppose that we can represent the functions ZI , J near the equilibrium as ZI , J = Z ( ϵI , ϵJ ) , where the parameter ϵ ≪ 1 defines the weakness of the dependence . It is convenient to denote x = ϵI , y = ϵJ , i = I − i0 , j = J − j0 , then we can expand the functions ZI , J around the steady state in Taylor series: Z I , J = Z i 0 , j 0 + z x i + z y j + 1 2 ( z x x i 2 + z y y j 2 + 2 z x y i j ) + ⋯ , ( 18 ) where the subscripts x and y denote the partial derivative of the function with respect to its argument , evaluated at ( i0 , j0 ) , and zx = Zx ϵ , zxx = Zxx ϵ2 , etc . In this description , while constants Zx = O ( 1 ) , Zxx = O ( 1 ) , etc are all of order one , all the first derivatives zx , zy contain a factor ϵ , and all the second derivatives zxx , zxy , zyy contain a factor ϵ2 . Define φ ˜ i , j = φ i + i 0 , j + j 0 = φ I , J , and Z ˜ i , j = Z i + i 0 , j + j 0 = Z I , J , then Eq ( 15 ) can be reformulated as: φ ˜ ˙ i , j = φ ˜ i + 1 , j - 2 L ˜ i + 1 , j - 2 S ˜ i + 1 , j - 2 P ˜ i + 1 , j - 2 + φ ˜ i - 1 , j L ˜ i - 1 , j S ˜ i - 1 , j ( 1 - P ˜ i - 1 , j ) + φ ˜ i , j - 1 L ˜ i , j - 1 ( 1 - S ˜ i , j - 1 ) + φ ˜ i , j + 1 D ˜ i , j + 1 - φ ˜ i , j ( L ˜ i , j + D ˜ i , j ) . ( 19 ) Using expansion Eq ( 18 ) in Eq ( 19 ) , we can derive the moment equations for this system . In what follows , we use the following notations for the moments: X α β = ∑ i , j i α j β φ ˜ i , j ( t ) . ( 20 ) Multiplying Eq ( 19 ) by i and by j , performing a summation in the two indices , and keeping only the highest order terms in ϵ , we obtain equations for the first moments in steady-state: 0 = - 2 L * S * ( p y X 01 + p x X 10 ) , ( 21 ) 0 = ( 2 L * S * p x + q x ) X 10 + ( 2 L * S * p y + q y ) X 01 . ( 22 ) For the second moments we have: 0 = ( S * l x + L * s x ) X 10 + ( S * l y + L * s y ) X 01 - 4 L * S * ( p y X 11 + p x X 20 ) + L * S * , 0 = - ( S * l x + L * s x + 2 L * S * p x ) X 10 - ( S * l y + L * s y + 2 L * S * p y ) X 01 + 2 L * S * [ p x X 20 - p y X 02 + ( p y - p x ) X 11 - 1 / 2 ] + q y X 11 + q x X 20 , 0 = [ S * ( 4 L * p x + l x ) + l x + d x + L * s x ] X 10 + [ S * ( 4 L * p y + l y ) + l y + d y + L * s y ] X 01 + ( 4 L * S * p x + 2 q x ) X 11 + ( 4 L * S * p y + 2 q y ) X 02 + L * ( 2 + S * ) . ( 23 ) Solving this system , we can obtain the expressions for the means and variances: E [ I ] = X 10 + i 0 = i 0 , E [ J ] = X 01 + j 0 = j 0 ; V a r [ I ] = X 20 - X 10 2 , V a r [ J ] = X 02 - X 01 2 . The highest order terms for the variances are given by V a r [ I ] = K x 4 B Δ , V a r [ J ] = K y 4 B Δ , ( 24 ) where we defined the quantities: Δ = q x p y - q y p x , ( 25 ) B = 2 L * S * ( p x - p y ) - q y , ( 26 ) K x = 2 L * S * Δ + q y 2 + 8 L * 2 S * p y 2 , ( 27 ) K y = 2 L * ( 2 + S * ) Δ + q x 2 + 8 L * 2 S * p x 2 . ( 28 ) Details of stability analysis are given in S1 Text . It follows that mixed division steady state is stable as long as Δ > 0 and B > 0; constants Kx and Ky are always positive quantities . Increasing Δ and B makes the system more robust by decreasing the variation of population sizes . A Mathematica file is provided in S1 File that calculates result Eq ( 24 ) symbolically . The equilibrium values for the numbers of stem and differentiated cells are unaffected by the presence of asymmetric divisions , as illustrated by Eq ( 16 ) . On the other hand , the probability of symmetric divisions , S* , can influence two important properties of the SC system: ( a ) stability of the equilibrium and ( b ) the size of fluctuations ( the amount of variance ) , which is related to the robustness of homeostatic control . | Stem cells have long been associated with their ability to divide asymmetrically , when one daughter cell retains stem cell properties of the parent cell , while the other daughter cell becomes more mature and loses its stemness . Recent findings , however , point at the existence of an alternative , symmetric division strategy for stem cells in mammalian tissues . Here we ask the question: what engineering design principles might be responsible for optimization of stem cell division strategies ? Although simple intuition may suggest that asymmetric divisions are better suited for stable maintenance of cell population numbers , our analysis shows that asymmetric divisions can also destabilize the system , depending on the particular “wiring” of the control loops that govern cellular fate decision making . We apply our theory to one particular unresolved question in mouse epidermis studies—why symmetric division percentage in paw epidermis is twice as high as that in ear and tail ? The answer may be related to the fact that paw epidermis lacks hair follicles ( a source of stem cells available in other types of skin ) , and it is also more injury-prone: symmetric divisions help stabilize skin in face of physical stresses from running , digging , grooming , and fighting—things that mice do with their paws . Thus , our theory offers quantitative explanations for the observed designs in stem cell lineages . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [] | 2015 | The Role of Symmetric Stem Cell Divisions in Tissue Homeostasis |
SUMOylation is associated with epigenetic regulation of chromatin structure and transcription . Epigenetic modifications of herpesviral genomes accompany the transcriptional switch of latent and lytic genes during the virus life cycle . Here , we report a genome-wide comparison of SUMO paralog modification on the KSHV genome . Using chromatin immunoprecipitation in conjunction with high-throughput sequencing , our study revealed highly distinct landscape changes of SUMO paralog genomic modifications associated with KSHV reactivation . A rapid and widespread deposition of SUMO-2/3 , compared with SUMO-1 , modification across the KSHV genome upon reactivation was observed . Interestingly , SUMO-2/3 enrichment was inversely correlated with H3K9me3 mark after reactivation , indicating that SUMO-2/3 may be responsible for regulating the expression of viral genes located in low heterochromatin regions during viral reactivation . RNA-sequencing analysis showed that the SUMO-2/3 enrichment pattern positively correlated with KSHV gene expression profiles . Activation of KSHV lytic genes located in regions with high SUMO-2/3 enrichment was enhanced by SUMO-2/3 knockdown . These findings suggest that SUMO-2/3 viral chromatin modification contributes to the diminution of viral gene expression during reactivation . Our previous study identified a SUMO-2/3-specific viral E3 ligase , K-bZIP , suggesting a potential role of this enzyme in regulating SUMO-2/3 enrichment and viral gene repression . Consistent with this prediction , higher K-bZIP binding on SUMO-2/3 enrichment region during reactivation was observed . Moreover , a K-bZIP SUMO E3 ligase dead mutant , K-bZIP-L75A , in the viral context , showed no SUMO-2/3 enrichment on viral chromatin and higher expression of viral genes located in SUMO-2/3 enriched regions during reactivation . Importantly , virus production significantly increased in both SUMO-2/3 knockdown and KSHV K-bZIP-L75A mutant cells . These results indicate that SUMO-2/3 modification of viral chromatin may function to counteract KSHV reactivation . As induction of herpesvirus reactivation may activate cellular antiviral regimes , our results suggest that development of viral SUMO E3 ligase specific inhibitors may be an avenue for anti-virus therapy .
Dynamic chromatin structure regulation by post-translational protein modifications modulates the accessibility of DNA and consequently the transcription of genes . Small ubiquitin-like modifier ( SUMO ) modification in epigenetic regulation of chromatin states has been intensively studied . SUMO modification of specific transcription factors or chromatin remodeling proteins , in most cases , is associated with repressive complex formation and a silencing role in transcription regulation [1 , 2] . Moreover , SUMOylation promotes de novo targeting of chromatin proteins to heterochromatin [3] . However , increasing evidence suggests that SUMO modification may also be associated with positive regulation of transcription [4] . These data highlight the complexity of chromatin-associated SUMO in gene expression modulation . To uncover the global epigenetic role of SUMO in transcription regulation , one study performed in yeast showed that SUMO associates with promoters of constitutively active and inducible genes . SUMO recruitment to inducible promoters during activation is required to shut-off inducible genes after elimination of the activating signal [5] . Unlike yeast , that contains only a single SUMO protein , human cells have three protein-conjugating isoforms . These isoforms include SUMO-1 , which is conjugated to proteins as a monomer , and highly related SUMO-2 and SUMO-3 ( SUMO-2/3 ) , which are known to form high molecular weight polymers on proteins [6 , 7] . Though earlier studies have pinpointed some important differences between SUMO-1 and SUMO-2/3 [8 , 9] , the functional specificity of SUMO isoforms in global epigenetic regulation of gene expression is just beginning to be uncovered . Several recent reports , including ours , using Chromatin Immunoprecipitation-Sequencing ( ChIP-seq ) in combination with transcriptome analysis of RNA-sequencing ( RNA-seq ) have comprehensively characterized the SUMO-1 and SUMO-2/3 genomic landscape and their global role in transcription regulation in human cells [10–12] . Neyret-Kahn et al showed that both SUMO-1 and SUMO-2/3 are strongly associated at promoters of actively transcribed genes and SUMOylation is responsible for restraining their expression during cell proliferation [12] . In contrast , Liu et al showed that SUMO-1 is associated with promoters of active genes and directly activates their transcription during interphase of the cell cycle [11] . The association of SUMO with the active histone mark H3K4me3 was identified in both studies . However , similar epigenetic alterations between SUMO paralogs were observed under physiological stimuli tested [12] . Interestingly , our recent report showed that SUMO-2/3 , when compared with SUMO-1 modifications around cellular promoter regions was significantly increased in B cells during Kaposi’s sarcoma associated herpesvirus ( KSHV ) reactivation . This enrichment prevents the activation of host genes during viral reactivation [10] . These findings indicate the existence of differential roles of SUMO paralogs in regulating chromatin and transcription during a stress response , such as virus infection . KSHV , also known as human herpesvirus type 8 ( HHV-8 ) , is a γ-herpesvirus associated with Kaposi’s sarcoma ( KS ) , a tumor of endothelial origin , and primary effusion lymphomas ( PEL ) , a B-cell lymphoma [13] . Similar to all herpesviruses , the KSHV lifecycle has distinct latent and lytic phases . KSHV can maintain a tightly latent infection in the majority of infected tumor cells . However , a small population of infected cells continues to undergo spontaneous lytic replication [14] . Establishing latency enables KSHV to evade host immune surveillance , establish persistent life-long infections and induce tumorigenesis [15] . Lytic reactivation is not only required for the proper spread of KSHV infection , but is also a prerequisite for the maintenance of a population of latently infected cells and KSHV pathogenesis [16] . After infection , KSHV genomic DNA in the host cells forms a chromatin-like structure . The latent–to-lytic switch involves global remodeling of viral chromatin from the heterochromatin to the euchromatin state . Recently , several studies including ours have begun to document the histone modification profiles on KSHV viral chromatin during viral infection or reactivation [17–20] . These studies indicate that activating and repressive histone marks are differentially located on KSHV latent and lytic genomes and these marks are involved in transcriptional regulation of viral genes . The bivalent states of chromatin marks on the KSHV-replication and transcription activator ( K-Rta ) promoter maintains a poised state for K-Rta rapid expression in response to reactivation stimuli . Latent genes possess only activating histone marks . Most early genes have either activating or repressive histone marks . The difference between chromatin modifications may contribute to the temporal expression of KSHV genes during reactivation from latency . However , one area that is complex and remains largely unknown is the function of additional post-translational modifications , such as SUMOylation , in regulating the viral epigenome . Analogous to ubiquitylation , SUMOylation is a multistep enzyme cascade including SUMO E1 activating enzyme ( SAE1/SAE2 ) , SUMO E2 conjugating enzyme ( Ubc9 ) , and SUMO E3 ligase ( i . e . , PIAS family , RanBP2 , and Pc2 ) . However , unlike ubiquitylation , an E3 ligase is not essential for SUMO conjugation , but instead provides specificity for SUMO modification . The SUMO interaction motifs ( SIMs ) in SUMO E3 ligases are responsible for its SUMO paralog specificity [21 , 22] and structure analysis shows potentially different specificity of SIMs toward SUMO paralogs [23] . This underlying complexity was increased by the identification of the downstream consequences of non-covalent interaction with effectors via SIMs , providing an additional interaction platform for selectively recruiting SUMO-1 or SUMO-2/3 specific SIM-containing proteins . As mentioned earlier , SUMO modification of chromatin proteins may formulate the fine-tuning of chromatin structure and transcription regulation . Like other DNA viruses , KSHV has evolved different mechanisms to directly or indirectly manipulate the SUMO machinery to advance their own growth ( reviewed in [24–26] ) . Most interestingly , we recently identified KSHV lytic protein K-bZIP as a SUMO E3 ligase with specificity toward SUMO-2/3 [27] . This unique specificity suggests the potential preferential usage of SUMO-2/3 by KSHV to globally modulate its epigenome and gene expression during lytic reactivation . Hence , KSHV represents an ideal model system to study the functional specificity of SUMO-2/3 in regulating global epigenetic changes and transcription . Moreover , this specificity makes KSHV an attractive model system to help distinguish SUMO paralog-specific effects in epigenetic regulation of transcription . In this study , we demonstrate SUMO-2/3 specific chromatin modification enrichment on the KSHV genome during lytic reactivation . RNA-seq results show a positive correlation between viral lytic gene transcription activation and SUMO-2/3 enrichment on the viral genome upon reactivation . SUMO-2/3 knockdown results in increased transcription of viral lytic genes located in high SUMO-2/3 enrichment regions of the KSHV genome . Interestingly , the overlaid SUMO-2/3 binding pattern and different epigenetic marks showed a positive correlation between SUMO-2/3 with the active histone mark H3K4me3 and a negative correlation between SUMO-2/3 with the repressive histone mark H3K27me3 in the latent viral genome . In addition , a negative correlation between SUMO-2/3 enrichment and H3K9me3 marks in viral lytic genomes during the early phase of KSHV reactivation was identified . These results suggest that SUMO-2/3 modification plays an essential role in fine-tuning genomic regions with active chromatin structure but not repressive heterochromatin regions . Since KSHV encodes a SUMO-2/3 specific E3 ligase , K-bZIP , we analyzed K-bZIP binding on the KSHV genome by a ChIP assay . A significant increase of K-bZIP binding in SUMO-2/3 enrichment region after KSHV reactivation was found . In addition , we used the BAC16 template to generate a new recombinant BACmid , BAC16 K-bZIP-L75A , which contains a SIM domain mutant of K-bZIP that has lost its SUMO E3 ligase activity . Cell lines stably transfected with BAC16 established latency and could produce infectious virus upon reactivation . The K-bZIP SUMO E3 ligase dead mutant showed increased expression level of viral lytic transcripts located in high SUMO-2/3 enriched regions and produced significantly more infectious viruses . These data strongly suggest SUMO-2/3 specific epigenetic regulation of viral gene expression during reactivation .
The doxycycline ( Dox ) -inducible TREx-BCBL-1 , 3x Flag- and 3x His-tagged K-Rta BCBL-1 ( TREx-F3H3-K-Rta BCBL-1 ) and Myc-His-tagged K-Rta BCBL-1 ( TREx-MH-K-Rta BCBL-1 ) cell lines were maintained in RPMI 1640 containing 15% FBS , 50 μg/ml blasticidin and 100 μg/ml Zeocin or 100 μg/ml hygromycin ( Invitrogen , Carlsbad , CA ) . TREx-BCBL-1 , TREx-F3H3-K-Rta BCBL-1 and TREx-MH-K-Rta BCBL-1 cells were induced with 0 . 2 μg/ml Dox . The SUMO-2/3 inducible knockdown cell line was generated in previous study [10] . Briefly , the shSUMO-2 and shSUMO-3 in pLenti4-H1/TO-shRNA plasmid were introduced into TREx-F3H3-K-Rta BCBL-1 cells by transduction and the stable TREx-F3H3-K-Rta-shSUMO-2/3 BCBL-1 cell line was maintained as described for TREx-F3H3-K-Rta BCBL-1 cells and supplemented with 300 μg/ml Zeocin . Induction of SUMO knockdown and K-Rta expression was confirmed by immunoblotting analysis . The SUMO-1 and SUMO-2 overexpression cell lines were generated by transfection using plasmids expressing Flag-SUMO-1 or Flag-SUMO-2 into TREx-MH-K-Rta BCBL-1 cells . Cells were selected for 21 days by 200 μg/ml G418 ( AMRESCO ) and purified by Ficoll . Expression of Flag-tagged SUMO-1 and SUMO-2 were tested by immunoblotting using anti-Flag antibody . iSLK-Puro cells were maintain in DMEM containing 10% FBS , 250 μg/ml G418 and 1 μg/ml puromycin ( Invitrogen ) . 293T cells were maintained in DMEM containing 10% FBS . ChIP was performed using the protocol from Dr . Farnham’s laboratory ( http://genomics . ucdavis . edu/farnham ) . Briefly , chromatin DNA from control and Dox-treated TREx-F3H3-K-Rta BCBL-1 cells were harvested . Chromatin DNA from 1 x 107 cells was used per antibody for each ChIP assay . ChIP grade anti-SUMO-1 ( Abcam , ab32058 ) and anti-SUMO-2/3 ( Abcam , ab3742 ) specific rabbit polyclonal antibodies , as well as rabbit non-immune serum IgG ( Alpha Diagnostic International ) , were used for the ChIP assays . 50 ng of ChIPed DNA suspended in 30 μl of ddH2O was used for ChIP-seq library preparation following the protocol from Illumina . Size-selected ( 400 bp ) DNA fragment libraries were used for paired-end high throughput sequencing on Illumina Genome AnalyzerII . The ChIP-Seq data was aligned with the KSHV genome build by Avadis NGS ( Strand Scientific Intelligence , San Francisco , CA ) . Approximately 1~3 x 106 reads were mapped for each sample after filtering and quality control ( QC ) . In this study , we used the target region QC detection method of Avadis NGS to delineate the SUMO-1 and SUMO-2/3 binding patterns . ChIP DNA was verified for successful IP by SYBR Green-Based real-time qPCR using CFX connect real-time PCR detection system ( Bio-Rad , Richmond , CA ) . Specific primer sets were designed around the potential binding sites . Primer sequences were listed in S1 Table . Total RNA was prepared from TREx-F3H3-K-Rta BCBL-1 cells harvested at 0 , 12 and 24 hours after Dox treatment using TRIzol ( Invitrogen , Carlsbad , CA ) according to the manufacturer’s instructions . RNA-seq was carried out at the Sequencing Core of National Research Program for Genomic Medicine at the National Yang-Ming University using an Illumina Genome AnalyzerII . Sequencing reads were processed as described previously [10] . In this study , the sequence reads that did not align with hg19 were mapped to KSHV . The transcript abundances were estimated in reads per kilobase of transcript per million mapped reads ( RPKM ) by Avadis NGS . Differential gene expression was analyzed by comparing RPKMs from each sample and verified using real-time RT-qPCR . 2 μg of total RNA was reverse-transcribed using Oligo-d ( T ) 18 and SuperScript III first-strand synthesis system ( Invitrogen ) . qPCR was carried out according to the manufacturer's protocol ( iQ SYBR Green Supermix , Bio-Rad ) . Primer sequences were listed in S2 Table . To mutate K-bZIP Leu 75 to Ala within BAC16 using recombineering , we first generated a targeting vector containing a KpnI/HindIII fragment of the KSHV genome that included partial K-Rta coding region and the complete K-bZIP coding region . The primer 5’-GGTCTGTGAAACGGTCATTGACGCTACAGCGCCTTCCCAAA-3’ containing the L75A mutation flanked by 14~15 bp homology were used to target mutagenesis of K-bZIP at Leu 75 . After confirmation of mutation , the FRT-flanked kanamycin cassette was inserted into the SalI site in between the K-Rta and K-bZIP coding region . A linear fragment for homologous recombination was generated by digestion targeting vector with KpnI and HindIII . The DNA fragment was gel purified by RECOCHIP ( Takara ) and electroporated into induced ( recombination + ) SW105 harboring BAC16 . Kanamycin resistant colonies were selected and the insertion of the targeting cassette was confirmed by PCR . For kanamycin cassette removal , positive colonies were inoculated in LB containing arabinose and incubated overnight at 32°C . Kanamycin-sensitive clones were further screened by PCR . After successful removal of the kanamycin cassette in K-bZIP-L75A mutants , clones were verified by sequencing . To make a wild-type ( WT ) revertant , K-bZIP-L75A in BAC16 was replaced by WT K-bZIP using the recombineering protocol as described above . KpnI and HindIII cleaved BAC16 DNA was separated on 1% agarose gel and visualized by ethidium bromide staining . The DNA in the gel was transferred to NC membrane using downward alkaline transfer . A K-bZIP probe was radiolabeled with [α-32P] dCTP ( Perkin Elmer ) using Rediprime II random primer labeling kit ( GE Healthcare , UK ) . DNA blots were hybridized overnight at 65°C with rotation . After washing , the blots were imaged using X-ray film ( Kodak , Rochester , NY , USA ) . iSLK-Puro cells were transfected with 2 μg of BAC16 DNA using FuGENE HD ( Roche ) . Forty-eight hours after transfection , the cells were expanded to a 15 cm Petri dish and selected by hygromycin ( 600 μg/ml ) . After three weeks of selection , the hygromycin-resistant and GFP positive colonies were picked and pooled to establish the iSLK-Puro-BAC16 K-bZIP-WT , -WT rev and -L75A cell lines . iSLK-Puro-BAC16 cell lines were maintained as described for iSLK-Puro cells and supplemented with 300 μg/ml hygromycin . To assess viral production , supernatants from control and Dox-induced TREx-F3H3-K-Rta BCBL-1 and SUMO-2/3 knockdown TREx-F3H3-K-Rta-shSUMO-2/3 BCBL-1 cells were collected before and after 48 hours treatment . KSHV virion DNA was prepared using QIAamp MinElute Virus Spin kits as described previously [28] . Quantification was performed by real-time qPCR using a TaqMan probe targeting orf73 ( LANA ) [29] .
To distinguish the genome-wide distribution of SUMO paralog conjugation on KSHV chromatin and determine their changes during viral reactivation , we conducted chromatin immunoprecipitation ( ChIP ) assays in combination with high-throughput next generation sequencing ( ChIP-seq ) , a gold-standard method for identifying the genome-wide sites of epigenetic marks . For this , we used the well-characterized KSHV-infected primary effusion lymphoma ( PEL ) cell line , TREx-F3H3-K-Rta BCBL-1 , that expressed a Dox inducible K-Rta protein whose expression switches KSHV from latent to lytic phase [30] . The induction of K-bZIP during KSHV reactivation by Dox induced K-Rta overexpression or by PKC agonist TPA treatment was first compared . A comparable level of K-bZIP expression was identified in cells receiving 0 . 2 μg/ml Dox or 20 ng/ml TPA treatment ( S1 Fig ) . Therefore , 0 . 2 μg/ml of Dox was used to treat K-Rta inducible TREx-F3H3-K-Rta BCBL-1 cells for our entire study to prevent induction of excessively high levels of K-bZIP by high dose Dox treatment . After Dox treatment for 12 hours , the global expression of SUMO-1 and SUMO-2/3 and successful induction of K-Rta were first confirmed by immunoblotting ( S2A Fig ) . The ChIP experiments were carried out using chromatin prepared from non-induced ( latency phase ) and Dox-induced ( lytic phase ) TREx-F3H3-K-Rta BCBL-1 cells . Next generation sequencing was then performed to measure the chromatin binding of SUMO-1 and SUMO-2/3 from a single run of ChIP assay . As shown in Fig 1 , the ChIP-seq result revealed a comparable binding level of SUMO-1 and SUMO-2/3 throughout the KSHV latent genome . Interestingly , SUMO-2/3 enrichment levels were significantly increased on most parts of the KSHV genome at 12 hours post induction ( hpi ) of viral reactivation when compared to the changes observed in SUMO-1 levels ( Fig 1 ) . However , it should be noted that a significant higher enrichment of SUMO-1 than SUMO-2/3 was observed in a few regions of the viral genome , including the promoter of latent gene orf73 ( LANA ) . The potential role of this SUMO-1 specific enrichment in regulation LANA gene expression is interesting and a subject for further inquiry . However , in this study , we focused on studying the global enrichment of SUMO-2/3 in regulating viral gene expression and reactivation . Surprisingly , when compared with our previous result of H3K9me3 modification on the KSHV latent genome ( 0 hpi ) [17] , we noticed that two distinct viral genomic regions , which contain a high level of the repressive heterochromatin mark H3K9me3 , displayed no increase of SUMO-2/3 occupancy at 12 hpi . To further investigate whether SUMO-2/3 enrichment at 12 hpi is indeed negatively correlated with H3K9me3 , we performed a ChIP-on-chip experiment by hybridizing DNA from H3K9me3-associated chromatin at 12 hpi ( Fig 1 ) with a KSHV genome tiling array we previously designed [17] . We first aligned the two binding profiles at 0 and 12 hpi on the KSHV genome and then analyzed the correlation by Pearson Correlation analysis . Applying this measure , a medium negative correlation ( r = -0 . 3~-0 . 5 ) was identified between SUMO-2/3 modification and H3K9me3 at 12 hpi ( r = -0 . 3 ) but not at 0 hpi ( r = -0 . 1 ) . These data suggest that SUMO-2/3 specific modification appears to have an epigenetic regulatory function separate from H3K9me3 . For further confirmation of the ChIP-seq results , we selected four genes from each of two viral genomic regions , one representing the region of high SUMO-2/3 enrichment with low H3K9me3 mark ( orf46 , K-bZIP , K8 . 1 and orf52 ) ( Fig 1 , red box ) and the other one representing the region of high H3K9me3 mark with little SUMO-2/3 enrichment ( orf19 , orf20 , orf23 and orf25 ) ( Fig 1 , blue box ) during KSHV reactivation . Consistent with the ChIP-seq results , real-time qPCR data showed that the genes in SUMO-2/3 highly enriched region tested here displayed significant enrichment of SUMO-2/3 after viral reactivation when compared with the non-induced control cells ( Fig 2 , upper panel ) . In contrast , the genes in high H3K9me3 mark region showed little increase in SUMO-2/3 modification ( Fig 2 , lower panel ) . The reproducibility of this assay was obtained with independent repeat ChIP-qPCR experiment ( S2B Fig ) . Expression of Flag-tagged SUMO-1 and SUMO-2 followed by ChIP using Flag antibody confirmed this phenomena ( S3 Fig ) . These data indicate that the KSHV genome undergoes SUMO-2/3-specific modification following reactivation . The discovery of the negative correlation between SUMO-2/3 and H3K9me3 prompted us to further explore the association of SUMO-1 and SUMO-2/3 with different histone marks . We compared the binding pattern of SUMO-1 and SUMO-2/3 before and after viral reactivation from this study with the previously published binding profiles of various chromatin marks [20] ( Table 1 ) . Pearson correlation showed that SUMO-2/3 is medium positive correlated with H3K4me3 ( r = 0 . 3 ) and medium negative correlated to H3K27me3 ( r = -0 . 3 ) at 0 hpi . A previous study of global SUMO modification on the human genome also showed that the majority of SUMO paralogs are highly correlated with the active histone mark H3K4me3 . Moreover , it was demonstrated that SUMO strongly associated at active promoters , with SUMOylation acting to restrain the gene expression [12] . In line with this study , the positive correlation of SUMO-2/3 with the active histone mark H3K4me3 and its negative correlation of repressive mark H3K27me3 in KSHV latent genomes suggest that SUMO-2/3 may be involved in maintaining a repressive environment in euchromatic regions of the viral episome to restrain viral gene expression during latency . The compelling correlation between SUMO-2/3 and activating histone marks on the KSHV genome prompted us to speculate that SUMO-2/3 may play a role in repressing lytic promoters associated with activating histone marks during latency . To study this , we transiently transduced a lentiviral vector expressing inducible shRNA for SUMO-2/3 into TREx-BCBL-1 [31] . Successful induction of partial SUMO-2/3 knockdown was detected at 24 hours ( S4A and S4B Fig ) . However , SUMO-2/3 knockdown did not induce the expression of viral lytic genes located in the high SUMO-2/3 enrichment region in latent KSHV infected BCBL-1 cells ( S4C Fig ) . This result indicates that knockdown SUMO-2/3 alone is not sufficient to induce KSHV lytic gene expression or viral reactivation from latency . Consistent with our ChIP-on-chip data , SUMO-2/3 showed medium negative correlation to H3K9me3 ( r = -0 . 3 ) at 12 hpi . Interestingly , no correlation was found between SUMO-2/3 enrichment with any other histone marks after KSHV reactivation . Moreover , no statistically significant correlation was found in SUMO-1 with histone marks on both latent and lytic KSHV genomes . To further explore if the negative correlation of SUMO-2/3 enrichment with heterochromatin mark H3K9me3 is also true on viral promoter regions , a number of promoters with high SUMO-2/3 enrichment at 12 hpi from this study were again compared with the previous study [20] . Among the 36 viral promoters that have high SUMO-2/3 enrichment ( 4-fold enrichment ) during KSHV reactivation , 17 ( 47% ) , 15 ( 42% ) and 14 ( 39% ) also contain H3K4me3 , AcH3 ( H3K9/K14ac ) , and H3K27me3 marks , respectively . Consistent with results shown in Fig 1 and Table 1 , SUMO-2/3 enrichment during viral reactivation is largely devoid of H3K9me3 . Only 5 ( 14% ) of the SUMO-2/3 enriched promoters have the H3K9me3 mark . Since H3K9me3 marks are associated with heterochromatin , these results again imply that SUMO-2/3 may be focused on tagging and regulating viral promoters in euchromatin sites during viral reactivation . In order to gain insight into transcriptional regulation by SUMO-2/3 during KSHV reactivation , an RNA-Seq assay was performed using total RNA purified from TREx-F3H3-K-Rta BCBL-1 cells before and after K-Rta induction . Several viral lytic genes , such as K7 and PAN , showed high expression in control cells . This may due to the spontaneous reactivation of KSHV in a small population of BCBL-1 cells . Viral gene expression changes revealed by this assay show that SUMO-2/3 enrichment was preferentially located at gene regions that show a higher level of transactivation during KSHV reactivation ( Fig 3 , red box ) . Interestingly , a relatively low level of viral gene expression before and after K-Rta induction was identified in the KSHV genomic region with high H3K9me3 mark ( Fig 3 , blue box ) . As mentioned earlier , the positive correlation between SUMO-2/3 enrichment and viral gene transactivation suggests that SUMOylation on the viral genome may be required for gene shut-off after viral lytic reactivation . To explore this idea , a SUMO-2/3 inducible knockdown TREx-F3H3-K-Rta-shSUMO-2/3 BCBL-1 cell line was used [10] . A time course analysis showed the earliest time point we were able to detect the partial knockdown of SUMO-2/3 was at 24 hours ( S5 Fig ) . Therefore , a 24 hour time point was used for further study . To study the role of SUMO-2/3 in transcriptional regulation , protein and mRNA samples were collected from control ( TREx-F3H3-K-Rta BCBL-1 ) and SUMO-2/3 knockdown TREx-F3H3-K-Rta-shSUMO-2/3 BCBL-1 cells before and after Dox ( 0 . 2 μg/ml ) treatment for 24 hours . Western blot analysis confirmed the successful induction of K-Rta , expression of K-bZIP and knockdown of SUMO-2/3 at 24 hours after Dox treatment ( Fig 4A ) . The viral gene expression in control ( TREx-F3H3-K-Rta BCBL-1 ) and SUMO-2/3 knockdown TREx-F3H3-K-Rta-shSUMO-2/3 BCBL-1 cells were then compared . Again , the viral genes representing the high SUMO-2/3 enrichment region ( orf46 , K-bZIP , K8 . 1 and orf52 ) and the high H3K9me3 mark region ( orf19 , orf20 , orf23 and orf25 ) were selected for RT-qPCR study . As shown in Fig 4B , SUMO-2/3 knockdown resulted in a higher induction of the expression of viral genes located in the high SUMO-2/3 enrichment region , but not the genes located in high H3K9me3 mark region . There are also KSHV genomic regions with modest SUMO-2/3 increase that are enriched in H3K9me3 , such as the region containing orf34 to orf39 ( Fig 1 ) . RT-qPCR results showed that SUMO-2/3 is also essential for restraining the transcription of genes such as orf35 , orf37 and orf39 in this region ( S6A Fig ) . This data suggest that SUMO may function at a level higher than histone marks , for example , by recruiting different transcription regulators . However , the latent transcript region containing orf73 , orf72 , and orf71 that contains both SUMO-1 and SUMO-2/3 enrichment does not show a significant difference in gene induction after SUMO-2/3 knockdown ( S6B Fig ) . This result suggests that the regulatory role of SUMO-2/3 in viral latent gene expression may differ from that of lytic genes . The differential role of SUMO paralogs in regulating KSHV latent genes is of interest and worth further exploration . These results together indicate that SUMOylation enrichment on the viral genome during reactivation is indeed required for diminution of the expression of select viral lytic genes after reactivation . To confirm whether global SUMO-2/3 knockdown results in a corresponding decline in SUMO-2/3 enrichment on the KSHV genome during reactivation , another ChIP assay was performed using control ( TREx-F3H3-K-Rta BCBL-1 ) and SUMO-2/3 knockdown TREx-F3H3-K-Rta-shSUMO-2/3 BCBL-1 cells . ChIP-qPCR result showed that SUMO-2/3 knockdown abolishes SUMO-2/3 enrichment on the high SUMO-2/3 enrichment region of KSHV genome , but has little effect on the high H3K9me3 mark region ( Fig 4C ) . This result again demonstrates the transcriptional regulation of orf46 , K-bZIP , K8 . 1 and orf52 located in the high SUMO-2/3 enrichment region is dependent on SUMO-2/3 . Taken together , these data indicate that KSHV uses SUMO-2/3 modification as an epigenetic modification to dampen lytic gene expression in regions where the H3K9me3 heterochromatin mark is low , in order to modulate transactivation during the lytic cycle . To determine if knockdown SUMO-2/3 led to changes in H3K9me3 mark or additional active and repressive chromatin marks in the KSHV genome , ChIP assays were performed in control ( TREx-F3H3-K-Rta BCBL-1 ) and SUMO-2/3 knockdown TREx-F3H3-K-Rta-shSUMO-2/3 BCBL-1 cells using antibodies specific for different histone marks . Knockdown of SUMO-2/3 did not change any of the interrogated histone marks in the KSHV genome ( S7 Fig ) . Consistent with our previous result ( S6A Fig ) , these data indicate that SUMOylation may modify protein binding at a level above the deposition of specific histone marks . Enhancement of viral lytic gene expression by SUMO-2/3 knockdown may lead to an increase in viral replication and elevate levels of virus production . To examine the level of KSHV virion production in SUMO-2/3 knockdown cells , the control ( TREx-F3H3-K-Rta BCBL-1 ) and the inducible SUMO-2/3 knockdown TREx-F3H3-K-Rta-shSUMO-2/3 BCBL-1 cells were treated with 0 . 2 μg/ml Dox for 48 hours , a condition that we can consistently detect relatively low ( ~ 2- to 3-fold virion induction ) but significant induction of KSHV virion production in control TREx-F3H3-K-Rta BCBL-1 cells . The successful induction of K-Rta , expression of K-bZIP and knockdown of SUMO-2/3 was first assessed by immunoblotting ( Fig 5A ) . The supernatants collected from different treatments were used for virion purification and the level of virion-associated DNA was determined using real-time TaqMan qPCR amplification . Consistent with our prediction , SUMO-2/3 knockdown significantly increased viral production by ~2-fold over control cells ( Fig 5B ) . Together , these data indicate that SUMO-2/3 modification may specifically target the KSHV genome to create a silencing chromatin environment ready for diminution of lytic gene expression and viral replication after induction . Our findings strongly suggest that SUMO-2/3 may play a major and critical role in regulating viral gene silencing after transactivation . The next question is how KSHV controls the SUMO-2/3 modification on its genome during lytic reactivation . Interestingly , our recent report demonstrated a unique mechanism by which KSHV modulates SUMOylation via expressing a viral lytic SUMO E3 ligase , K-bZIP [27] . This viral SUMO E3 ligase has specificity towards SUMO-2/3 . Together with this finding , we speculated that during KSHV reactivation , KSHV expresses the lytic protein K-bZIP and simultaneously conjugates SUMO-2/3 to viral genome regions with low heterochromatin marks . To address this , a ChIP assay was performed using a K-bZIP specific antibody and chromatin prepared from TREx-F3H3-K-Rta BCBL-1 cells before and after Dox induction for 12 hours . Again , primer pairs for orf46 , K-bZIP , K8 . 1 and orf52 representing the high SUMO-2/3 enrichment region and orf19 , orf20 , orf23 and orf25 representing the high H3K9me3 mark region were used to determine the chromatin binding of K-bZIP using real-time qPCR . Consistent with our hypothesis , a significant higher increase of K-bZIP binding in SUMO-2/3 enrichment region after KSHV reactivation was observed ( Fig 6 ) . Note that some K-bZIP binding on its own promoter was observed in non-induced cells . ChIP data reported in Ellison et al . 2009 stated that the K-bZIP promoter was the most enriched target of K-bZIP following its overexpression [32] . Thus , the interaction between K-bZIP and its promoter may be high and readily detectable when compared with other KSHV promoters . A small population of BCBL-1 cells continuously emerge from latency into lytic replication resulting in the expression of K-bZIP in a small fraction of cells , thus some binding of K-bZIP to its promoter is observed in control cells . Together , these results suggest that K-bZIP may be the SUMO E3 ligase catalyzing the addition of SUMO-2/3 to the KSHV genome during reactivation . To uncover the potential functional role of the SUMO E3 ligase activity of K-bZIP in regulating KSHV gene expression during reactivation in a virus context , we generated a SUMO E3 ligase dead mutant of K-bZIP in BAC16 , a KSHV BAC clone generated from Dr . Jung’s laboratory [33] . The leucine 75 ( L75 ) in the SIM domain of K-bZIP was mutated to alanine in the targeting vector using site-directed mutagenesis [27] . This K-bZIP-L75A mutant was introduced into the wild-type ( WT ) KSHV genome in BAC16 bacmid by recombineering ( Fig 7A ) . Recombination was carried out using a SW105 transformant containing BAC16 . A targeting vector containing WT K-bZIP was also used to replace the K-bZIP-L75A allele on BAC16 to generate revertant viruses for use as WT control ( WT rev ) . The BAC16 constructs were first checked by PCR analysis . The positive clones were then digested with KpnI or HindIII and analyzed by agarose gel electrophoresis ( Fig 7B; left panel ) and subsequently by Southern blot analysis ( Fig 7B; right panel ) . Both KpnI and HindIII digestion analysis shows a ~1 . 8 Kb band shift in BAC16 intermediate and return to similar position of BAC16 parental band upon FRT-mediated removal of kanamycin selection cassette ( Fig 7B ) . Sequencing was used to confirm successful mutagenesis and no unexpected changes were detected . After confirmation , WT rev and L75A mutant viruses were reconstituted in iSLK cells , a cell line that inducibly expresses K-Rta by Dox treatment . Following hygromycin selection for 21 days , comparable GFP expression was observed in iSLK-Puro-BAC16 cells ( Fig 7C ) . The expression of KSHV latent protein LANA was analyzed by immunoblotting , indicating stable propagation of BAC16 in mammalian cells ( S8A Fig ) . The KSHV genome copy number in stable iSLK-Puro-BAC16 K-bZIP-WT rev and -L75A mutant cell lines was analyzed by qPCR using orf19 and orf20-specific primer pairs . Similar relative copy numbers of KSHV genome were observed from both primer pairs ( S8B Fig ) and the values were used to normalize all the following real-time RT-qPCR and virus production quantification experiments . To evaluate the potential role of K-bZIP SUMO E3 ligase , iSLK cells harboring different recombinant KSHV-BACmids were induced with Dox for 24 and 48 hours for gene expression and virus production analysis , respectively . Total protein lysates were collected for immunoblotting analysis to assess the successful induction of K-Rta and expression of K-bZIP ( Fig 8A ) . Next , we measured the accumulation of KSHV orf46 , K-bZIP , K8 . 1 and orf52 mRNA , representing the potential SUMO-2/3-regulated genes and orf19 , orf20 , orf23 and orf25 mRNA , representing non-SUMO-2/3-regulated genes . A significant higher increase in expression of the SUMO-2/3-regulated viral genes was found in iSLK-Puro-BAC16 K-ZIP-L75A mutant compared with K-bZIP-WT rev ( Fig 8B; upper panel ) . Consistent with the significant increase of K-bZIP transcript by approximately 5-fold in iSLK-Puro-BAC16 K-ZIP-L75A mutant , a higher level of mutant K-bZIP protein was also observed during K-Rta-induced viral reactivation ( Fig 8A ) . In line with our previous data , this result indicates that K-bZIP may mediate the SUMOylation of viral promoters in the low H3K9me3 region which results in a diminution of viral gene expression after reactivation . Unexpectedly , viral genes in the high H3K9me3 mark region showed defects in transactivation in the iSLK-Puro-BAC16 K-bZIP-L75A mutant ( Fig 8B; lower panel ) . Since no SUMO enrichment was identified in the high H3K9me3 region , SUMO E3 ligase activity of K-bZIP should not influence the transactivation of genes in this region . As shown in our previous report , a high affinity direct interaction between K-bZIP and H3K9me3 was found [17] . The loss of gene transactivation in the high H3K9me3 region early after reactivation ( 24 hours ) may relate to the direct binding of K-bZIP to this region and enhance late gene expression by recruiting SUMOylated transcription factors in a SIM-dependent manner . It should also be noted that K-bZIP is known to be required for lytic DNA replication [34–36] , and that DNA replication induces high level expression of late genes , including orf19 , orf20 , orf23 and orf25 . The SIM domain of K-bZIP may be required for KSHV replication and that is why the SIM mutation results in less expression of orf19 , orf20 , orf23 and orf25 . To determine if the K-bZIP SIM plays a role in its binding on the viral genome , ChIP assays of K-bZIP by using chromatin prepared from non-induced and Dox-induced iSLK-Puro-BAC16 K-bZIP-WT rev and -L75A mutant were performed . Consistent with our ChIP result of K-bZIP occupancy from TREx-F3H3-K-Rta BCBL-1 cells ( Fig 6 ) , a significant higher increase in K-bZIP binding in SUMO-2/3 enrichment region compared with high H3K9me3 mark region was also observed ( S9 Fig ) . However , the binding of K-bZIP to viral promoters shows no significant differences between the WT rev and L75A mutant after KSHV reactivation ( S9 Fig ) . This result suggests that the K-bZIP SIM is not involved in K-bZIP binding on the KSHV genome but may influence its recruitment of other SUMOylated proteins . Since KSHV lytic reactivation is accompanied by transcriptional reprogramming , we extended the observations to virus production . Supernatants from non-induced and Dox-induced ( 48 or 72 hours ) iSLK-Puro-BAC16 K-bZIP-WT rev and -L75A cells were collected and used for detection of virion-associated DNA using real-time TaqMan qPCR amplification or to infect 293T cells , respectively . In agreement with our previous findings , K-bZIP-L75A mutant showed significantly higher viral production ( Fig 8C ) . Consistently , K-bZIP-L75A mutant also exhibited higher numbers of recombinant KSHV ( rKSHV ) infected 293T cells by ~3-fold over than WT rev KSHV ( Fig 8D and 8E ) . In addition , when examined on a single-cell basis , K-Rta-expressing cells show more lytic protein expression in iSLK-Puro-BAC16 K-bZIP-L75A cells compared to WT rev during reactivation ( S10 Fig ) . To confirm the higher viral production in iSLK-Puro-BAC16 K-bZIP-L75A cells , the infection experiment was repeated by including the original BAC16 construct . The parental BAC16 shows similar virus production as K-bZIP-WT rev ( S11 Fig ) . K-bZIP-L75A mutant showed higher virus production compared to both parental and WT rev KSHV ( S11 Fig ) . The data from SUMO-2/3 knockdown and K-bZIP SUMO E3 ligase-dead mutant strongly suggest that SUMO E3 ligase activity of K-bZIP is responsible for the SUMO-2/3 enrichment on the KSHV genome during viral reactivation . To confirm this hypothesis , another ChIP assay was performed with chromatin prepared from iSLK-Puro-BAC16 K-bZIP-WT rev and -L75A mutant using anti-SUMO-2/3 antibody . Consistent with all data above , a significant increase of SUMO-2/3 was observed on the promoters of orf46 , K-bZIP , K8 . 1 and orf52 but not of orf19 , orf20 , orf23 , and orf25 . Moreover , the SUMO-2/3 enrichment on orf46 , K-bZIP , K8 . 1 and orf52 promoters was completely abolished in K-bZIP SUMO E3 ligase-dead mutant ( Fig 9 ) . Consistent with Fig 8A , a higher level of K-bZIP protein was observed during K-Rta-induced viral reactivation ( Fig 9A ) . In line with our data of SUMO-2/3 knockdown ( S7 Fig ) showing that SUMO modification is involved in gene transcription regulation without changing histone marks , K-bZIP SUMO E3 ligase-dead mutant did not change the H3K9me3 pattern ( S12 Fig ) . Together , these data show that conjugation of SUMO-2/3 on the KSHV genome by the viral SUMO E3 ligase K-bZIP plays an essential role in alleviating transactivation of viral genes and production of virus .
One distinct feature of herpesviruses is that there are two phases , latent and lytic , in their lifecycle . Establishment of latency is a common property for herpesvirus to evade host immune responses and establish life-long infection . In addition to viral propagation , lytic reactivation has also been found to be required for maintaining herpesviral persistence [37] . This is substantiated by mouse experiments showing that passive transfer of anti-lytic cycle antibody , but not anti-latent cycle antibody , into B-cell deficient mice decreased the number of cells harboring latent virus [37] . This idea is further supported by the finding that the antiviral drug cidofavir reduces the frequency of latently infected cells [38] . For KSHV , both latent and lytic cycles are essential for not only its persistent infection but also for its tumorigenesis ( reviewed in [39 , 40] ) . Therefore , maintaining an exquisite balance between latency-to-lytic cycle switch is important for persistent viral infection . The SUMO-2/3 enrichment identified here during viral reactivation demonstrates an unexpected SUMO function in the diminution of active KSHV lytic gene expression . Although the benefit of KSHV suppressing itself during reactivation is unclear , single cell analysis of cells undergoing reactivation have noted that only 20% of K-Rta positive cells also expressed the late gene K8 . 1 suggesting the existence of additional commitment factors required for K-Rta positive cells to advance through complete reactivation [41] . Although K-Rta is necessary and sufficient for lytic reactivation , it has been considered an inefficient reactivating switch , subject to positive and negative regulation by viral and cellular factors [42] . Thus , there is precedent for cells putatively undergoing reactivation ( i . e . , K-Rta positive ) to enter sub-lytic , abortive or full lytic pathways . Thus , we speculate SUMO-2/3 deposition may influence viral pathway fate by effects on other viral gene products feeding back to the level of K-Rta . Alternatively , as posited by studies in yeast [5] , SUMOylation may function in promoter clearance after each round of activated transcription , allowing another cycle of transcription to proceed if sufficient activator signal is present . Our results ( Figs 4B , 4C , 8B and 9B ) are consistent with this mechanism; failure to efficiently clear a promoter , through SUMO-2/3 knockdown or K-bZIP mutation would be expected to result in prolonged transcriptional activation and elevated viral reactivation . Lack of SUMO-2/3 enrichment at high H3K9me3 regions ( Fig 1 ) would account for the differential promoter responses observed . Though disruption the balance of latency-to-lytic cycle toward lytic activation might lead to increased infectivity and viral loads , it may also result in host immune activation and viral clearance . Moreover , current anti-herpes viral drugs only target lytic replicating viruses . This concept has been explored as a potential therapy for herpesvirus . Treatment strategies consisting of lytic induction of Epstein–Barr virus ( EBV ) using doxorubicin and gemcitabine , or both EBV and KSHV by bortezomib , followed with chemo- or radiotherapy has been described [43 , 44] . In addition , the report of KSHV reactivation following knockdown of Tousled-like kinases [45] suggests that development of small molecule inhibitors targeting these kinases may act as lytic inducers . A somewhat similar approach ( the so-called “shock and kill” strategy ) that shock latent proviruses with pharmacological agents such as histone deacetylase ( HDAC ) inhibitors and kill emergent viruses with combined anti-retroviral therapy and/or host cytolytic T cells is currently under evaluation as a means to eradicate latent proviruses present in patient HIV reservoirs ( reviewed in [46] ) . An inhibitor of K-bZIP E3 ligase activity might function similar to L75A and increase lytic replication that favors host clearance by making KSHV visible to the immune system and anti-viral drugs . Understanding the molecular mechanisms that regulate the KSHV latent-to-lytic switch not only holds the key to developing effective therapy for KSHV but also for other oncogenic herpesvirus . In latent phase , KSHV genome persists as a transcriptionally silent extrachromosomal episome resembling heterochromatin . During lytic phase , many regions of the viral genome adopt a state of euchromatin organization and almost all viral genes are transcribed in a temporally ordered manner . Epigenetic modifications of herpesvirus chromatin very likely play key roles in regulation viral gene expression as well as controlling the switch between latency and lytic replication . This notion is supported by the fact that viral lytic reactivation can be induced by inhibitors of DNA methyltransferases ( DNMTs ) [47] and HDACs [48 , 49] . Several recent studies comprehensively analyzed the epigenetic marks , including DNA methylation and histone modifications , in the KSHV latent and lytic genome [18–20] . Upon de novo infection , a quick transition from euchromatin mark to heterochromatin mark was detected in KSHV genomes [19] . In latent KSHV genomes , both activating as well as repressive histone marks were identified at certain viral loci . This “bivalent” state of chromatin generated a poised state of repression that can likely be quickly reverted to fully active state upon induction of viral lytic cycle under stimuli [18] . During viral reactivation , the activating marks located on genomic regions encoding the immediate-early ( IE ) genes were increased whereas the repressive H3K27me3 mark was decreased [20] . Together , these results highlight the importance of epigenetic modifications in the regulation of the KSHV lifecycle . Although significant effort has been devoted to find key epigenetic marks and their roles in the establishment of KSHV latent and lytic chromatin , a detailed understanding of the post-translational modifications involved in mediating the switch between KSHV latent infection and lytic replication is still largely unknown . SUMO modification is a post-translational modification that not only modulates the function of many transcription factors but also the chromatin organization by recruiting chromatin remodeling enzymes , including DNA and histone modification enzymes , to regulate gene expression . Global analysis of SUMOylation in epigenetic regulation has been done in different eukaryotic cells and indicates that SUMOylation can either restrain [10 , 12] or promote [11] the expression of actively transcribed genes under different environmental stimuli . These results indicate that depending on the stimuli , SUMO paralogs are able to create repressive or active chromatin states to regulate host gene expression . It comes as no surprise that herpesvirus KSHV has evolved ways to modulate the SUMO machinery to epigenetically regulate viral chromatin to benefit its lifecycle . Indeed , our recent report showed that KSHV encodes a lytic SUMO-2/3 specific E3 ligase , K-bZIP [27] . Moreover , Cai et al showed that the KSHV latent protein LANA also contains a SUMO-2 specific SIM that is essential for the recruitment of the SUMOylated chromatin remodeling protein KAP-1 which aids in the maintenance of viral latency [50] . These studies prompted us to study how SUMOylation regulates viral gene expression during reactivation and whether there is preferential SUMO paralog usage . ChIP-seq results showed similar global SUMO-1 and SUMO-2/3 binding patterns on KSHV latent , but not lytic , viral genomes . SUMO-2/3 , compared with SUMO-1 , was significantly increased across the KSHV genome during reactivation ( Fig 1 ) . The significant enrichment of SUMO-2/3 on the viral genome provided us with an opportunity to uncover the epigenetic role of SUMO-2/3 . The role of SUMO-2/3 in regulating viral gene expression during KSHV reactivation was revealed by RNA-seq conducted in TREx-F3H3-K-Rta BCBL-1 cells before and after KSHV reactivation ( Fig 3 ) . The higher induction of viral gene expression in high SUMO-2/3 enriched regions during viral reactivation allows for two potential hypothesis; ( 1 ) SUMO-2/3 activates viral gene transcription during reactivation and ( 2 ) SUMO-2/3 restrains viral gene expression after reactivation . To elucidate the functional role of SUMO modification in transcription regulation , SUMO-2/3 knockdown experiments were conducted . Reduction of SUMO-2/3 binding at a high SUMO-2/3 enrichment region on the KSHV genome was first confirmed in SUMO-2/3 knockdown TREx-F3H3-K-Rta-shSUMO-2/3 BCBL-1 cells ( Fig 4C ) . During reactivation , the expression of viral genes in the high SUMO-2/3 enrichment region was activated to a higher level relative to a low SUMO-2/3 enrichment region after SUMO-2/3 knockdown ( Fig 4B ) . Together with previous reports [10–12] , these results indicate the preferential usage of SUMO-2/3 by KSHV in restraining viral gene expression during reactivation . Histone marks have long been studied and believed to be essential in maintaining chromatin structure and therefore manipulating gene expression . Though global SUMO modifications in epigenetic regulation have begun to be studied in eukaryotic cells , the correlation between histone marks and SUMO modifications has not yet been elucidated . Due to the relatively low complexity of the viral genome and the availability of genome-wide histone modification landscapes from KSHV , we were able to compare the genome-wide distribution of SUMO modification with different histone marks [20] on KSHV latent and lytic genomes . Our SUMO DNA binding patterns were overlaid with different epigenetic marks on the KSHV genome . Correlation with histone marks was only identified in chromatin binding of SUMO-2/3 , but not SUMO-1 ( Table 1 ) . An interesting finding is that our ChIP-seq data showed that during lytic reactivation , SUMO-2/3 levels increased mostly on KSHV genomic regions devoid of H3K9me3 , thus negatively correlating with H3K9me3 occupied late gene-rich regions ( Fig 1 ) . The genome-wide correlation analysis also showed that SUMO-2/3 enrichment only correlated with H3K9me3 ( r = -0 . 3 ) but not any other histone marks . This finding suggests that SUMO modification may be involved in transcriptional regulation of genes at low heterochromatin loci . However , there are also few regions with increased SUMO-2/3 as well as H3K9me3 modifications . Our preliminary data showed that SUMO-2/3 also restrained the transcription of genes in these regions ( S6A Fig ) . This data suggest that SUMO may function at a level higher than histone marks . The regulation of gene expression in those regions may be more complex and worthwhile for detailed analysis in the future . A previous report from Toth et al . noted that the repressive H3K9me3 mark was restricted to two regions of the KSHV genome that mainly encode late viral genes [20] . The absence of SUMO-2/3 enrichment in these two regions indicates that SUMO-2/3 may be less involved in regulating transcription of viral lytic late genes located in heterochromatin regions than early expressed viral lytic genes in euchromatin regions . This idea is further supported by a recent study analyzing the genome-wide SUMOylation sites in human cells . Little SUMO-1 and SUMO-2/3 was found at repressive chromatin regions marked by H3K9me3 [12] . The significant enrichment of SUMO-2/3 in low H3K9me3 region ( Fig 1 ) in conjunction with the higher transcription activation of genes in this region after SUMO-2/3 knockdown ( Fig 4B ) suggests a novel level of interpretation that , in order to attenuate reactivation , expression of immediate early and early gene located in open chromatin regions may be repressed by SUMO-2/3 modification upon stimulation for viral reactivation . These results are consistent with the study in yeast suggesting that SUMO functions in the shut-off of induced genes after stimuli are no longer present [5] . However , we cannot completely exclude the possibility that SUMO-2/3 may participate in transcriptional activation of some viral genes during viral reactivation as one recent global SUMO-1 study mentions that SUMO-1 modification is responsible for stimulation of promoter activity [11] . Together , these findings illustrate the complexity of SUMOylation-mediated epigenetic regulation of transcription under different environmental conditions . SUMO E3 ligases occasionally display paralog-specificity toward certain targets , however , none of the cellular SUMO E3 ligases , including protein inhibitor of activated STAT ( PIAS ) family , Ran-binding protein 2 ( RanBP2 ) , and Pc2 , have been demonstrated to have selectivity towards a specific SUMO paralog . Therefore , despite many studies intent on analyzing paralog-specific SUMO modifications of epigenetic and transcription regulators , the distinctive functional specificity of SUMO isoforms in global epigenetic regulation in relationship to gene expression remains largely unknown . This indicates that a specific stimulus that can induce SUMO paralog-specific modification is required to elucidate the SUMO paralog-specific epigenetic regulatory function . By encoding a viral lytic SUMO-2/3 specific E3 ligase , K-bZIP , KSHV possesses as an ideal tool to distinguish the functional specificity of SUMO-2/3 in epigenetic regulation [27] . To study the functional role of SUMO-2/3 specific E3 ligase activity in KSHV gene expression , we generated a SUMO E3 ligase-dead mutant of K-bZIP ( K-bZIP-L75A ) in the context of the KSHV genome . Analyses of cells containing the K-bZIP-L75A mutant bacmid during KSHV reactivation demonstrated a repressive role of K-bZIP SUMO-2/3 specific E3 ligase activity in regulating the expression of viral genes located in a high SUMO-2/3 enrichment genome region ( Fig 8B ) . This repressive function of K-bZIP helped to dampen KSHV viral production upon reactivation stimuli ( Fig 8C and 8E ) . Moreover , our result demonstrated that E3 ligase activity of K-bZIP is indeed responsible for SUMO-2/3 enrichment on euchromatin regions of the KSHV genome during reactivation ( Fig 9B ) . These results suggest that K-bZIP may mediate the SUMOylation of DNA or histone binding proteins located in KSHV genome euchromatin regions and alleviate transactivation during viral reactivation . The identity of the protein ( s ) modified by SUMO and responsible for global SUMO-2/3 enrichment on the viral genome during reactivation is an important unanswered question . All K-bZIP interacting chromatin binding proteins could be potential candidates , such as the H3K9me3 demethylase JMJD2A [17] . Another potential SUMO-2/3 target might be K-bZIP itself . K-bZIP is a SUMO-2/3 specific E3 ligase and residue Lys158 is a SUMOylation site . SUMOylation at this site is responsible for the transcription repression activity of K-bZIP [27 , 51] . This is consistent with our current finding that viral gene expression is elevated with loss of SUMO-2/3 modification ( Fig 4B ) . However , there may be multiple potential SUMO-2/3 targets that are responsible for global SUMO-2/3 enrichment on the viral genome during reactivation . The factors responsible for SUMO-2/3 enrichment on the host genome can be more complex . Several potential SUMO-2/3 targets we identified in our recent study provide clues for identifying proteins that are responsible for global SUMO-2/3 enrichment on the host genome [10] . Moreover , we cannot exclude the possibility that those same factors may also bind to the viral genome and contribute to the SUMO-2/3 enrichment on the KSHV genome . This is a very interesting topic for further study as it can explore how SUMO modification functions in the regulation of epigenetic status . Surprisingly , during reactivation in K-bZIP-L75A mutated rKSHV , the viral genes in the high H3K9me3 mark region where there is little SUMO-2/3 enrichment , showed defective transcription activity . This implies that K-bZIP may participate in transactivation of viral genes located in high H3K9me3 regions in a SUMO E3 ligase-independent manner . Although K-bZIP is widely accepted as a transcriptional repressor , K-bZIP has also been found to activate gene transcription through direct DNA binding [52] . Direct K-bZIP binding results in a low level of gene transcription . Consistent with our current finding showing relatively low expression level of genes in high H3K9me3 region of KSHV genome ( Fig 3 ) , K-bZIP may directly bind to some of the viral promoters located in this heterochromatin region and mediate a low level of gene transactivation ( Fig 10 ) . This hypothesis is further supported by our recent finding showing a direct binding of K-bZIP to H3K9me3 [17] . The difference in epigenetic context may provide an explanation for the opposing functions of K-bZIP in transcription regulation . Consistent with this notion , a previous report of cytomegalovirus showed that the SIM of IE2 is required for recruiting a SUMOylated transcription initiation factor and this recruitment is essential for the transactivation function of IE2 [53] . Our data here showed that this SIM-dependent recruitment phenomenon may also be true for KSHV K-bZIP . The underlying mechanism for SUMOylation-independent SIM-mediated transactivation in K-bZIP is interesting , however , it still requires for further study . In addition to transcription regulation , the role of K-bZIP in KSHV replication is also controversial . Several reports have showed that K-bZIP is able to regulate KSHV replication through interacting and regulating Ori-Lyt-binding proteins , such as LANA [54] or HDAC [34] . Though it has also been reported that K-bZIP can directly bind to KSHV Ori-Lyt [55 , 56] , the same group showed that K-bZIP is not absolutely required for Ori-Lyt-mediated KSHV DNA replication [36] . In contrast , other studies showed that K-bZIP is essential for Ori-Lyt-mediated replication and virion production [35 , 57] . However , one group also reported that over-expression of K-Rta can overcome the absence of K-bZIP [58] . It has been long known that K-bZIP and K-Rta play an antagonist role against each other . K-Rta , the major transactivator of KSHV , was recently identified as a SUMO-Targeted Ubiquitin Ligase ( STUbL ) . The STUbL activity of K-Rta is a prerequisite for its transactivation activity and for reactivation of KSHV [59] . Consistent with the notion that K-bZIP may oppose the activating function of K-Rta , we found that the E3 ligase activity is essential for K-bZIP to repress activation of KSHV lytic genes and virus reactivation . This finding uncovers a novel mechanism of antagonism between K-Rta and K-bZIP in regulating KSHV life cycle . Although SUMOylation of individual transcription factors is responsible for transcription repression , global SUMO modification of chromatin was found to be essential for either activation of genes [11] or shut-off of active genes after induction [5 , 10] . The multiple isoforms of SUMO proteins may contribute to this discrepancy . However , a recent study on global SUMO modifications of the human genome consistently showed similar epigenetic alterations between SUMO paralogs under physiological stimuli tested [12] . Using KSHV as a model , a SUMO-2/3 specific function in epigenetic regulation of transcription has been revealed . To our knowledge , this is the first report showing that a virus targets SUMO-2/3 specifically for epigenetic modification of its genome and repression of its lytic gene expression . Our results suggest that viruses have evolved a unique way to hijack the SUMO machinery in a paralog-specific manner to alleviate reactivation that may benefit their own survival . | Establishment of KSHV persistent infection requires a dynamic balance between latency , a phase where most viral genes are silenced , and lytic cycle , a phase when nearly all viral genes are expressed . Disruption of this balance may augment virus clearance . During the latent-to-lytic switch , KSHV genomes are subjected to profound epigenetic changes . SUMOylation promotes targeting of proteins to different DNA sites , thereby helping to create specific epigenetic patterns that switch genes between active and inactive stages . It comes as no surprise that SUMOylation may be involved in chromatin remodeling of the KSHV genome during the latent-to-lytic switch and SUMOylation inhibition may disrupt the balance between KSHV latent and lytic cycle . In this study , we identified a profound SUMO-2/3 enrichment in KSHV genome euchromatin regions upon reactivation . SUMO-2/3 modification is responsible for diminishing viral gene expression after reactivation . KSHV SUMO-2/3-specific E3 ligase K-bZIP mediates the SUMO-2/3 enrichment during reactivation . Loss of E3 ligase activity of K-bZIP in the viral context increases viral lytic gene expression and virus production . Our findings demonstrate , for the first time , a SUMO-2/3-specific modification affecting transcription which regulates viral lytic gene expression , and uncovers a novel therapeutic strategy to disrupt persistent infection . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [] | 2015 | K-bZIP Mediated SUMO-2/3 Specific Modification on the KSHV Genome Negatively Regulates Lytic Gene Expression and Viral Reactivation |
Pentavalent antimonials ( Sb5 ) are the first-line drugs for treating cutaneous leishmaniasis in Colombia; however , given problems with toxicity , compliance , availability , and cost , it is imperative to look for better therapeutic options . Intravenous amphotericin B ( AmB ) has been used extensively to treat visceral leishmaniasis; however , evidence on its topical use for cutaneous leishmaniasis is limited . Anfoleish is a topical formulation based on 3% AmB , which was developed following GMP standards by HUMAX and PECET . Anfoleish was shown to be safe and efficacious in animal model and in an open label study in CL patients . Hereafter we show the results of the first controlled and randomized study assessing the safety and efficacy of Anfoleish administered topically , two or three times per day for 28 days , for the treatment of non-complicated cutaneous leishmaniasis in Colombia . An open-label , randomized , non-comparative phase Ib/II clinical trial was performed . Adult volunteers with a parasitologically confirmed diagnosis of cutaneous leishmaniasis were randomly allocated to receive Anfoleish cream either 3 ( TID group ) or 2 ( BID group ) times per day for 4 weeks . 80 out of 105 subjects screened were included in the study . In intention to treat analysis , final cure was observed in 13 ( 32 . 5% ) out of 40 subjects ( IC 95% = 20 . 1–48 ) and in 12 ( 30% ) out of 40 subjects ( IC 95% = 18 . 1–45 . 5 ) in the BID and TID group respectively . In the per protocol analysis , cure rates were 39 . 4% ( n = 13 ) ( IC 95% = 24 . 7–56 . 3 ) and 35 . 3% ( n = 12 ) ( IC 95% = 21 . 5–52 . 1 ) in the BID and TID groups respectively . Anfoleish proved to be safe , and the few adverse events reported were local , around the area of application of the cream , and of mild intensity . Anfoleish showed to be a safe and well-tolerated intervention . Its efficacy results however do not support at this time continuing with its clinical development or recommending it for the treatment of CL . Additional , studies to improve its current formulation are needed before thinking in conducting additional studies in patients . Registered in clinicaltrials . gov NCT01845727 .
Cutaneous leishmaniasis ( CL ) is caused by over 15 different species of the protozoan parasite Leishmania . CL typically begins as a papule at the site of a sand fly bite , enlarges to a nodule , and ulcerates over 1–3 months [1–3] . The exact incidence of CL is not known . An estimated 1 . 2 million cases/year from approximately 90 countries worldwide suffer from different forms of CL [1 , 4] . In the New World , ulcerative lesions are most common . Clinical syndromes of CL vary according to the infecting species and geographic distribution , but no species is uniquely associated with a particular clinical syndrome . Among the different parasites causing CL , L . tropica in the Old World and L . braziliensis in the New World are considered the most important because of the difficulty to cure , public health importance , and severity of the disease [1 , 5] . Colombia now ranks second after Brazil , with around 15 , 000 reported cases per year [4 , 6] . Most cases are caused by L . panamensis , L . braziliensis , and L . guyanensis [1 , 4] . The standard treatment is still parenteral injections of meglumine antimoniate , 20 mg/kg/day/20 days , despite its cost ( USD 60–550 per course ) [7] , variable efficacy of between approximately 36% and 95% , and toxicity [8–10] . Miltefosine , 2 . 5 mg/kg/day for 28 days is also available , but is only used in cases where subjects do not respond to meglumine antimoniate or its use is contraindicated [11] . Pentamidine isothionate is sometimes available but is mainly used for subjects with lesions due to L . guyanensis [1] . There are issues with all these drugs , including mild to serious adverse reactions , compliance , availability , and cost [1 , 12 , 13] . Although Amphotericin B ( AmB ) has been used extensively over the last decades to treat VL , its use for treatment of CL has been limited due to its availability , the need to keep the patient hospitalized during administration , and toxicity [14 , 15] . Liposomal amphotericin B has also been used with good results , however its cost hampers its generalized use [16 , 17] . Topical therapy of CL is a promising approach . Several options have been tested , yet none have been shown to be safe , efficacious , or easy to use in the field such as would enable their adoption in all possible CL epidemiological situations . Other parameters such as frequency and duration of treatment in the presence or absence of an occlusive dressing , and whether lesions are open or closed may influence the efficacy or safety of topically applied formulations [18] . Until recently , the use of topical treatments in the New World CL patients was controversial due to the risk of further development of mucocutaneous leishmaniasis ( MCL ) . The evidence supporting it however was assessed during the 2010 WHO Expert Committee on Leishmaniasis and it was concluded that the evidence was weak and even the use of systemic treatment does not prevent the development of MCL , hence it was recommended to include the use of topical treatment for uncomplicated CL cases due to L . braziliensis . Anfoleish is a topical formulation of a semi-solid oil in water ( O / W ) emulsion containing 3% AmB , developed and manufactured in accordance with Colombian regulations by Humax Pharmaceutical S . A and PECET ( Programa de Estudios y Control de Enfermedades Tropicales ) , Medellín , Colombia . All preclinical studies were performed by PECET following international OECD pre-clinical protocols and standardized animal models . Since there was no previous information about safety and efficacy of Anfoleish in humans , this initial study was designed as a randomized , non-comparative phase Ib/II randomzed study , aiming to evaluate the safety and efficacy of topical Anfoleish administered either two ( BID ) or three ( TID ) times per day for 28 days for the treatment of Colombian patients with CL caused by either L . panamesis or L . braziliensis .
This is an open-label , randomized , non-comparative , two armed exploratory study to evaluate the safety , and efficacy of two regimens of Anfoleish . Subjects who met the following criteria were included in the study: Males and females , aged ≥18 and ≤60 years old , confirmed parasitological diagnosis of CL , subjects with ≤ 3 ulcerative lesions of ≥ 0 . 5 cm and ≤ 3 cm ( longest diameter ) not located on the ear , face , close to mucosal membranes , joints , or on a location that in the opinion of the principal investigator was difficult to maintain application of the study drug topically . Criteria for exclusion were females with a positive serum pregnancy test , breast-feeding , or of a fertile age but not agreeing to take appropriate contraception during treatment period up to D45; history of clinically significant medical problems as determined by history or laboratory studies; previous use of antileishmanial drugs ( within 8 weeks ) ; or abnormal laboratory values at baseline ( Hb < 10g; serum creatinine above normal level; ALT / AST 3 times above normal range ) . Proof of infection was documented either through microscopic identification of amastigotes in stained lesion tissue , the demonstration of motile promastigotes in aspirate cultures , or demonstration of Leishmania by PCR following already published protocols [19 , 20] . Study subjects were adult males serving in the Colombian Army attending a leishmaniasis recovery center or adults attending the PECET Clinic , both locations in Colombia . Participants were randomized and allocated in a 1:1 ratio to receive either Anfoleish applied 3 times per day for 4 weeks ( TID group ) or Anfoleish applied twice a day for 4 weeks ( BID group ) . Cream Application: Prior to the first Anfoleish application , lesions were cleaned with soap , water , and sterile 0 . 9% saline , debrided , and then dried . Anfoleish was applied topically with a gloved finger to cover the whole area of the ulcer , the raised area , and rubbed into the lesion . The lesion ( s ) was covered and left undisturbed until the next application . At each subsequent application of the cream , the previous application and dressing was removed . Study staff members applied the cream to all lesions through Day 28 . The application continued until day 28 even if the lesion had obtained 100% re-epithelialization prior to Day 28 . Rescue therapy: Meglumine antimoniate at doses of 20 mg/SbV /kg body weight per day for 20 days as recommended by Colombian Ministry of Health guidelines was provided free of charge to all subjects who met the failure criteria and those who , for whatever reason , decided to withdraw from the study . Subjects were evaluated on a weekly basis during the treatment , at the end of treatment ( day 28 ) and then on day 45± 5 days and on Days 90± 14 and 180± 14 to assess initial and final cure respectively . The response to treatment was evaluated clinically . The following definitions were used for each lesion: Initial cure: Complete re-epithelialization of all ulcers and complete disappearance of the induration at Day 90 after the start of treatment . Final Cure: Initial cure plus the absence of relapses at Day 180 . Relapse: Lesion that achieved 100% re-epithelialization by Day 90 that subsequently reopened by Day 180 . Failure was defined as <50% re-epithelialization of lesion by nominal Day 45; <100% re-epithelialization of the lesion by nominal Day 90 , and relapse of the lesion at any time between D90 and D180 and an increase of ≥100% in ulcer area as compare to baseline , at any time before D90 . The percentage of re-epithelialization of the lesion ( s ) was calculated by comparing the size of the ulcer at baseline against the size at the follow up visit . Measures were taken after cleaning the lesion and removing the crust . Measures were done in two perpendicular directions using an electronic caliper . The area of ulceration was calculated using the area calculation for an ellipse as follows: Area = A/2*B/2*π mm2 , where A = longest diameter of ulceration in mm; B = perpendicular to “A” diameter of ulceration in mm and π = 3 . 14 . It was calculated that a sample size of 36 subjects per treatment arm ( 36 TID and 36 BID ) would provide a precision estimate of 15% with 95% CI , based on an anticipated cure rate at Day 90 of 70% . Accounting for 10% subjects lost during follow up , four more subjects were added resulting in sample size of 40 subjects per regimen , or 80 subjects in total . Blood samples for PK analysis were collected from the initial 30 subjects ( 15 subjects in each study arm ) as follow: Day 1: Prior to treatment onset and at approximately 2 and 6 hours after the first Day 1 Anfoleish application; on days 14 , 21 and 28 , samples were obtained 2 hours after the first application of the day of Anfoleish and on day 45 , at the end of the assessment visit . Plasma levels of Amphotericin B were determined using HPLC . UV methods and used to calculate the following PK parameters: Cmax , Tmax , AUC; t1/2; and λz: It was calculated that a sample size of 36 subjects per treatment arm ( 36 TID and 36 BID ) would provide a precision estimate of 15% with 95% CI , based on an anticipated cure rate at Day 90 of 70% . Accounting for 10% subjects lost during follow up , 4 more subjects were added resulting in an increase of the sample size to 40 for each regimen . The overall samples size was 80 subjects . Analyses included all randomized participants under the intention-to-treat principle . Subjects’ baseline characteristics were tabulated and analyzed for each treatment group . The efficacy of the treatments was calculated by intention to treat ( ITT ) and per protocol ( PP ) . The relative risk was calculated using 2 × 2 tables . The χ2 test or Fisher’s exact test was used for hypothesis testing of dichotomous variables . Taking into account the distribution of the variables , a Student’s t test or Mann-Whitney U test was used for analyses of continuous data . Potential confounding factors and interactions were controlled with stratified analyses for the species of parasite responsible for the infection , number of lesions , anatomic location of the lesion , type of lesion , and geographic location where the infection occurred . Due to the lack of information about the safety of Anfoleish when applied to Cl subjects , an interim analysis once the first 15 subjects per treatment arm had completed their 28 days treatment was planned . An ad-hoc independent safety-monitoring group reviewed safety and PK data . A list of treatments , generated randomly in blocks of six ( EpiInfo , version 3 . 1 , CDC , Atlanta , GA ) , was used to assign each subject to a treatment group . Numbered opaque envelopes were used to conceal the random allocation sequence . Only the study coordinator had access to the list and was in charge of assigning the treatments . The protocol was approved by the bioethics committee for research on humans in the Sede de Investigación Universitaria ( CBEIH-SIU ) of the University of Antioquia , by the Ethics Committee of the Military Hospital of the Colombian Army and by the National Regulatory Authorities ( Instituto de Vigilancia de Medicamentos y Alimentos—INVIMA ) , and carried out according to international norms of good clinical practice . For the participation of military , a military staff from the Sanidad Militar was invited to participate in the discussion of the project at the Universidad de Antioquia’s Ethics Committee . Approvals from Army’s Research Unit and their Institutional Ethics Committee was also obtained . Recognizing the influences of the military command structure ( in Colombia ) , the study consent was obtained by a study staff not affiliated to the army . The presence of army officers or any superior ( in Colombia ) , at the time of the recruitment or during the consenting process was not allowed . Before entry into the study , investigators obtained written informed consent from all participants . After the patients signed the informed consent form for participation , a clinical form containing demographic information , data on the lesions , and a summary of the inclusion/exclusion criteria was prepared for each patient . A photographic record was also made of each lesion . Clinical samples were taken from all subjects for the parasitological confirmation of leishmaniasis . Leishmania species identification was done using polymerase chain reaction–restriction fragment length polymorphism ( PCR-RFLP ) , following established procedures [19–21] . PECET is certified by Colombian Health Authorities and quality control is conducted by the Antioquia Local Health Department . Each application of the cream was performed by the investigators for the first 30 patients . For the additional patients , only the first application of the cream ( usually in the morning ) was performed by a member of the study team . Investigators observed each participant for 30 minutes after application of study drug . Lesions and surrounding skin were evaluated for pain , pruritus , erythema , and edema daily throughout treatment administration and at follow-up study visits . Clinical and laboratory evidence of side effects was determined on D7 , D14 , and D28 by changes from baseline in liver enzymes and serum creatinine . Patients were not given incentives to come back for follow-up visits; patients were actively followed-up .
The study was carried out between February 2014 and June 2016 . Of the 105 subjects screened , 80 were enrolled and randomly assigned to receive Anfoleish either twice or three times a day for 28 days . Fig 1 shows the number of subjects per treatment group , followed and that constitute the PP and ITT population . A total of 79 subjects completed their treatment . Table 1 shows baseline subject characteristics by treatment group . Apart from lesion size , randomization successfully allocated subjects with similar characteristics , into both treatment groups . Lesions in subjects assigned to the BID group were significantly larger than the lesions of subjects assigned to the TID group ( p = 0 . 04 ) . All 80 subjects had their diagnosis confirmed by smear or biopsy . All but two subjects were male adults . All lesions were ulcerative , and most subjects had only one lesion ( n = 72 , 90% ) . All 80 subjects but one completed their 28 days of treatment . One patient in the BID group withdrew his consent before completing all 28 days of treatment . In the ITT analysis , final cure was observed in 13 ( 32 . 5% ) out of 40 subjects ( IC 95% = 20 . 1–48 ) and in 12 ( 30% ) out of 40 subjects ( IC 95% = 18 . 1–45 . 5 ) in the BID and TID groups respectively . PP analysis cure rates were 39 . 4% ( 13/33 ) ( IC 95% = 24 . 7–56 . 3 ) and 35 . 3% ( 12/34 ) ( IC 95% = 21 . 5–52 . 1 ) in the BID and TID groups respectively . Even though the study was not designed to determine differences in cure rate by Leishmania species and the number of patients per group were small , it was observed that more subjects with infection due to L . panamesis were cured ( 35% ) in comparison with subjects with infection due to L . braziliensis ( 8% ) . The main reason for failure was an absence of initial improvement by day 45 and day 63 in 21 and 18 subjects in the BID and TID groups respectively . Seven subjects withdrew their consent to continue in the study; six were lost during the follow-up period , and three subjects were removed from the study because of the appearance of new lesions . None of the subjects declared cured at day 90 experienced a relapse of their lesion by day 180 . All subjects who fail or withdraw their consent to continue participating in the study were rescued treated as per the national treatment recommendations , using meglumine antimoniate at doses of 20 mg/kg/day for 20 days . All subjects remained under observation during their treatment and were discharged once their lesions were declared cured . Survival analysis , using time to heal of CL lesions as endpoint . Lesion size data time to healing of cutaneous leishmaniasis lesions at baseline ( n = 80 ) ; day 45 ( n = 14 ) ; day 63 ( n = 3 ) and day 90 ( n = 8 ) was analyzed ( Fig 2 ) . Only five subjects , two in the BID group and three in the TID group , reported 12 adverse events related to the cream: burning sensation , itching and rash . All were mild , affecting the area around the lesion where the cream was applied . application’s site . Three subjects experienced mild and transitory elevation of transaminases ( 2 subjects ) or creatinine ( one ) , all classified as of non-clinical significance , In all instances , the values returned to normal at the following control . There was one serious adverse event , varicella , which was classified as not related to the study drug . Seven subjects in each treatment group reported 36 adverse events which were classified as not related to the study drug , including flu related symptoms ( 7 ) muscle aches ( 6 ) and gastrointestinal symptoms . ( 10 ) . Regarding the PK analysis , with a minimum level detection of 4 . 02ng/ml , Amb was not detected in any blood plasma sample from study subjects .
This study was conducted according to the protocol and following the Good Clinical Practices principles . All Subjects voluntarily accepted to participate in the study and received their treatment according to the treatment arm they were randomly allocated . The follow up rate at 6 months was 92 . 5% . By protocol analysis , the therapeutic response of Anfoleish was 39 . 4% for BID group and 35 . 5% for TID group . Topical treatments offer significant advantages over systemic therapy , including easier administration , fewer adverse effects , and cost-effectiveness . They are especially attractive for uncomplicated CL cases , and the use of a number of different approaches has been widely recommended , such as liquid nitrogen , local heat , or intralesional antimonials . The development of topical formulations containing AmB , the most potent anti-Leishmania compound identified so far , seems logical as it would reduce the toxicity of the drug compared to when it is used systemically , and also because of the scarcity of new active compounds against Leishmania parasites . There are different potential explanations for the low therapeutic response to Anfoleish , including the physiochemical properties of the drug , the vehicle , the delivery system , etc . AmB has a high molecular weight and consequently transcutaneous absorption is difficult [22] . Potential problems with its penetration and absorption may have been demonstrated , given that no AmB was detected in any of the blood samples collected for pharmacokinetic analysis from the study subjects , indicating that either the levels were too low for detection by the methods used for the analysis , or that the AmB did not penetrate . It would be interesting to consider the development of new formulations of Anfoleish by evaluating other concentrations of the compound and/or modifying the vehicle to improve its absorption and retention within the tissue [22–25] . Although the therapeutic response was not as anticipated , the absence of a control group in the study makes it difficult to estimate the actual performance of the cream . The exploratory analysis performed in the group of patients who failed revealed a median decrease in the area of lesions in at least 25% of these patients ( Table 2 ) . This finding might suggest some activity of Anfoleish , another option that might worth exploring is the use of Anfoleish in combination with other treatments to determine if the overall efficacy can be improved . There were no differences in the therapeutic response observed between BID and TID groups , which might lead us to believe that in the future two applications per day might be sufficient . Since the vast majority of patients had infections due to L . panamensis , it was impossible to perform an analysis of cure rates by species of Leishmania . Of note , the cure rate achieved by Anfoleish in subjected with lesions due to L . braziliensis ( 8 . 3% ) was close to the rate of spontaneous cure ( 6 . 4% ) reported using placebo interventions in patients with the same infecting specie [26] . In terms of safety , Anfoleish proved to be safe , and the few adverse events reported were local and of mild intensity ( mainly burning , pruritus , rash , and erythema ) in the zone of application of the cream . In conclusion: Although Anfoleish showed to be a safe and well tolerated option , its efficacy results do not support continue with its clinical development as a therapeutic option for CL , Additional formulation studies are needed to improve its current presentation before conducting more clinical studies . | Cutaneous leishmaniasis is a disease caused by Leishmania parasites and transmitted by sand flies . It is a complex disease and is underresearched disease , that depending on the parasite species , can take multiple forms , which can be less or more aggressive . It manifests as ulcers , nodules or other lesions mostly on the face and extremities . Cutaneous leishmaniasis is found across South and Central America , Africa , Europe , the Middle-East and Asia , and generally it is classified into forms of the Old World and New World . Colombia now ranks second after Brazil , with around 15 , 000 reported cases per year . Most cases are caused by L . panamensis , L . braziliensis , and L . guyanensis . Since the 1940’s , Pentavalent antimonials ( Sb5 ) parenteral treatment have been the first-line of drugs for treating cutaneous leishmaniasis , however , the cost , the variable efficacy and the toxicity associated with its use , make it necessary to search for new treatments . Although there are other approved treatments for the disease , They also show issues of adverse reactions , and are used is mainly as a second option of treatment depending on the availability of these in health programs . Topical therapy of cutaneous leishmaniasis is a promising approach . Anfoleish is a cream than developed in Colombia . In this paper , we report the results of phase Ib/II , randomized trial in Colombia to assess the safety and therapeutic response of topical Anfoleish for the treatment of Cutaneous leishmaniasis . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"medicine",
"and",
"health",
"sciences",
"pathology",
"and",
"laboratory",
"medicine",
"clinical",
"research",
"design",
"tropical",
"diseases",
"geographical",
"locations",
"parasitic",
"diseases",
"parasitic",
"protozoans",
"research",
"design",
"protozoans",
"signs",
... | 2018 | A phase II study to evaluate the safety and efficacy of topical 3% amphotericin B cream (Anfoleish) for the treatment of uncomplicated cutaneous leishmaniasis in Colombia |
Cytoplasmic assembly of ciliary dyneins , a process known as preassembly , requires numerous non-dynein proteins , but the identities and functions of these proteins are not fully elucidated . Here , we show that the classical Chlamydomonas motility mutant pf23 is defective in the Chlamydomonas homolog of DYX1C1 . The pf23 mutant has a 494 bp deletion in the DYX1C1 gene and expresses a shorter DYX1C1 protein in the cytoplasm . Structural analyses , using cryo-ET , reveal that pf23 axonemes lack most of the inner dynein arms . Spectral counting confirms that DYX1C1 is essential for the assembly of the majority of ciliary inner dynein arms ( IDA ) as well as a fraction of the outer dynein arms ( ODA ) . A C-terminal truncation of DYX1C1 shows a reduction in a subset of these ciliary IDAs . Sucrose gradients of cytoplasmic extracts show that preassembled ciliary dyneins are reduced compared to wild-type , which suggests an important role in dynein complex stability . The role of PF23/DYX1C1 remains unknown , but we suggest that DYX1C1 could provide a scaffold for macromolecular assembly .
Motile cilia ( also known as flagella ) are antenna-like organelles protruding from many types of cells and required for motility and cell signaling [1 , 2] . Motile cilia are essential for normal vertebrate development , fertility , and organ homeostasis [3–7] . The movement of motile cilia is driven by the ciliary dynein motors , which are composed of outer ( ODA ) and inner dynein arms ( IDA ) [8–10] . ODA and IDA motors are composed of different proteins and have different structures . The ciliary dyneins are assembled in the cytoplasm , a process called “preassembly” , before they are transported into cilia for docking on axonemal doublet microtubules [11 , 12] . Defects in preassembly , transport or docking of the dynein complexes can cause abnormal ciliary motility and disorders in humans including primary cilia dyskinesia ( PCD ) [6 , 13 , 14] . Multiple non-dynein cytoplasmic proteins have been identified that are required for preassembly of the axonemal dyneins ( reviewed in [15 , 16] ) . These include ODA7/LRRC50/DNAAF1 , PF13/KTU/DNAAF2 , PF22/DNAAF3 , HEATR2 , IDA10/MOT48 , ODA16 , ZYMD10 , PIH1D3 and LRRC6 [17–24] ( see S1 Table ) . However , neither the molecular mechanism of the preassembly process , nor the complete genetic/functional relationship between these factors is known . Dyslexia is a disorder in which the patients show normal intelligence but have problems in reading , writing and/or spelling words . The causative genes for dyslexia are controversial [25–32] . One candidate gene is DYX1C1 ( dyslexia susceptibility 1 candidate 1 ) , which was disrupted by a translocation in dyslexia patients [33] , but there is little validation of the role of this gene [34] . Recently , the DYX1C1 gene has also been implicated in ciliary assembly [35 , 36] , but these patients did not show dyslexia [35] . Chandrasekar et al [36] found that knockdown of the DYX1C1 homolog in zebrafish shows ciliary phenotypes that included a failure in assembly of some ciliary dyneins and abnormal ciliary motility . DYX1C1 mutations in human cause PCD due to abnormal ciliary motility and failure in assembly of axonemal dyneins [35] . The authors postulated that DYX1C1 ( also called DNAAF4 ) is required for preassembly of ciliary dyneins , possibly working in concert with PF13/KTU/DNAAF2 [35] . However , the mechanism by which DYX1C1 affects dynein preassembly remained to be fully elucidated . In this study , we report that the Chlamydomonas homolog of DYX1C1 is defective in the classical Chlamydomonas mutant pf23 ( paralyzed flagella 23 ) , a paralyzed mutant deficient in ciliary dynein assembly [37] . pf23 has been widely used to study the function and composition of ciliary dyneins ( e . g . [38–42] ) . In pf23 , DYX1C1 contains a mutation resulting in deletion of 27 amino acids within the DYX domain . The pf23 phenotype is rescued by transformation with the wild-type DYX1C1 gene , which indicates the deletion mutation in PF23/DYX1C1 is sufficient to cause ciliary dynein assembly defects and the non-motile phenotype . Furthermore , cryo-electron-microscopic tomography ( cryo-ET ) and biochemical analyses reveal that the defects in axonemal dynein assembly are more profound than had previously been appreciated [37 , 39 , 40] . PF23/DYX1C1 is localized to the cytoplasm and is required for preassembly of most axonemal dyneins . In addition , missing dyneins in pf23 display a range of defects in their subunit composition , and suggest PF23/DYX1C1 is essential for proper stability of axonemal dynein heavy chains . Thus , PF23/DYX1C1 is a conserved , cytoplasmic ciliary dynein assembly protein .
To determine the function of DYX1C1 in cilia-related processes , we searched the Chlamydomonas genome database ( version 4 ( http://genome . jgi . doe . gov/Chlre4/Chlre4 . home . html ) and version 5 ( https://phytozome . jgi . doe . gov/pz/portal . html%23 ! info ? alias=Org_Creinhardtii ) ) for DYX1C1 homologs . BLAST search using human DYX1C1 protein as the query revealed one copy of a DYX1C1 homolog in Chlamydomonas ( Cre11 . g467560 at Phytozome Chlamydomonas v5 . 5 ) . The predicted DYX1C1 sequences in both genome databases contained gaps , and the correct Chlamydomonas DYX1C1 sequence was determined by PCR ( the sequenced Chlamydomonas DYX1C1 sequence has been deposited in the DNA Data Bank of Japan ( DDBJ ) under the accession number LC149873 ) . The Chlamydomonas DYX1C1 gene maps to chromosome XI near the PF23 locus and is predicted to encode a protein with 816 amino acids and a theoretical pI/MW of 5 . 04/82751 . 23 ( S1A and S1B Fig ) . The Chlamydomonas DYX1C1 protein is longer , with a lower pI , compared to the human DYX1C1 protein , which has 420 amino acids and a theoretical pI/Mw of 8 . 88/48526 . 88 ( Figs 1A , S1A and S1B ) . A search of Chlamydomonas DYX1C1 , using the SMART ( http://smart . embl-heidelberg . de/ ) and pfam ( http://pfam . xfam . org/ ) servers , revealed a CS domain in the N-terminus; CS domains are predicted to be an interacting module for HSP90 ( discussed further below; [18 , 43] ) . The analysis also reveals a central coiled-coil domain and multiple TPR ( tetratricopeptide repeat ) motifs in the C-terminal half of the molecule , which are involved in protein-protein interactions ( S1A Fig ) . Although there are some differences in the number of TPR motifs , the overall DYX1C1 domain structure is conserved in ciliated organisms that include human , mouse , zebrafish , Trypanosoma and Chlamydomonas ( S1A and S1B Fig ) . This domain organization suggests that DYX1C1 is designed for protein interactions [35 , 36] . Cross sections of ciliary axonemes in DYX1C1 morpholino knockdown zebrafish and the PCD patients reveal a partial failure of assembly of dynein arms [35 , 36] . This structural phenotype is reminiscent of the axonemal phenotype of the Chlamydomonas pf23 mutant , which lacks several inner dynein arms [37] . Therefore , we sequenced DYX1C1 in pf23 . Analysis of the DYX1C1 genomic DNA in pf23 reveals a 494 bp deletion that removes all of exon 5 and part of the flanking introns . Using a marker designed for this deletion and crosses between pf23 and a mapping strain ( S1D2: CC-2290 ) , we confirmed that this deletion cosegregates with the Pf23 phenotype ( 62/62 ) . The pf23 DYX1C1 cDNA lacks 81 nucleotides of exon 5 . The mutation results in the removal of 27 amino acids in DYX1C1 , which forms part of the “DYX domain” [36] ( Figs 1A and S1A ) . In addition , cDNA sequencing revealed that in pf23 DYX1C1 exons 4 and 6 are directly connected in-frame ( S1A Fig ) . Consistent with the sequencing data , immunoblots , using anti-Chlamydomonas DYX1C1 antibody ( CT299; see Materials and methods ) , revealed that pf23 expresses an altered , non-functional form of DYX1C1 ( ~ 92 kDa ) while wild-type expresses a ~ 95 kDa form ( Figs 1A , 1B , S2A and S2B ) . To further test whether the defective DYX1C1 gene is responsible for the non-motile phenotype , we transformed pf23 with the wild-type DYX1C1 genomic DNA with its endogenous promoter or cDNA with the PsaD promoter . Both the wild-type DYX1C1 genomic DNA and cDNA successfully rescued the pf23 phenotype . The rescued strains ( listed in S2 Table ) show nearly wild-type motility ( Fig 1C ) . Consistently , immunoblots of whole cells from the rescued strains show both the wild-type DYX1C1 together with the shorter , mutant DYX1C1 ( Fig 1B ) . Also , the ciliary lengths of the rescued strains are near wild-type in the SG ( Sager and Granick ) media ( wild-type , 10 . 9 ± 1 . 1 μm; pf23 , 4 . 8 ± 0 . 6 μm; pf23cR-NT , 11 . 0 ± 1 . 2 μm , pf23cR-3×HA , 11 . 1 ± 2 . 0 μm; pf23gR-T5 , 11 . 6 ± 1 . 4 μm; pf23gR-T9 , 10 . 3 ± 1 . 4 μm; n = 10 , see also [44] ) . These results strongly suggest that DYX1C1 is the gene responsible for the pf23 phenotype , and that an intact DYX domain is essential for ciliary dynein assembly in pf23 [37] . One transformant , pf23gR-T9 ( S2 Table ) , displays slightly slower swimming speeds compared to wild-type or other transformants ( Fig 1C ) . In addition , immunoblots revealed a DYX1C1 fragment of ~ 67 kDa in cells from pf23gR-T9 ( Fig 1A ) . Sequence analysis of the inserted , exogenous DYX1C1 sequence in pf23gR-T9 revealed that during transformation the full-length DYX1C1 gene was not integrated , which predicts the loss of 208 amino acids at the C-terminus of the DYX1C1 protein ( S1A Fig ) . The predicted pI/Mw of exogenous DYX1C1 in pf23gR-T9 was 5 . 22/62780 . 39 , consistent with the last two TPR motifs of DYX1C1 being lost and/or disrupted ( Figs 1A and S2B ) . Conclusions are confounded by the observation that the DYX1C1 protein in pf23gR-T9 is present at a lower level . These results may indicate that while the DYX domain is required for function , either the C-terminal two TPR motifs are required for the assembly of only specific inner dynein arms , or inner dynein arm assembly is affected more by the amount of DXY1C1 , or both ( Figs 1A and S1A ) . This C-terminal truncation would produce a protein that is similar to that encoded by the translocation found in the Finnish family with dyslexia [33] . Ciliary dyneins can be divided into the outer dynein arms ( ODA ) that reside on the outer circumference of the doublet microtubules and the inner dynein arms ( IDA ) that reside on the inner circumference ( Fig 2A ) . Chlamydomonas has only one typeof ODA . There are seven major subspecies of IDAs termed “a” to “g” and three minor subspecies of IDAs called “DHC3” , “DHC4” and “DHC11” ( Fig 2A ) [8 , 45 , 46] . ODAs are particularly important for high ciliary beat frequency , while the IDAs are essential for ciliary waveform control [8 , 41] . The three minor subspecies have been shown to replace the major subspecies in proximal part of the axoneme [46 , 47] , and are predicted to be important for bend initiation in cilia . Although thin-section electron micrographs and high-resolution 2D gel electrophoresis showed a reduction in dynein arm assembly in axonemes from pf23 [37 , 39 , 40] , the exact dynein species that failed to assemble was not determined . Furthermore , the original pf23 was described as a mutant lacking inner dynein arm proteins and structures , but retaining most of the ODAs [37] . Subsequent analysis predicts that four IDAs , “a” , “c” , “d” and “f/I1” are missing in axonemes from pf23 [48 , 49] . To determine precisely which IDA subspecies are missing from pf23 cilia , we used urea-PAGE [50 , 51] to examine the dynein heavy chain composition . As described previously [37] , urea-PAGE reveals that IDA bands are missing or reduced in density compared to wild-type ( Fig 2B ) . To semi-quantitatively estimate the amounts of each dynein in cilia from pf23 , we performed spectral counting of dynein heavy chain bands cut from the gel samples of wild-type and pf23 as well as four of the pf23 rescued strains described above ( pf23cR-NT , pf23cR-3×HA , pf23gR-T5 and pf23gR-T9 ) ( S2 Table ) . Based on spectral counting , IDA subspecies “b” , “c” , “d” , “e” , “f/I1” and “g” are present at less than 20% of the wild-type level ( Fig 3A and 3B ) , whereas inner dynein arm “a” is present at ~70% of the wild-type level ( Fig 3B ) . We also found that ODAs have a modest defect in pf23 as well as IDAs , ODAs are reduced to 50–60% of the wild-type level ( Fig 3B ) . These results were also confirmed by HPLC chromatography using a Mono-Q column , which show clear reductions in all dynein peaks except inner dynein arm “a” ( Fig 2C ) . In addition , we found that the minor IDA subspecies ( “DHC3” , “DHC4” and “DHC11” ) are missing or greatly reduced in pf23 ( Fig 3B ) . Taken together , the dynein defects in pf23 are far more profound than previously described . Partial loss of the DYX domain in DYX1C1 results in an assembly defect for the majority of ciliary IDAs including the minor IDA subspecies and a fraction of the ODAs . Many of the mutations found in patients with PCD are premature stop codons ( Y128X/W162X , V132X , and I195X ) . These stop codons may lead to mRNA decay . By the immunofluorescence microscopy with antibodies to ODA heavy chains , or to DNALI1 ( dynein axonemal light intermediate chain 1 ) in IDAs , the dynein arms are not assembled in these patients [35] . These patients are likely to define the null phenotype , but the presence of the DYX1C1 protein was not monitored . At present , there is not a protein null for the PF23 locus in Chlamydomonas . However , we would argue that the current pf23 allele is not a complete null since some ODAs are assembled unlike in the PCD patients . To further assess defects of dynein structure in pf23 axonemes , we performed cryo-electron tomography ( cryo-ET ) and subtomogram classification/average analysis of isolated axonemes . Sub-volumes containing a 96-nm periodic structure on the doublet microtubule were computationally extracted , aligned in 3D orientation , masked at dyneins of interest , classified and averaged ( Fig 4 and [52] ) . The whole averaged map of pf23 axonemes demonstrates a loss of most of IDAs , consistent with our biochemical analyses , which show a failure in assembly of most inner dynein arms ( Fig 4A ) . A structure with an elongated and a globular density is seen at the position of dynein “a” ( red arrow in Fig 4A , [52] ) , which suggests the presence of dynein “a” . To further characterize the structure located in the dynein “a” position ( referred to as dynein “a” ) , and further assess occupancy of dyneins in the pf23 axoneme , we employed a newly developed image classification technique [52] . In 71% of sub-tomograms , the dynein “a” structure appears more clearly in the sub-averages ( Fig 4B: left ) , while in the remaining sub-tomograms dynein “a” is missing in sub-averages . Thus , consistent with the biochemical analysis in Fig 3A and 3B , cryo-ET reveals partial occupancy of the dynein “a” site in pf23 axonemes . With the exception of dynein “a” , cryo-ET analysis does not resolve other inner arm dyneins . There is a structure in pf23 axonemes at the dynein “f/I1” position ( blue arrows in Fig 4A and 4B ) . However , the structure does not show the morphology of a dynein . Features showing the head and the tail are not found and density corresponding to the IC/LC complex is missing ( Fig 4A and 4C ) . The density at the dynein “f/I1” site is located at the position of the dynein “f/I1” tether , which is associated with the dynein “f/I1” motor domain: the tether is assembled in axonemes from other mutants missing f/I1 dynein [53] . Although the sites of dyneins “c” and “e” have diffuse intensity ( yellow arrow in Fig 4B ) , we do not find any structure , before and after classification that look like a dynein . These densities may represent the IDA heavy chains that are not properly folded or represent density from other proteins . Similarly , no significant density is found at the location of the other inner dynein isoforms . Therefore , we conclude that only dynein “a” is found with partial occupancy in pf23 axonemes . Given the large difference of intermediate/light chain compositions between the dynein species lacking in pf23 ( “b” , “c” , “d” , “e” , “f/I1” , “g” ) , these results strongly suggest that DYX1C1/PF23 plays an important role in assembly of ciliary IDA heavy chains . We also observed that , compared to wild-type axonemes , pf23 axonemes show a tendency to be compressed during cryo preservation . This observation suggests the IDAs play a role in maintaining axonemal integrity . By cryo-ET observation and image classification , we can detect the decrease of ODAs , which is consistent with our biochemical analyses ( Fig 5A ) . Subtomograms containing ODA were classified into five classes . One subclass ( C4 ) shows no density of outer arm dynein heads . Since 19% of subtomograms were categorized in C4 , we conclude that the total occupancy of ODAs in pf23 cilia is about 81% , slightly more than our spectral counting results . Structural differences between other subclasses indicates heterogeneity of ODA in pf23 axonemes . As partly described in Obbineni et al . , [52] , we noticed that one ( β ) of the three ( α , β and γ ) ODA heavy chains are tilted in some populations of ODAs in pf23 axonemes . The observed ODA heterogeneity suggests DYX1C1/PF23 functions in the assembly of ODA heavy chains/heads ( Fig 5B ) . In addition , partial failure in ODA assembly or heavy chain orientation may be a consequence of missing the outer-inner dynein ( OID ) linkers [47 , 53–56] . Other obvious axonemal structures , such as radial spokes , central pair , N-DRC , and MIA complex [57] appear unaffected in the pf23 axonemes . To ask if the cytoplasmic localization of DYX1C1 in mammals [35] is conserved in Chlamydomonas , we examined the cellular localization of DYX1C1 by immunoblots of Chlamydomonas cilia , cell bodies and whole cells using anti-DYX1C1 antibody ( Fig 6A ) . DYX1C1 is found in cell bodies and whole cells , but not in cilia ( Fig 6A ) . Although previous studies postulated that DYX1C1 could function in dynein assembly in the cytoplasm [35 , 36] , the exact state of ciliary dyneins in DYX1C1-deficient organisms remains unresolved . DYX1C1 could be required to maintain the stability of dynein subunits during assembly . To test this , we compared selected dynein subunits from the axoneme and cell body in wild-type and pf23 cells by immunoblots ( Fig 6B ) . As expected , IC2 ( an intermediate chain of ODA ) , IC138 ( an intermediate chain of IDA “f/I1” ) and p28 ( a light chain of IDAs “a” , “c” , “d” ) are reduced in pf23 axonemes ( Fig 6B ) . Importantly , the relative amount of each protein is reduced in the cytoplasmic extracts ( compare red and blue arrowheads , Fig 6B ) . Thus , the loss of part of the DYX domain appears to also be required for stability of axonemal dyneins before transport to and docking in the axoneme ( Figs 1A and S1A ) . To further test the idea that DYX1C1 is required for preassembly of ciliary dyneins in cytoplasm , we examined the preassembly state of ciliary dyneins in wild-type and pf23 cytoplasmic extracts ( Fig 6C ) . As previously reported for sucrose gradient fractionation of wild-type cytoplasmic extracts , IC2 sediments at ~20S and ~12S , IC138 sediments at ~20S , and p28 sediments at ~15S , respectively [11 , 12] . In contrast , following sucrose density gradient centrifugation of pf23 cytoplasmic extracts , the amount of preassembled ODA is greatly reduced compared to wild-type and the preassembly of IDAs is nearly undetectable ( Fig 6C ) . These results indicate that Chlamydomonas DYX1C1 functions in cytoplasmic preassembly of ciliary dyneins . Based on a yeast two-hybrid experiment , Tarker et al [35] suggested a potential interaction between DYX1C1 and PF13/KTU/DNAAF2 . Also , yeast two-hybrid and immunoprecipitation studies suggest interaction between DYX1C1 and PIH1D3 , which is another preassembly factor responsible for X-linked primary ciliary dyskinesia [23 , 24 , 58] . To identify proteins that function with DYX1C1 in the preassembly/cytoplasmic stability of ciliary dyneins , we immunoprecipitated DYX1C1-containing complexes from the cytoplasm of the pf23cR-3×HA strain; an antibody to HA was used for precipitation from cytoplasmic extracts either with or without Bis ( sulfosuccinimidyl ) suberate ( BS3 ) -crosslinking ( see Materials and methods ) . From the immunoprecipitation experiments , we could not identify interactions between DYX1C1 and previously identified dynein preassembly factors reviewed in [16] . In addition , immunoblot analysis of whole cell extracts from wild-type and pf23 do not reveal significant changes in the abundance of the previously identified dynein preassembly factors PF13/KTU/DNAAF2 and IDA10/MOT48 ( Fig 6D ) . In Chlamydomonas , many double mutants that lack inner and outer dynein arms fail to assemble cilia . For example , the pf9×oda4 , ida3×oda2 or pf9×oda3 double mutants lack cilia [44 , 59 , 60] . We made double mutants of pf23 with pf13 ( PF13/KTU/DNAAF2 ) or with oda7 ( ODA7/LRRC50/DNAAF1 ) . The double mutants lacked cilia in both pf23×pf13 and pf23×oda7 strains . Furthermore , DYX1C1 levels are unaffected in the known preassembly deficient mutants , oda7 , ida10 , pf13 and pf22 ( Fig 6E and S2 Table; reviewed in [16] ) . These results suggest that the stability and function of DYX1C1 in Chlamydomonas cells can be independent of the other preassembly factors . Given the differences in the dynein species missing in pf23 compared to other preassembly mutants ( pf13 , pf22 , oda7 and ida10 ) ( for the dynein defect ( s ) in each mutant see S2 Table ) , the DYX1C1/PF23 complex may have a unique molecular function in ciliary dynein pre-assembly . However , it is also possible that DYX1C1 functions both independently and cooperatively with previously identified preassembly factors for dynein assembly . We report that DYX1C1 is a ciliary dynein preassembly factor that is needed for dynein arm assembly; the DYX domain plays a major role in the preassembly of the inner dynein arms . We propose that DYX1C1 is needed for stability of ciliary dyneins both in cytoplasm and cilia . This action could be mediated by effects on protein folding as has been proposed for other preassembly factors , or by providing a scaffold for the association of various essential dynein components . Many of the previously identified preassembly factors have domains identified as playing roles in protein-protein interactions ( e . g . TPR , HEAT , LRR ) , and may provide a unique staging area for the assembly of these large molecular complexes . Additional genetic , biochemical and structural analyses are required to determine the precise mechanistic role of DYX1C1 in dynein assembly .
Chlamydomonas reinhardtii wild-type ( 137c ) and the mutants that were used are listed in S2 Table . Mutants were purchased from Chlamydomonas Resource Center ( University of Minnesota ) or generously provided by Dr . Ritsu Kamiya ( Chuo University ) . Cells were grown in TAP ( Tris-Acetate-Phosphate ) and/or SG ( Sager and Granick ) media under constant illumination . Live Chlamydomonas cells were deciliated using the dibucaine method [61] . In order to prepare axonemes , detached cilia were collected by centrifugation , and demembranated with 0 . 2% IGEPAL or Nonidet P-40 in cold HMDEK ( 30 mM HEPES , 5 mM MgSO4 , 1 mM DTT , 1 mM EGTA , and 50 mM potassium acetate , pH7 . 4 ) or HMDS ( 10 mM HEPES , 5 mM MgSO4 , 1 mM DTT , 4% Sucrose , pH7 . 4 ) buffer . Membrane and matrix fractions were separated from the axonemes by centrifugation to yield a pure axoneme fraction . Extracts from cell body or whole cell samples were treated with methanol and chloroform to remove DNA/RNA fractions and washed with methanol several times until the resultant pellets became nearly white in color . In some experiments , to reduce possible protein degradation , isolated cell bodies or whole cells were directly added to SDS-PAGE sample buffer , mixed well , and heated at 95°C for 10 min . SDS-PAGE and immunoblotting were performed with standard protocols described previously using 5 to 10% acrylamide gels [57] . SDS-PAGE gels were stained with CBB ( Coomassie brilliant blue ) or silver . For immunoblotting , samples separated by SDS-PAGE were transferred to a nitrocellulose or PVDF membrane , stained with CBB or Ponceau S if necessary , incubated with specific primary antibodies and subsequently HRP ( horseradish peroxidase ) -conjugated secondary antibodies . Immuno-reaction was detected using a TMB ( 3 , 3′ , 5 , 5′-tetramethylbenzidine peroxidase ) substrate kit ( Vector Laboratories ) , Pierce ECL immnoblotting substrate ( Thermo Scientific ) or ECL prime immunoblotting detection reagent ( GE Healthcare ) . Primary antibodies used were as follows: anti-DYX1C1 CT299 ( Rabbit: this study ) , anti-PF13/KTU ( Rabbit: [17] ) , anti-IDA10/MOT48 ( Rabbit: this study ) , anti-IC138 ( Rabbit: [62] ) , anti-IC2 ( Mouse: [63] ) , anti-Actin ( Rabbit: [64] ) , anti-p28 ( Rabbit: [59] ) , and anti-HA 3F10 ( Rat: Roche Applied Science ) or anti-HA Y11 ( Rabbit: Santa Cruz ) . HRP-conjugated goat anti-rabbit or goat anti-mouse secondary antibodies were commercially purchased from Invitrogen . Anti-PF13/KTU antibody was a generous gift from Dr . David R . Mitchell ( SUNY Upstate Medical University ) . Sucrose density gradient centrifugation of Chlamydomonas cytoplasmic extracts was performed as described in [65] . Briefly , cells were broken by the glass beads method [66] and crude supernatant was obtained by centrifugation at 10 , 000 rpm for 10 min . The supernatant was clarified at 22 , 500 rpm for 2 hr using a Type 40 Beckman fixed-angle rotor , and the cytoplasmic extract was collected . The cytoplasmic extract was loaded on a 5–20% sucrose density gradient , centrifuged at 32 , 500 rpm for 16 hr using a Beckman SW41Ti rotor , and equal volume fractions were collected into Eppendorf tubes . Anti-DYX1C1 antibody ( CT299 ) was raised against maltose-binding protein fused to the 206-residue DYX1C1 sequence predicted by JGI Chlamydomonas genome version 4 ( http://genome . jgi . doe . gov/Chlre4/Chlre4 . home . html ) . This DYX1C1 antigen was synthesized by GenScript USA Inc . and corresponds to residues 1–142 and 324–350 of both the DYX1C1 sequence determined in this study and the full-length protein sequence predicted by Phytozome Chlamydomonas genome version 5 . 5 ( https://phytozome . jgi . doe . gov/pz/portal . html# ! info ? alias=Org_Creinhardtii ) . Antibody was blot-purified [67] prior to use . An anti-MOT48 antibody was produced against the full-length MOT48 protein ( see[18] ) fused with the glutathione S-transferase protein located at the N-terminus . The MOT48 antiserum was blot-purified using the recombinant MOT48 protein . Crude DNA was obtained from about ~1×106 cells from each progeny of matings and used for mapping PCR as previously described [68] . The matings were carried out between the S1D2 strain ( CC-2290 ) and a pf23 strain [69] . Phenotypic rescue of pf23 using a genomic DNA fragment containing the wild-type DYX1C1 sequence was performed as described previously [70] . Briefly , BAC29H10 was isolated from Chlamydomonas BAC library ( bacterial artificial chromosome ) , and pf23 cells were transformed with 29H10 by the glass bead method [71] . For screening of transformed pf23 cells , the cells were grown in liquid TAP medium and swimming cells were isolated for further analyses . Phenotypic rescue of pf23 using wild-type DYX1C1 cDNA and the pGenD vector was carried out as described previously [18 , 72] . The wild-type DYX1C1 cDNA was synthesized by GenScript Japan ( http://www . genscript . jp/gene_synthesis . html ) , and inserted into the NdeI-EcoRI sites of the modified pGenD vector , which contains the APHVIII gene conferring paromomycin resistance to Chlamydomonas . For some experiments , the 3×HA epitope sequence was also inserted into the pGenD plasmid so that the exogenous DYX1C1 gene would express a protein with the 3×HA tag at the C-terminus . The modified pGenD vector was transformed into pf23 cells by the electroporation method [73] . Transformants were selected by growing the transformed pf23 cells on TAP plates containing 10 μg/ml paromomycin . Swimming cells were then isolated for further study . For semi-quantitative estimation of amounts of axonemal dynein heavy chains , 15 μg axonemes were run on 8% acrylamide SDS-PAGE gels . Gel regions above 250 kDa containing all the axonemal dynein heavy chains were cut out and processed for spectral counting analyses using LC/MS/MS ( liquid chromatography/MS/MS ) . The averages of two independent experiments are summarized in Fig 3 . To determine the potential interacting partners of DYX1C1 , immunoprecipitation experiments were performed using an anti-HA antibody and cytoplasmic extracts from pf23cR-3×HA expressing 3×HA tagged exogenous DYX1C1 . Immunoprecipitation was performed from extracts without chemical crosslinking or following BS3 ( bis ( sulfosuccinimidyl ) suberate ) crosslinking ( 0 . 5 mM ) . Cytoplasmic extracts of pf23cR-3×HA were obtained by sonicating cells that had been deciliated ( by pH shock ) 30 min to induce DYX1C1 expression . The cell extracts were clarified by centrifugation and the supernatants collected . Immunoprecipitation under non-crosslinked conditions was performed as described previously [57] . Cytoplasmic extracts were mixed with HA antibody-conjugated agarose beads ( Anti-HA ( 3F10 ) affinity Matrix: Roche ) , incubated from several hours to overnight , and proteins precipitated with HA antibody-conjugated agarose beads run ~ 10 mm into an SDS-PAGE gel , stained with CBB , cut out from the gel and their identity determined by ESI/LC/MS/MS ( electrospray ionization/liquid chromatography/MS/MS ) analysis . For immunoprecipitation under BS3-crosslinked conditions , cytoplasmic extracts were crosslinked with 0 . 5 mM BS3 for 30 min , and SDS-PAGE sample buffer containing 1% SDS added . Samples were then heated at 95°C for 10 min , diluted 10 times with 1% IGEPAL in TBS [74] , and mixed with an HA-antibody conjugated agarose beads ( Anti-HA ( 3F10 ) affinity Matrix: Roche ) . The mixture was incubated overnight and the proteins precipitated with the HA antibody-conjugated agarose beads processed as described above . For experimental controls , untagged wild-type cytoplasmic extracts were processed in the same way as experimental samples , both under non-crosslinked and BS3-crosslinked conditions . The proteins identified in precipitates from pf23cR-3×HA , but not from wild-type , were considered as potential interacting partners of DYX1C1 . The protein identification thresholds in Scaffold 4 software ( http://www . proteomesoftware . com/products/scaffold/ ) were set as follows: Protein Threshold: 90% , Minimal Peptides Number: 2 , Peptide Threshold: 80% . Purified cilia from pf23 were mounted on lacey carbon grids ( Plano , Germany ) and plunged into liquid ethane at liquid nitrogen temperature using a hand-made manual plunger . Frozen grids were transferred to the JEM2200FS transmission electron microscope ( JEOL , Japan ) by a cryo-transfer system 626 ( Gatan , USA ) . Electron micrographs were recorded at a nominal magnification of 20 , 000 , accelerating voltage of 200 kV and total dose ~40e-/Å2 , by in-column energy filtering ( JEOL ) and a CMOS detector F416 ( TVIPS , Germany ) , using Serial EM [75] . Tomograms were reconstructed using IMOD . Subtomograms were extracted from axonemes located nearly parallel to the tilt axis of the goniometer of the microscope , aligned based on the 96-nm periodicity as described elsewhere [76] , and classified using a missing wedge-free algorithm developed in part using axonemes from pf23 [52] . For ODA classification , we set the number of subclasses to six . Since the sixth subclass contains only nine subtomograms which seem misaligned , we did not pursue further classification and show only five subaverages . 3D maps are presented using UCSF Chimera [77] and IMOD . The analyses of wild-type/pf23 tomograms in Figs 4 and 5 were based on and further characterized/refined from [52] under the publisher’s permission ( Elsevier ) . The density maps used in Figs 4 and 5 have been deposited at EMDataBank ( http://www . emdatabank . org/ ) under the accession IDs EMD3779 , EMD3786 , and EMD3787 . Protein concentration of samples was measured by the Coomassie staining method [78] . Protein motifs and domains were analyzed using SMART ( http://smart . embl-heidelberg . de/ ) or pfam ( http://pfam . xfam . org/ ) searches . For sequence comparisons , data were aligned using ClustalW ( http://www . genome . jp/tools/clustalw/ ) and the output was processed with BioEdit ( http://www . mbio . ncsu . edu/bioedit/bioedit . html ) . Images were equally adjusted for contrast and/or brightness , if necessary . Figures were assembled using Photoshop ( Adobe Systems ) , Microsoft Paint ( Microsoft Windows ) , Illustrator ( Adobe Systems ) , and/or Power Point ( Microsoft Corporation ) . | Most animal cells have antenna-like organelles called “cilia” . These organelles have various important functions both in motility and sensing the environment . Motile cilia are essential for moving cells as well as moving fluids across a surface . The waveform of motile cilia requires large macromolecular motors; these are the ciliary dyneins . These dynein complexes are assembled in the cytoplasm in a pathway called preassembly , and then transported into cilia . Defects in this process cause a heterogeneous human disease called primary ciliary dyskinesia that results , for example , in the disruption of the motility of respiratory tract cilia , sperm and nodal cilia during development . The mechanisms of the preassembly pathway are not fully understood . In this study , we use a mutation in the well-conserved DYX1C1/PF23 gene of the green alga , Chlamydomonas reinhardtii . Loss of a conserved domain ( DYX ) reveals a failure to assemble most ciliary dyneins . Preassembly of inner arm dyneins is particularly affected . We find that if dynein arms are not assembled , dynein subunits in the cytoplasm are unstable . We suggest that DYX1C1 may play a role as a scaffold for other preassembly factors and the dynein subunits . | [
"Abstract",
"Introduction",
"Results",
"and",
"discussion",
"Materials",
"and",
"methods"
] | [
"molecular",
"probe",
"techniques",
"immunoblotting",
"parasitic",
"protozoans",
"dyneins",
"molecular",
"motors",
"protozoans",
"sequence",
"motif",
"analysis",
"molecular",
"biology",
"techniques",
"cellular",
"structures",
"and",
"organelles",
"motor",
"proteins",
"res... | 2017 | Chlamydomonas DYX1C1/PF23 is essential for axonemal assembly and proper morphology of inner dynein arms |
Mating between different species produces hybrids that are usually asexual and stuck as diploids , but can also lead to the formation of new species . Here , we report the genome sequences of 27 isolates of the pathogenic yeast Candida orthopsilosis . We find that most isolates are diploid hybrids , products of mating between two unknown parental species ( A and B ) that are 5% divergent in sequence . Isolates vary greatly in the extent of homogenization between A and B , making their genomes a mosaic of highly heterozygous regions interspersed with homozygous regions . Separate phylogenetic analyses of SNPs in the A- and B-derived portions of the genome produces almost identical trees of the isolates with four major clades . However , the presence of two mutually exclusive genotype combinations at the mating type locus , and recombinant mitochondrial genomes diagnostic of inter-clade mating , shows that the species C . orthopsilosis does not have a single evolutionary origin but was created at least four times by separate interspecies hybridizations between parents A and B . Older hybrids have lost more heterozygosity . We also identify two isolates with homozygous genomes derived exclusively from parent A , which are pure non-hybrid strains . The parallel emergence of the same hybrid species from multiple independent hybridization events is common in plant evolution , but is much less documented in pathogenic fungi .
Hybridization or mating between different species can promote the emergence of new species by creating extreme ( transgressive ) phenotypes allowing adaptation to new ecological niches [1] . In the human fungal pathogen Cryptococcus neoformans , hybridization has been associated with phenotypic evolution and increased virulence [2 , 3] , and in plant fungal pathogens hybridization is associated with increased host range and the emergence of new species [4–6] . Hybridization is particularly common in yeast species used in the preparation of food and drink , such as Zygosaccharomyces and Saccharomyces [7–9] . Natural hybrids between many of the members of the Saccharomyces species complex have been identified [10 , 11] . For example , Saccharomyces pastorianus formed at least twice from recent hybridizations between Saccharomyces cerevisiae and Saccharomyces eubayanus , and this event has been associated with the acquisition of cold tolerance in the lager yeast [12–14] . Homoploid hybrid speciation ( without an increase in chromosome number ) can lead to the formation of new species , for example in natural populations of Saccharomyces paradoxus [15] . Polyploidization was probably important for speciation of up to 1/3 of plants , and has been reported in both plants and animals [16] . The increased use of whole genome sequencing has made it relatively easy to identify hybrids and to study their genome evolution at high resolution [9] , and indeed recent evidence suggests that the whole-genome duplication in the S . cerevisiae lineage arose from an ancient hybridization between two closely related species [17] . Here , we investigate hybridization in members of the yeast CTG-Ser clade ( species that translate the codon CTG as serine and not leucine [18] ) . Several of these species are human fungal pathogens , including Candida parapsilosis , which is particularly associated with infections of neonates [19–21] . The C . parapsilosis sensu lato species complex consists of three defined species: C . parapsilosis sensu stricto , C . orthopsilosis and C . metapsilosis [22] . C . parapsilosis sensu stricto is the most frequently isolated from human infections , followed by C . orthopsilosis ( up to 26% of C . parapsilosis sensu lato isolates ) and C . metapsilosis ( up to 11% of C . parapsilosis sensu lato isolates ) [23 , 24] . There is however a large variation in the frequency of isolation of the individual species , which may be related to geographic region . Several studies fail to identify any C . metapsilosis isolates [23 , 24] , whereas in a 12-year study in Taiwan , approximately equal numbers ( 10% ) of C . parapsilosis sensu lato isolates were identified as C . orthopsilosis and C . metapsilosis [25] . A recent study in Chinese hospitals identified more C . metapsilosis than C . orthopsilosis isolates [26] . The C . parapsilosis sensu lato species vary significantly in virulence and drug susceptibility , with C . parapsilosis being the most virulent , followed by C . orthopsilosis and C . metapsilosis [25 , 27 , 28] . C . parapsilosis sensu lato species are obligate diploids , and mating and meiosis have never been observed [29–31] . The level of heterozygosity in C . parapsilosis sensu stricto isolates is much lower than in other CTG clade species [29 , 32–34] . For example , SNP frequency in one sequenced C . parapsilosis isolate is approximately 1 SNP per 15 kb , which is 70 times lower than in the related species Lodderomyces elongisporus [29] . Low levels of heterozygosity were confirmed by sequencing three additional genomes , though some copy number variations were identified [34] . In addition , all C . parapsilosis sensu stricto isolates characterized to date contain only one mating idiomorph ( MTLa ) at the Mating-Type Like locus , and MTLa1 is a pseudogene [31] . Genome structure in C . metapsilosis however suggests a different evolutionary history in that species . Sequencing genomes of 11 clinical C . metapsilosis isolates showed that they were all highly heterozygous , and most likely resulted from hybridization between two parental species that differed by approximately 4 . 5% at the genome level [35] . Although earlier analysis suggested that C . metapsilosis isolates contained only MTLα idiomorphs [31] genome sequencing revealed that a second idiomorph was formed by introgression at MTLa generating a chimeric locus , containing the MTLa regulatory genes a1 and a2 , and MTLα2 [35] . The authors suggested that a single ancient interspecies hybridization event was followed by global expansion of C . metapsilosis and loss of heterozygosity [35] . In C . orthopsilosis , AFLP ( amplification fragment length polymorphism ) analysis and sequencing of ITS sequences identified some heterogeneity among isolates , suggesting the presence of at least two sub-groups [31 , 36–38] . This was supported by our identification of two MTLa and two MTLα idiomorphs in 16 C . orthopsilosis isolates which differed by approximately 5% [31] . Some isolates were heterozygous at MTL , and we suggested that the two different MTLa/α combinations represented two distinct subspecies , named Type 1 and Type 2 . Sequencing of a putative C . orthopsilosis Type 2 genome ( isolate 90–125 ) showed that it is highly homozygous , similar to C . parapsilosis [29 , 39] . However , further studies identified two highly heterozygous isolates , which were suggested to result from the same hybridization event , possibly between Type 1 and Type 2 parents [40] . Here , we carried out a population genomics analysis of 27 worldwide C . orthopsilosis isolates . We report that most C . orthopsilosis isolates are hybrids most likely formed by mating between two parental species that are about 5% different in sequence , followed by loss of heterozygosity ( LOH ) to form mosaic genomes . Although some aspects of C . orthopsilosis evolution are remarkably similar to the recently described structure of C . metapsilosis populations [35] , we show that C . orthopsilosis has arisen from at least 4 distinct hybridizations between the same two parental species , whereas all known C . metapsilosis strains derive from a single ancestral hybridization . We propose a model for C . orthopsilosis hybrid origins that places Type 1 and Type 2 strains in the hybrid context , and shows that Type 1 and Type 2 are not useful descriptions of the parental species , but are reciprocal combinations of mating partners . The existence of recombinant mitochondrial genomes and a non-hybrid “Parent A” lineage indicates that the formation of C . orthopsilosis by hybridization is recent and probably ongoing . Recurrent hybridization has been shown to lead to increased virulence of some plant and animal fungal pathogens [2 , 4 , 41–44] , but our study is the first to show that it is also occurring in the Candida clade .
We sequenced the genomes of 27 C . orthopsilosis clinical isolates from around the globe , including Europe , US and Asia ( Table 1 ) . Four isolates were sequenced at >400X coverage with the reminder at >70X with Illumina technology . One isolate ( Sample 427 ) was also sequenced using PacBio technology , which was used to characterize genome structure . Previously sequenced isolates 90–125 and MCO456 [39 , 40] were included in the subsequent analysis . We identified an assembly error in the 90–125 reference genome , an artefactual translocation between chromosomes 2 and chromosome 6 ( S1 Fig ) . The corrected assembly has been submitted to GenBank ( BioProject number PRJEA83665 ) . A description of the main findings from the genome data , including an analysis of copy number variation , is provided in S1 File and is summarised in Table 1 . Homozygous and heterozygous single nucleotide polymorphisms and insertions and deletions relative to the reference genome 90–125 were identified as described in Methods ( Fig 1 ) . Only one strain has a highly homozygous genome similar to 90–125 ( Sample 428 , <4 , 000 heterozygous SNPs , mostly derived from incorrectly assembled regions in the reference strain ) . In contrast , the heterozygosity levels of the remaining isolates varied from approximately 100 , 000 ( Sample 1799 ) to 400 , 000 ( Sample 498 ) heterozygous sites , and are much more similar in number to isolate MCO456 described by Pryszcz et al [40] . Pryszcz et al [40] suggested that high levels of heterozygosity in C . orthopsilosis MCO456 and in a related isolate AY2 reflects their origin from a hybridization between two related , but different , species or sub-species , where one parent is highly similar to 90–125 . This proposal was supported by their observation that there is a bimodal distribution of differences between homozygous regions of MCO456 and 90–125 . Approximately 41% of the MCO456 genome is similar to 90–125 ( <1 . 8% divergence ) , and approximately 32% is different ( >1 . 8% divergence ) . We carried out a similar analysis of our data , and we found that all the heterozygous isolates exhibited a similarly bimodal pattern of differences to 90–125 ( Fig 1B , S2 Fig ) . This result indicates that the majority of C . orthopsilosis isolates arose from hybridization between two parents . The average nucleotide difference is 5 . 1% . One of the parents is very similar to 90–125 . To further characterize the parental lineages , we identified the regions of each genome derived from each parent , and used these in a phylogenetic reconstruction . We first identified highly heterozygous regions ( defined as 1 kb regions that were heterozygous in at least 20 isolates , see Methods ) . We then identified the homozygous and heterozygous SNPs from these regions and assigned them to haplotypes . Those that were identical to isolate 90–125 were assigned to haplotype A , and those that were different were assigned to haplotype B . Maximum likelihood trees were generated separately from each haplotype using RAxML [45] ( Fig 2 ) . When all SNPs are considered together , the isolates fall into only two groups ( Fig 2A ) , suggesting that they all arose from the same two parents , where haplotype A is derived from one parental species , and haplotype B from the other . The phylogenetic trees derived independently from SNPs in each of the two haplotypes are very similar ( Fig 2 ) . The isolates fall into 4 major clades , with the two homozygous isolates forming outgroups to Clades 1 and 3 ( Sample 428 ) and Clade 4 ( 90–125 ) . Clade 4 is somewhat divergent , and for discussion we divide it into sub-clades 4 . 1 and 4 . 2 , with a single isolate ( Sample 498 ) in subclade 4 . 3 . The overall sequence divergence in the trees for haplotype B is almost twice as large as for haplotype A ( indicated by scale in Fig 2 ) . We next assigned all sections of each genome ( in 1 kb windows ) into one of four categories: homozygous haplotype A , homozygous haplotype B , heterozygous ( AB ) , and undefined , based on the distribution of homozygous and heterozygous SNPs relative to 90–125 ( Fig 3 , see Methods ) . Loss of heterozygosity ( LOH ) events ( homozygous A or homozygous B ) are shared by members of the same clade , though some events are specific to individual isolates within each clade . Clade 1 isolates have the highest amount of LOH , followed by isolates in Clade 2 , 3 and 4 . The majority of large LOH events occurred towards telomeric regions , particularly in the most heterozygous isolates in Clade 4 . To test whether A and B are the only two haplotypes present in the sequenced C . orthopsilosis isolates , we investigated whether there is any evidence for genomic regions that originate from a third source . To do this , we used the 90–125 strain as a reference for haplotype A , and inferred an almost-complete ( 84 . 6% ) reference for haplotype B from sample 424 , which has the lowest amount of homozygous haplotype A regions . We then repeated the SNP analysis looking for genomic regions that are divergent from both the A and B references ( S3 Fig ) . The majority of the genomes are similar to either haplotype A or haplotype B , except for regions where haplotype B cannot be inferred ( S3 Fig ) . Some short regions on chromosomes 7 and 8 in some isolates differ from both haplotypes A and B , but this is an artifact of short LOH events ( < 500 bp ) in sample 424 ( S3 Fig ) . We therefore conclude that all heterozygous isolates descended from the same two parental species , A and B . For the heterozygous isolates we determined the contribution from each parent by calculating the percentage of each strain’s genome that is homozygous AA , homozygous BB , heterozygous AB or undefined ( Fig 4 , see Methods ) . Most isolates have approximately equal contributions from haplotype A and haplotype B genome-wide . Isolates with the biggest difference include Sample 498 ( 12% more haplotype A than B ) and Sample 436 ( 11% more haplotype B than A ) . Overall however , LOH events appear mostly random in C . orthopsilosis isolates with little preference towards one or the other parental species . To determine if haplotypes A and B underwent recombination we took advantage of the PacBio data from Sample 427 . We restricted the analysis to five regions where two haplotypes were assembled , ranging in size from 20 kb to 64 kb . In all five cases one contig from the PacBio assembly matched 90–125 ( the A genome ) , and the other represented the B parent , indicating that no recombination occurred , at least at these regions in Sample 427 relative to 90–125 ( S1 Table , S4 Fig ) . We previously characterized the Mating-Type-Like ( MTL ) locus of 16 isolates of C . orthopsilosis and showed that both the MTLa and MTLα idiomorphs occurred in two types [31] . The types diverged in sequence by approximately 5% , for both MTLa and MTLα . The majority of isolates in that study were homozygous for either MTLa or MTLα and only two were MTLa/α heterozygotes . In that study , we assumed that these MTL heterozygotes resulted from mating between isolates in a single type or group; the MTLa and MTLα idiomorphs from isolate J981224 were designated as Type 1 , and those from isolate CP125 ( Sample 425 ) as Type 2 . The 5% sequence divergence between the Type 1 and Type 2 MTL idiomorphs is similar to what we now observe between haplotypes A and B at other loci . We analyzed the MTL idiomorphs in our genome sequences ( which included 14 of the strains studied by Sai et al [31] ) . Six isolates are MTLa/α heterozygotes ( S5 Fig ) . Analysis of adjacent SNPs shows that MTLα Type 1 and MTLa Type 2 are both in physical linkage with haplotype A in different strains ( S5 Fig ) . Similarly , MTLα Type 2 and MTLa Type 1 are both linked to haplotype B . Isolates that are homozygous at MTL ( either MTLa/a or MTLα/α ) have undergone LOH , and SNP analysis shows that these tracts of LOH extend into the regions flanking the MTL locus itself so that both chromosomes are derived from the same parental haplotype ( e . g . Samples 423 and 426 are MTLα/α homozygotes derived from haplotypes A and B respectively by LOH; S5 Fig ) . Some of the isolates heterozygous at MTL show a small amount of LOH in a region to the right of the MTL ( Sample 425; S5 Fig ) . We can now see that the MTLa/α heterozygotes designated “Type 1” by Sai et al [31] ( such as Sample 498 ) are heterozygotes containing MTLα from haplotype A and MTLa from haplotype B . The heterozygotes that were designated as “Type 2” ( such as Sample 425 ) have the reciprocal combination of MTLa from haplotype A and MTLα from haplotype B . Therefore , the “Type 1” and “Type 2” labels for MTL heterozygotes represent the two complementary ways that cells from two putative parental species corresponding to haplotypes A and B could combine by mating . We refer to these parental species as ‘Parental Species A’ and ‘Parental Species B’ . Although many Candida species are asexual , a parasexual cycle has been described in some diploid species [46] . Cells of opposite mating type hybridize to form a tetraploid , followed by concerted chromosome loss to regenerate a diploid [47 , 48] . We propose that similar events occurred during hybridization of Parental Species A and Parental Species B during the evolution of C . orthopsilosis , though we cannot rule out the possibility that the ancestral species were fully sexual . For isolates in Clades 1 , 3 and 4 , Parental Species A contributed MTLα , and Parental Species B contributed MTLa . Conversely , for Clade 2 isolates , Parental Species A contributed MTLa , and Parental Species B contributed MTLα ( S5 Fig; Fig 5 ) . This discrepancy in MTL genotypes indicates that Clade 2 and Clades 1/3/4 cannot be descendants of a single ancestral mating event between two cells from the two parental species , so at least two separate hybridization events occurred . By analysis of other datasets described below , we inferred a model for C . orthopsilosis evolution that postulates at least 4 , and possibly 5 separate hybridizations between the parental species ( Fig 5 ) . We assembled the mitochondrial genome sequences from the 27 sequenced isolates and from strain 90–125 [39] , and compared them to the three previously published C . orthopsilosis mtDNAs [49 , 50] . Twenty-eight isolates have linear mitochondrial genomes , and three have circular genomes ( S2 Table ) . In the linear genomes , all the genes are located in a central 24 kb region , which is flanked on each side by a large inverted terminal repeat ( Fig 6A ) . This repeat consists of a subterminal region ( sub ) followed by multiple tandem copies of a telomeric repeat ( tel ) . The PacBio assembly of mtDNA from Sample 427 has nine complete copies of tel on the left arm , and six on the right , followed by an incomplete copy on both arms ( 100 to 103 bp ) . C . parapsilosis mtDNA molecules , which have a similar organization , have been reported to contain up to eight telomeric repeats on each arm [51] . The circular genomes in Samples 436 and 90–125 are caused by recombination at microhomologies in the subterminal repeats , similar to those previously described [49] . Although the Illumina assemblies of the linear mtDNAs are incomplete in the telomeric regions , there are at least two different types of subterminal region ( 451 and 396 bp ) and two types of telomeric repeat ( 565 and 777 bp ) in C . orthopsilosis ( S2 Table ) . The phylogenetic analysis showed that the C . orthopsilosis mitochondrial genomes belong to four mitotypes ( Fig 6B ) . Isolates from three mitotypes ( mt1 , mt2 and mt3 ) correspond to the nuclear Clades 1 , 2 and 3 . We designated strain 90–125 as mitotype mt4 because this strain is closely related to nuclear Clade 4 and its mtDNA appears not to be recombinant ( see below ) . There is almost no variation within the mitotypes ( 0–6 SNPs in the whole genome ) whereas divergence between the mitotypes is significant ( up to 222 nucleotide substitutions or 1 . 1% between mt2 and mt3 ) . In contrast to Clades 1–3 , nuclear Clade 4 isolates have mtDNAs that fall at two distinct positions on the tree . Most belong to a single clade designated as clade mtR1 , but Sample 498 forms a distinct lineage ( mtR2 ) that is more closely related to mt4 ( 90–125 ) and to the mt1 and mt3 clades ( Fig 6B ) . The predicted phylogenetic relationship of mt1 , mt2 , mt3 and mt4 isolates is the same , irrespective of which region of the mtDNA is used for the phylogenetic analysis . However , the placement of the mtR1 and mtR2 isolates varies when different mtDNA regions are used to construct trees ( Fig 6C–6F ) . This suggests that the mtDNA in mtR1 and mtR2 resulted from an inter-lineage recombination , as was previously proposed by Valach et al [50] for the strain MCO471 ( mtR1 ) . The mtR1 and mtR2 mtDNAs are both derived from recombination between the mt2 and mt4 mitotypes , but they were formed by two separate events because the recombinations occurred at different sites in the genome ( Fig 6G and 6H ) . Analysis of diagnostic SNP sites in mtDNA ( Fig 6G ) shows that in mtR1 the center of the mitochondrial genome ( between rrnL and rrnS ) comes from mt2 whereas both arms come from mt4 . In contrast , in mtR2 only the right arm ( from nad5 to the telomere ) comes from mt2 . The left recombination event in mtR1 can be mapped to within the rrnL gene for the large subunit rRNA , because mtR1 shares a 44 bp insert at the 5’ end of rrnL with mt2 , but lacks the rI1 and rI2 introns that are present at the 3’ end of rrnL only in mt2 . A further polymorphism in C . orthopsilosis mtDNA concerns the cox1 intron ai3 , which is present only in the mt2 , mt3 and mtR1 clades ( S2 Table ) . Analysis of the nuclear polymorphisms , MTL loci and mitochondrial genomes together shows that there were at least four independent hybridizations between Parental Species A and B ( Fig 5 ) . We infer that Parental Species A and B were both quite diverse , with multiple populations having distinct nuclear lineages ( lineages B1 to B4 within Parental Species B; and lineages A1 to A4 , 90–125 and 428 within Parental Species A ) and distinct mitochondrial lineages ( mt1-mt4 ) . Each hybridization was a mating event that gave rise to one of the four nuclear clades . After each clade was formed by hybridization , it diversified ( probably clonally , undergoing LOH ) , resulting in congruent SNP trees for its A and B subgenomes . Because hybridization has been associated with increased virulence of the human fungal pathogen Cryptococcus neoformans [2] and has been postulated as a virulence mechanism in C . metapsilosis [35] , we measured the virulence of C . orthopsilosis isolates using the model host Galleria mellonella ( Fig 7 ) . We identified substantial variation , ranging from avirulent to highly virulent isolates ( Fig 7A ) . Isolates in Clade 2 have significantly reduced virulence compared to isolates in Clade 3 and Clade 4 ( Fig 7B ) . However , we did not identify any correlation between levels of heterozygosity and virulence , or between heterozygosity and doubling time in rich media , either by comparing survival endpoints or by using Kaplan-Meier analysis ( Fig 7 , S6 Fig ) . Notably , one homozygous C . orthopsilosis isolate ( Sample 428 ) is virulent ( survival rate <25% ) , whereas the other ( 90–125 ) is not ( Fig 7 , S6 Fig ) . All the isolates studied here would be classified as C . orthopsilosis based on their ribosomal DNA sequences , but our population genomics data shows that the name ‘C . orthopsilosis’ has been applied to two quite different types of isolates . A minority ( 2 of the 29 studied here ) have ‘pure’ genomes that are simply Parental Species A . The first C . orthopsilosis genome sequenced , strain 90–125 [39] was fortuitously a strain of this type . In contrast , the majority of C . orthopsilosis isolates ( 93%; 27 of 29 ) are hybrids formed by mating between this species and a second species ( Parental Species B ) that is 5% different in genome sequence and that has not yet been isolated in a ‘pure’ ( non-hybrid ) form . This relationship is reminiscent of the beer yeasts , where a minority of strains are pure S . cerevisiae , the majority are interspecies hybrids ( S . cerevisiae x S . eubayanus ) , and for a long time there was no known example of a ‘pure’ lineage of the S . eubayanus parent [12] . Our data show unequivocally that hybridization between the A and B parents of C . orthopsilosis occurred by mating , and that it occurred on at least four separate occasions . Since the pure A lineage was found in only 7% of isolates , and the pure B lineage was not found at all , this observation suggests that the interspecies hybrids are significantly more successful ( i . e . , more viable or more virulent ) than their parents . Hybridization has been proposed to increase the virulence capacity of human fungal pathogens [2 , 35] and the host range of plant fungal pathogens [4 , 42] . However , our assays show no consistent correlation between virulence and heterozygosity in C . orthopsilosis , and a large difference in virulence between the two homozygous representatives of lineage A . Our results also indicate that C . orthopsilosis must be capable of mating , even though it has never been seen to mate in the laboratory . The inference that multiple hybridizations occurred means that the species C . orthopsilosis does not have a single clonal origin . Although hybridization is common in yeasts , to our knowledge the only other known example of an ascomycete yeast species with multiple origins by parallel hybridizations is the beer yeast S . pastorianus . Multiple interspecies hybridizations have been reported in other pathogenic fungi ( such as hybridizations between Cryptococcus neoformans and C . deneoformans , or between C . neoformans and C . gattii ) , but in these examples the hybrid lineages are a minority and the parental species are abundant and readily identifiable [2 , 41] . We suggest that , for both C . orthopsilosis and C . metapsilosis , the parental species of these hybrids may be yeasts that are pathogens of other mammalian species and are not normally ( or frequently ) associated with humans . Formation of the hybrid may have facilitated a change in host range and pathogenicity to humans [2 , 4 , 35 , 42] . Further investigation into geographical and ecological variation in the C . parapsilosis sensu stricto clade will be needed to understand the circumstances in which the parental species encounter one another and these pathogenic hybrids can emerge .
For Illumina sequencing , the 27 C . orthopsilosis isolates were cultured overnight in a shaking incubator at 30°C in 5 ml YPD medium ( 1% yeast extract , 2% peptone , 2% glucose ) . Genomic DNA was extracted using the Qiagen Genomic Tip-kit ( 20/G , product code 10223 ) . For PacBio sequencing , Sample 427 was grown in 100 ml synthetic complete media ( 2% glucose , 6 . 7% yeast nitrogen base , 2% Bacto agar , 0 . 2% dropout mix ) overnight to reduce the carbohydrate concentrations . DNA was extracted using the Qiagen Genomic-tip kit ( 500/G , product code 10262 ) using a modified protocol for yeast ( available at PacBio SampleNet , www . pacbiosamplenet . com ) . C . orthopsilosis strains ( Table 1 ) and C . parapsilosis CLIB214 were grown overnight in a shaking incubator at 30°C in 5 ml YPD medium ( 1% yeast extract , 2% peptone , 2% glucose ) . Yeast strains from overnight cultures were centrifuged and washed in Phosphate Buffered Saline ( PBS , Oxoid ) and diluted to 108 cells/ml for C . parapsilosis , and 5 × 107 cells/ml for C . orthopsilosis strains in PBS . G . mellonella in their final larval stage were obtained from Lifefoods Direct Ltd , Sheffield , UK , and stored at 15°C in the dark for use within 7 days from shipment . Twenty larvae , similar in size and weight , were used to analyze the virulence of each fungal strain , in two separate experiments . Larvae were injected with 10 μl of the diluted strains through the last left proleg , using insulin syringes . Untreated larvae and larvae injected with PBS were used as negative controls to assess the general viability and the effect of injection , respectively . After inoculation , larvae were placed into two petri dishes with filter paper ( 10 larvae per dish ) and incubated at 30°C in the dark . Viability of larvae was monitored every 24 hours , for four days . To determine whether the larvae are alive or dead , they were gently touched with tweezers . If no movement was observed , larvae were considered to be dead [52] . Virulence was analysed by comparing the endpoint survival means ( using one-way ANOVA with Bonferroni correction ) and comparing the survival curves over time ( by Kaplan-Meier estimate with log-rank test ) . The statistical analyses were performed using the SPSS Statistics software package and survival package implemented in R . To compare doubling times , 28 C . orthopsilosis isolates were grown overnight in YPD broth in a shaking incubator at 30°C . Cultures were diluted to an A600 of 0 . 1 and incubated in a 96 well round bottom plate . A600 measurements were taken every 15 min over 48 h using a shaking plate reader ( Biotek Synergy HT: Multi-Detection Microplate Reader ) , to generate growth curves . The doubling times of the isolates were calculated from three biological replicates , each with three technical replicates , using the software tool GATHODE [53] . Library preparation and Illumina sequencing of 27 C . orthopsilosis isolates in two Illumina HiSeq 2500 lanes ( 150 bp paired-end ) was carried out by the DNA Core Facility , University of Missouri , USA . Five isolates ( Sample 427 , Sample 831 , Sample 422 , Sample 282 and Sample 320 ) were sequenced on one lane , and the remaining samples were multiplexed in the second lane . Raw read numbers ranged from 40 . 59 to 69 . 22 million reads for the first five genomes and from 6 . 19 to 13 . 17 million reads for the remainder ( Table 1 ) . Raw reads were downloaded for previously published C . orthopsilosis genomes 90–125 [39] ( Illumina GAIIX , 75 bp single end ) and MCO456 [40] ( Illumina HiSeq 2000 , 100 bp paired-end ) . For each isolate , raw reads were trimmed using Skewer v0 . 1 . 117 [54] with parameters—quiet ( without progress updates ) -m pe ( paired-end mode ) -l 36 ( minimum read length allowed after trimming ) -q 15 ( trim 3’ end of read until a quality of 15 or higher is reached ) -Q 15 ( lowest mean quality for a read allowed before trimming ) and -t 4 ( number of threads ) . Reads were mapped to the reference genome ( 90–125 ) using the mem algorithm from bwa ( with -t 8 threads and default options ) , generating BAM files for each isolate . AddOrReplaceReadGroups from Picard Tools v1 . 82 from the Broad Institute [55] was used to add read groups to BAM files in order to pass requirements of Genome Analysis Toolkit ( GATK ) for BAM files . BAM files were indexed using Samtools [56] . The GATK HaploTypeCaller [57] ( with -nct 42 threads and default options ) was used to obtain a high quality set of single nucleotide variants ( SNVs ) . Identified SNVs from all samples were merged using GATK CombineVariants . For the SNP analysis , insertions and deletions were removed using a custom script . Each genome was divided into 1 kb windows and assigned to haplotype A , haplotype B or haplotype A/B depending on the number of homozygous and heterozygous SNPs compared to the reference 90–125 , using a custom R script . Regions were defined as homozygous A if they had <10 homozygous SNPs and <10 heterozygous SNPs , homozygous B if >10 homozygous SNPs and <10 heterozygous SNPs , heterozygous A/B if <10 homozygous SNPs and >10 heterozygous SNPs , and undefined ( >10 homozygous SNPs , >10 heterozygous SNPs ) . Undefined regions ranged from 0 . 93% ( Sample 424 ) to 4 . 81% ( Sample 434 ) , which correlates with the overall heterozygosity levels ( Pearson correlation coefficient of 0 . 86 ) . Undefined regions probably arise because the 1 kb windows used sometimes span the start or the end of a LOH event . These 1 kb regions then correspond partly to a heterozygous region and partly to a homozygous region . To infer haplotype B , all 1 kb regions in Sample 424 defined as either haplotype B or haplotype A/B were extracted . For heterozygous SNPs , the base that differed from 90–125 was used . In total , SNPs from 10 . 71 Mb were extracted . For each extracted SNP the base in the 90-125-reference genome was substituted with the base from haplotype B . GATK HaplotypeCaller was used to call SNPs for all isolates against the inferred parent B . SNPs in 1 kb regions were binned as described above . Mitochondrial genomes were identified as contigs in genome assemblies made using Platanus v1 . 2 . 1 [58] and annotated by reference to Kosa et al [49] . Linear mtDNAs assembled as contigs ending in telomeric ( tel ) repeats ( Fig 6 ) . The circular mtDNAs present in two strains assembled as contigs containing junctions between the left and right subtelomeric ( sub ) regions and lacking telomeric repeats . All 28 mitochondrial genomes that we sequenced have been submitted to the EMBL database ( accession numbers LT594353-LT594380 ) . BAM files generated in the variant analysis step were used to characterize depth of coverage using the DepthOfCoverage tool in GATK with default parameters . Expected genome-wide coverage was calculated using the total number of mapped reads multiplied by the read length and divided by the size of the reference genome ( 90–125 ) . Log2 ratios for copy number analysis for each isolate were calculated in 1 kb windows using the formula: log2 ( observed coverage in 1 kb window / expected coverage ) + log2 ( total expected coverage / total observed coverage ) . Log ratios were then smoothed using the smooth . CNA ( ) function of the DNAcopy package in Bioconductor ( R package version 1 . 42 . 0 ) . We used circular binary segmentation as implemented in the DNAcopy package to extract regions of equal copy number . All identified CNVs were manually verified . The genomes were divided into 1 kb windows and , for each isolate , all SNPs ( heterozygous and homozygous SNPs ) in a specific 1 kb region were extracted if that region contained more than ten heterozygous SNPs in 20 or more of the analyzed isolates ( excluding the two homozygous strains 90–125 and Sample 428 ) . We identified 57530 SNPs from 1195 kb ( 9 . 43% of the genome ) . The SNPs were converted to a FASTA file and split into parental alleles ( named A and B ) using a custom script . In brief , if a base was identical to 90–125 it was assigned to A and an N was inserted for B; a homozygous SNP was assigned to B and an N was inserted for A; a heterozygous SNP was split into A ( equal to 90–125 ) and B . If neither base of a heterozygous SNP matched 90–125 they were alphabetically ordered ( A to T ) and the first base was assigned to A and the second to B . RAxML v8 . 1 . 21 ( raxmlHPC-PTHREADS ) [45] was used to generate 20 maximum likelihood trees ( options -m GTRCAT -p 12345 ) and 1000 bootstraps ( -m GTRCAT -p 12345 -b 12345 ) . Bipartitions were calculated by drawing all bootstraps onto the best maximum likelihood tree using RAxML ( -m GTRCAT -p 12345 -fb ) . Trees were visualized using FigTree v1 . 4 . 2 ( http://tree . bio . ed . ac . uk/software/figtree/ ) To identify structural variations , including insertions , deletions and intra- and inter-chromosomal rearrangements , we utilized BreakDancer [59] with the BAM files generated in the variant analysis step . The script bam2cfg . pl from BreakDancer was used with 10000 random paired-end reads for each isolate to generate configuration files that list read length as well as lower , upper and mean insert size of the paired-end read fragments and its standard deviation . The insert size ranged from 412 . 01 to 442 . 29 bases , with a standard deviation between 95 . 09 and 109 . 56 bases . BreakDancer was then executed with default parameters and the results were analyzed manually . For PacBio sequencing of Sample 427 , SMRT-bells generation , quality control and sequencing on two SMRT Cells using P6-C4 chemistry was outsourced to GATC Biotech Ltd . , Constance , Germany . PacBio SMRT Portal version 2 . 3 . 0 . 140936 . p2 . 144836 was used for quality assessment of reads , generation of subreads , genome assembly using the Hierarchical Genome Assembly Pipeline ( HGAP3 ) algorithm and AHA ( A Hybrid Approach ) scaffolding . The SMRT portal was locally modified to allow execution of commands on a single 48 core Linux server with 256 GB of memory . A total of 300584 polymerase reads were generated from two SMRT Cells , with a total read base count of 1 . 09 billion and a N50 of 18093 bases . After filtering ( with parameters minimum subread length of 500 , minimum polymerase read quality of 0 . 8 and a minimum polymerase read length of 100 ) , 79850 polymerase reads with a total read base count of 894 . 95 million and a N50 value of 19514 bases were used to extract 165541 subreads with a total read base count of 875 . 25 million and a N50 of 6992 bases . The subreads were used as input for the RS_HGAP_Assembly . 3 protocol in the SMRT Portal ( with parameters minimum seed read length of 6000 , number of seed read chunks of six , alignment candidates per chunk of 10 , total alignment candidates of 24 and minimum coverage for correction of six ) . The draft assembly contained 263 polished contigs with a N50 of 292 . 25 kb , a length of 17 . 14 Mb and a mean coverage of 46 . 62x . Scaffolding with the RS_AHA_Scaffolding . 1 protocol using five iterations resulted in 241 scaffolds , a N50 of 322 . 40 kb , a length of 17 . 16 Mb and 22 gaps with a total length of 19 . 04 kb . The 34 longest scaffolds ( ranging from 988 . 53 kb to 91 . 93 kb ) had a total sum of 12 . 65 Mb , compared to a total sum of 12 . 66 Mb for the Co_90–125 assembly . The mitochondrial genome was represented in a single 33 . 78 kb contig . | The genus Candida is one of the leading causes of fungal morbidity in humans . Many pathogenic Candida species are diploid , and do not have have a full sexual cycle . The evolutionary origin of Candida orthopsilosis is unclear . Here , we use whole genome sequencing of 27 C . orthopsilosis isolates from around the world to show that C . orthopsilosis arose from hybridization ( or mating ) of two distinct parental species . Unusually , the hybridization event did not occur only once; we identify at least four events , and we suggest that hybridization is ongoing . The “species” C . orthopsilosis therefore does not have one single origin . We have identified one of the parental lineages involved , but the other remains elusive . Our results suggest that inter-species hybridization has an evolutionary advantage . However , unlike in plant pathogens , it does not appear to result in increased virulence of C . orthopsilosis . | [
"Abstract",
"Introduction",
"Results",
"and",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"heterozygosity",
"fungal",
"genetics",
"population",
"genetics",
"developmental",
"biology",
"plant",
"science",
"phylogenetic",
"analysis",
"energy-producing",
"organelles",
"mitochondria",
"population",
"biology",
"bioenergetics",
"molecular",
"biology",
"techniques",
"pla... | 2016 | Multiple Origins of the Pathogenic Yeast Candida orthopsilosis by Separate Hybridizations between Two Parental Species |
A range of molecular amplification techniques have been developed for the diagnosis of Human African Trypanosomiasis ( HAT ) ; however , careful evaluation of these tests must precede implementation to ensure their high clinical accuracy . Here , we investigated the diagnostic accuracy of molecular amplification tests for HAT , the quality of articles and reasons for variation in accuracy . Data from studies assessing diagnostic molecular amplification tests were extracted and pooled to calculate accuracy . Articles were included if they reported sensitivity and specificity or data whereby values could be calculated . Study quality was assessed using QUADAS and selected studies were analysed using the bivariate random effects model . 16 articles evaluating molecular amplification tests fulfilled the inclusion criteria: PCR ( n = 12 ) , NASBA ( n = 2 ) , LAMP ( n = 1 ) and a study comparing PCR and NASBA ( n = 1 ) . Fourteen articles , including 19 different studies were included in the meta-analysis . Summary sensitivity for PCR on blood was 99 . 0% ( 95% CI 92 . 8 to 99 . 9 ) and the specificity was 97 . 7% ( 95% CI 93 . 0 to 99 . 3 ) . Differences in study design and readout method did not significantly change estimates although use of satellite DNA as a target significantly lowers specificity . Sensitivity and specificity of PCR on CSF for staging varied from 87 . 6% to 100% , and 55 . 6% to 82 . 9% respectively . Here , PCR seems to have sufficient accuracy to replace microscopy where facilities allow , although this conclusion is based on multiple reference standards and a patient population that was not always representative . Future studies should , therefore , include patients for which PCR may become the test of choice and consider well designed diagnostic accuracy studies to provide extra evidence on the value of PCR in practice . Another use of PCR for control of disease could be to screen samples collected from rural areas and test in reference laboratories , to spot epidemics quickly and direct resources appropriately .
Human African trypanosomiasis ( HAT ) , also known as sleeping sickness , is a parasitic disease caused by single-celled , eukaryotic protozoa called trypanosomes . Two subspecies of Trypanosoma brucei namely T . b . gambiense and T . b . rhodesiense , cause the disease in West and Central Africa and in East Africa respectively [1] . In recent years the number of HAT patients has fallen due to the renewal of control programs in the late 1990's; however the current number of patients reported for treatment per year in Africa is still approximately 10 , 000; with an estimated number of infected patients around three times that number [1] . The reference standard diagnostic test for HAT is microscopy , whereby demonstration of parasites in the body fluids confirms active infection [2] , [3] . Microscopy is a compelling diagnostic tool due to its high specificity , ease of use , lack of cold chain , lack of electricity requirements and hence ability to be taken into rural areas where HAT is prevalent . However , its lack of sensitivity ( approximately 10 , 000 parasites/ml for wet blood film examination ) means that many patients may not be positively diagnosed ( false negative ) which may lead to death of patients in the absence of treatment [2] . Only with concentration methods such as microhaematocrit centrifugation [3] , quantitative buffy coat technique ( QBC ) [4] and mini-anion-exchange centrifugation technique ( mAECT ) [5] , [6] can microscopy detect parasitaemia as low as 50 parasites/ml . This limits the utility of microscopy in resource-poor settings , as these concentration methods require electricity and other laboratory logistics . Regardless , microscopy still remains the basis of HAT diagnosis , disease staging and after-treatment follow-up due to its high specificity and availability . HAT comprises two stages of disease; stage one affects the blood , lymph and peripheral organs; stage two occurs when parasites enter the central nervous system . Currently , staging of HAT is performed by microscopic examination of cerebrospinal fluid ( CSF ) for presence of parasites and an increased white blood cell ( WBC ) count ( WHO 1986 ) . Patients with stage one HAT should be treated with pentamidine ( T . b . gambiense ) or suramin ( T . b . rhodesiense ) [7] . Stage two drugs must be able to cross the blood brain barrier ( BBB ) ; melarsoprol is a commonly administered drug for treatment of this stage but can cause reactive encephalopathy with sometimes fatal outcome [8] . The newly recommended treatment for stage two T . b . gambiense HAT , i . e . nifurtimox-eflornithine combination is less toxic but administration is still complex [9] . It is therefore , crucial to reduce false positives and , subsequently also , determine the appropriate treatment by accurate disease stage determination . Recently , a range of molecular amplification techniques have been developed for the diagnosis of HAT , with polymerase chain reaction ( PCR ) at the forefront [10]–[12] . These tests are not commonly used in endemic areas due to the necessity of continuous electricity , trained staff , sophisticated equipment , and the requirement of a cold chain . Isothermal reactions such as loop-mediated isothermal amplification ( LAMP ) [13] , and nucleic acid sequence-based amplification ( NASBA ) [15] , [16] have also been proposed for the diagnosis of HAT . These diagnostic tests may be more applicable for HAT diagnosis because they need less expensive equipment and post-amplification handling requirements that are imposed by PCR testing . If the available molecular amplification diagnostic tests are to be safely used to support HAT diagnosis , they must have high diagnostic specificity as well as sensitivity to ensure that the dangers of inappropriate treatment are avoided . As laboratory strengthening in endemic areas increases , it is expected that the applicability of molecular tests will increase . However , careful evaluation of these tests against the current reference standard , microscopy , must precede implementation . Therefore , we have investigated the published diagnostic accuracy of molecular amplification tests for HAT compared to microscopy for both initial diagnosis as well as for disease staging . Furthermore , we investigated reasons for variation in accuracy amongst HAT diagnostic tests .
In order to find all relevant articles assessing the diagnostic accuracy of molecular assays for HAT , MEDLINE and EMBASE databases were searched with a combination of the following search terms as MeSH ( Medical Subject Headings ) terms and/or free text words; see Appendix S1 . Abstracts of study articles published until the 4th March 2011 were identified electronically in Medline and Embase . Unpublished data were sought from scientific conference abstract books , symposia , books and experts ( Institute of Tropical Medicine , Antwerp , Belgium; Makerere University Kampala , Uganda and Centre International de Recherche-Dévelopement sur l'Elevage en Zone Humide , Bobo Dioulasso , Burkina Faso ) . The reference lists of included studies and of review articles were checked to identify additional studies for inclusion . Articles were initially screened on the title and secondly upon reading the abstract . At this stage , articles not using molecular techniques for diagnostic purposes , case-studies ( only patients with confirmed HAT ) , review articles , serological diagnostics studies and studies only diagnosing animal trypanosomiasis or other non-HAT trypanosomes were excluded . All studies highlighted by at least one of the two review authors were selected; if either reviewer was unsure about exclusion then the article was included to the next stage . The full text of appropriate articles was read and taken forward for study selection . Study selection was conducted by two authors ( CM and EA ) independently , in the case of disagreements a third author ( either KB or ML ) acted as a mediator . We included all studies that evaluated the accuracy of molecular tests for either HAT , for one of the two subspecies of trypanosomes ( i . e . East Africa or Central and Western Africa ) , or for stage two HAT . Studies were included if they involved clinical specimens of patients suspected of any form of HAT and fulfilled the following inclusion criteria: Diagnostic accuracy data for two-by-two contingency tables , patient spectrum data and quality assessment data were extracted by two independent review authors ( CM and EA ) and recorded onto a standard form . Discrepancies were resolved by mediation of a third researcher ( ML ) . From each study , the following characteristics were extracted: i ) molecular test type; ii ) clinical material assessed ( blood , cerebrospinal fluid; iii ) the sub-species detected ( T . b . gambiense or T . b . rhodesiense ) ; iv ) read-out method of index test e . g . oligochromatography ( OC ) ; v ) target gene of the index test; vi ) study design i . e . whether the patients were equally suspected ( ‘consecutive design’ ) or if cases and controls were selected from different populations ( ‘case-control study’ ) . Quality assessment was based on QUADAS ( Quality Assessment of Diagnostic Accuracy Studies ) [17] . The estimates of sensitivity and specificity and their 95% confidence interval were plotted in forest plots and receiver operating characteristic ( ROC ) space in Review Manager version 5 . For the meta-analysis , we used the bivariate random effects model through Proc NLMIXED in SAS for Windows , version 9 . 2 ( Cary , NC ) . This model pools sensitivity and specificity in one model , while accounting for the correlation between the two [18] . Studies that evaluated the diagnostic value of the tests were analyzed separately from studies that evaluated the staging value of the tests . Articles in which two-by-two contingency tables could not be completed were excluded from the meta-analyses . Summary estimates of sensitivity and specificity for diagnosis and staging for the different assays were calculated . Meta-analysis was performed if at least three studies evaluated the same assay in the same sample type ( either blood or CSF ) . Real-time assays were considered as different assays than standard assays , because of significant differences in protocol and design of primer/probe mixes . The results in diagnostic accuracy reviews are expected to show much heterogeneity , mainly due to threshold effects . It is therefore more common to investigate the sources of heterogeneity , without formally testing whether heterogeneity is present or not [19] . For the same reasons , a standard random effects model was used . Heterogeneity was investigated by adding the following covariates to the meta-regression models , if appropriate and possible: i ) type of detection system; ii ) tissue used e . g . blood versus CSF; iii ) sub-species detection T . b . gambiense or T . b . rhodesiense; iv ) target gene of the index test; v ) study design and quality indicated by consecutive versus case-control studies . All reporting in this review is in accordance with the MOOSE guidelines [20] .
The electronic searches yielded a total of 282 articles ( see Figure 1 ) . After reading the title and abstract , thirty-six articles were taken forward and the full text article was read . Twenty articles were excluded at this stage; 4 articles used molecular methods for other purposes e . g . genotyping data , 5 articles did not test patient samples and 11 articles reported case series where the specificity could not be calculated . Sixteen articles were selected for inclusion in the systematic review . The index tests assessed were; PCR ( n = 12 ) [11] , [21]–[31] , NASBA ( n = 2 ) [15] , [16] , LAMP ( n = 1 ) [13] and a study comparing PCR and NASBA ( n = 1 ) [23] . Two studies assessed PCR combined with Oligochromatography ( PCR-OC ) and three studies assessed NASBA combined with Oligochromatography ( NASBA-OC ) . One study [15] assessed a real-time NASBA assay ( RT-NASBA ) . Ten publications focused on the primary diagnosis of HAT in blood , one of these used CSF and blood for diagnosis of HAT . Two publications reported on both diagnosis and staging and used blood for diagnosis and CSF for staging . The two publications focusing only on staging both used CSF for this purpose . See Table 1 for full details . All articles were scored with the QUADAS tool ( quality assessment for diagnostic accuracy ) which included , amongst other , scoring based upon patient spectrum , blinding , exclusion and inclusion criteria ( Figure 2 ) . Studies performed badly when assessed for using representative patient populations . The majority of the studies seemed to enroll their patients in a consecutive way , although they did select them from highly skewed populations: in most articles , patients with confirmed HAT were enrolled , after which these patients underwent both the reference standard ( microscopy ) and the index test . This could artificially increase the clinical accuracy of tests . Only seven out of 16 articles included a representative patient spectrum , that is , patients suspected of infection with HAT . In addition , all studies were scored ‘unclear’ when assessed for blinding of the reference standard to the index test results and vice versa ( items 10 and 11 of QUADAS ) . There is a chance of bias if readers had prior knowledge of either the index or reference test outcome . The verification process ( items 3 to 7 of QUADAS ) raised no problems in most of the articles and the execution of the index test was sufficiently described ( item 8 ) in all articles except one [30] . The aspect of withdrawals ( item 14 ) was not applicable for most of the studies; 2 articles explained the withdrawal of patients from the study ( Figure 2 ) . Two publications did not report sufficient data to construct the complete 2×2 tables , so these were excluded from the meta-analyses [13] , [32] . The ten papers that reported on the accuracy of molecular tests for the diagnosis of HAT , included 15 separate studies and their respective , complete 2×2 tables . Their sensitivity varied from 82% to 100% and the specificity ranged from 59% to 100% ( Figure 3 ) . Eleven studies analysed PCR or PCR-OC in blood; their pooled sensitivity was 99 . 0% ( 95% CI 92 . 8 to 99 . 9% ) and the pooled specificity was 97 . 7% ( 95% CI 93 . 0 to 99 . 3% ) ( Figure 4 ) . There was no significant difference between the clinical accuracy of PCR and PCR-OC performed on blood samples ( Table 2 ) . Two studies assessed NASBA-OC , their sensitivities were 90 . 2% and 97 . 2%; their specificities were 98 . 9% and 59 . 3% respectively . The only study evaluating NASBA-RT in blood had a sensitivity of 93 . 9% and a specificity of 61 . 5% . The largest group of studies evaluated PCR ( including PCR-OC ) on blood . It was performed on five different targets: T . gambiense specific glycoprotein ( TgsGP ) [27] , [31] , serum resistance associated gene ( SRA ) [27] , expression-site–associated genes 6 and 7 ( ESAG 6/7 ) [11] , 18S ribosomal DNA [23 , 23] and the satellite DNA [26] , [28]–[30] . Target genes differ in copy number from TgsGP and SRA as single copy targets , ESAG with 10 copies , 18S rDNA with 40–200 copies and the satellite DNA with approximately 10 , 000 copies . We compared satellite sequences versus the other target sequences , which showed that using the satellite sequences as a target had a significantly lower specificity ( p = 0 . 002 , see Table 2 ) . Another source of heterogeneity is the infecting sub-species ( T . b . rhodesiense or T . b . gambiense ) as patients with T . b . g usually have a lower parasitaemia than patients with T . b . r . In addition , detection of the sub-species specific genes rather than the abundant genes that may appear in both sub-species also changes the diagnostic accuracy . Of the 11 PCR studies conducted on blood , one amplified T . b . rhodesiense-specific genes [27] and two amplified T . b . gambiense-specific genes [27] , [31] . The remaining nine studies were species-specific amplifying T . brucei s . l . , thus amplifying the genes from both subspecies . The advantage of this method is that it is known to increase sensitivity . A separate analysis of the seven studies in patients infected with T . b . gambiense , using a PCR detecting both subspecies revealed a sensitivity of 97 . 6% ( 95% CI 90 . 8 to 99 . 4% ) and a specificity of 95 . 8% ( 95% CI 88 . 9 to 98 . 5% ) . Of the 11 PCR studies on blood , six were diagnostic accuracy studies that enrolled consecutive suspects , the other five were case-control studies . The non case-control studies showed a pooled sensitivity of 98 . 6% ( 95% CI 90 . 7 to 99 . 8% ) and a pooled specificity of 94 . 5% ( 95% CI 86 . 8 to 97 . 8% ) . In the case-control studies , the pooled specificity was significantly higher: 99 . 8% ( 95% CI 95 . 5 to 100% ) . The sensitivity did not significantly differ between the different types of study design: 98 . 7% ( 95% CI 82 . 9 to 99 . 9% ) . See also Table 2 . Four studies evaluated the accuracy of molecular tests to differentiate between stage one and stage two HAT . Three of these evaluated PCR in CSF while one evaluated NASBA-OC . The sensitivity of the PCR tests varied from 88% to 100% , while their specificity varied from 56% to 83% . The sensitivity of the NASBA-OC study was 88 . 6% and its specificity was 14 . 3% .
Our results suffer from two main limitations , one regarding the representativeness of the included patients and the other regarding the reference standard . Of the 11 studies in our main analysis ( accuracy of PCR tests ) , only four included a representative patient spectrum . This may be a threat for the validity of the results shown here and for the translation of the results into practice . Diagnostic accuracy is not a fixed property of a test and may change over populations , especially when these populations are suffering from selection bias [19] , [37] , [38] . The most severe form of selection bias is using a case-control design in which the cases are confirmed and known cases and the controls are healthy people . Four out of eleven PCR studies were case-control studies and these showed a significantly higher specificity , which is expected as the included healthy controls are known to lead to an overestimation of accuracy [39] , [40] . Future studies should think carefully about the patients to include and choose the patient spectrum most closely matching the situation as found in practice , otherwise health workers are forced to rely on accuracy data that are not representative . We recommend the inclusion of clinically or serologically suspected persons; e . g . persons living in endemic regions with enlarged lymph nodes , irregular fever , headaches or other neurological symptoms or positive in a serological test . The other limitation of the studies that are presently available is that most use microscopy as the reference standard . Microscopy , itself , may have a relatively low sensitivity , although most of the studies we included used a form of centrifugation in order to increase sensitivity [34] , [41] . However , the highly toxic treatment administered to HAT patients should only be given after demonstration of the parasites , and therefore , microscopy remains the accepted reference standard for HAT . For this review it means that sensitivity is the percentage of microscopy-positive patients with a positive molecular test and specificity is the percentage of microscopy-negative patients with a negative molecular test . In reality , it is possible that the index tests have correctly diagnosed patients who have been missed by microscopy due to its low sensitivity . In such cases the accuracy , and especially the specificity , of the index test is underestimated . However , in diagnostic studies , if there are any disagreements between the reference standard and the index test then it is assumed that the index test is incorrect . Therefore , in diagnostic accuracy study designs the index tests , by definition , can never be better than the reference standard . Other study designs or analytic techniques are needed to get more information about the relative accuracy of PCR versus microscopy . Examples may be latent class analyses , decision analyses or longitudinal studies using another reference standard to compare both PCR and microscopy with [42] . Even if the accuracy of PCR tests may be close to perfect , implementation of molecular diagnostic tests in the low and middle income countries that are most affected by HAT will be a difficult and arduous task . Role-out could be hampered by more practical issues; the time it may take before a diagnosis is made , the need for a cold-chain , continuous electricity or expertly-trained staff . Development of simple and more appropriate molecular tests , such as LAMP , that may show the same high accuracy in due course , may be a solution . For now , an important role for PCR in the control of HAT may be in screening samples from serologically positive patients collected from the field in a central reference laboratory; the high accuracy , shown here , would allow epidemics of HAT to be spotted early and treatment directed towards these specific areas . Longitudinal impact studies , feasibility studies and cost-effectiveness studies may be warranted to gain further information about the practical application of molecular diagnostics for HAT and their position within the diagnostic algorithm . In conclusion , PCR tests seem to have an acceptably high specificity and sensitivity for diagnosis of stage I HAT . This conclusion is , however , based on microscopy as reference standard and a patient population that was not always representative . Future studies should , therefore , first and foremost include those patients for which PCR may become the test of choice . More certainty about the practical value of PCR tests for HAT diagnosis should come from non-accuracy design studies , like feasibility or cost-effectiveness studies . | A range of molecular amplification techniques has been developed for the diagnosis of HAT , with polymerase chain reaction ( PCR ) at the forefront . As laboratory strengthening in endemic areas increases , it is expected that the applicability of molecular tests will increase . However , careful evaluation of these tests against the current reference standard , microscopy , must precede implementation . Therefore , we have investigated the published diagnostic accuracy of molecular amplification tests for HAT compared to microscopy for both initial diagnosis as well as for disease staging . Here , PCR tests seem to have an acceptably high specificity and sensitivity for diagnosis of stage I HAT . This conclusion is , however , based on multiple-microscopy based techniques as reference standards , which may have low sensitivity , and a patient population that was not always representative . Future studies should , therefore , first and foremost include those patients for which PCR may become the test of choice . More certainty about the practical value of PCR tests for HAT diagnosis should come from non-accuracy design studies , like feasibility or cost-effectiveness studies . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [
"medicine",
"public",
"health",
"and",
"epidemiology",
"diagnostic",
"medicine",
"biology",
"zoology",
"public",
"health"
] | 2012 | Diagnostic Accuracy of Molecular Amplification Tests for Human African Trypanosomiasis—Systematic Review |
Antimicrobial proteins and peptides ( AMPs ) are important effectors of the innate immune system that play a vital role in the prevention of infections . Recent advances have highlighted the similarity between AMPs and amyloid proteins . Using the Eosinophil Cationic Protein as a model , we have rationalized the structure-activity relationships between amyloid aggregation and antimicrobial activity . Our results show how protein aggregation can induce bacteria agglutination and cell death . Using confocal and total internal reflection fluorescence microscopy we have tracked the formation in situ of protein amyloid-like aggregates at the bacteria surface and on membrane models . In both cases , fibrillar aggregates able to bind to amyloid diagnostic dyes were detected . Additionally , a single point mutation ( Ile13 to Ala ) can suppress the protein amyloid behavior , abolishing the agglutinating activity and impairing the antimicrobial action . The mutant is also defective in triggering both leakage and lipid vesicle aggregation . We conclude that ECP aggregation at the bacterial surface is essential for its cytotoxicity . Hence , we propose here a new prospective biological function for amyloid-like aggregates with potential biological relevance .
Antimicrobial proteins and peptides ( AMPs ) represent a wide family that contributes to the host defense system with multiple pathogen killing strategies [1]–[3] . Their fast and multitarget mechanism of action reduces the emergence of bacteria resistance and represents a valuable alternative for common antibiotics [4] , [5] . The mechanism of action of AMPs has been systematically investigated , suggesting that AMPs bind to bacteria cell membranes and disrupt cell homeostasis . However , more investigations are needed to completely understand how different structures determine the function of AMPs [6]–[12] . Membrane damage is a multifaceted mechanism that can involve different peptide assemblies and ultimately promotes membrane permeabilization when achieving a critical concentration [13] , [14] . Several authors have highlighted the striking resemblance of membrane disrupting mechanisms with those observed for amyloid peptides and proteins [15]–[17] . In both cases , membrane composition ( e . g . cholesterol content ) and biophysical properties ( e . g . membrane fluidity and curvature ) were found critical for the peptide action [13] , [15] , [18]–[26] . Furthermore , we have recently suggested that antimicrobial activity could have arisen through cationization of amyloid-prone regions [27] . In this light , some AMPs have been described to form amyloid structures in vitro [28] , [29] and some amyloid peptides have also been considered as putative AMPs [30] , [31] . In fact , we have proposed that inherent AMP aggregation properties can modulate antimicrobial activity [32] . Interestingly , some antimicrobial proteins and peptides have been found to agglutinate bacteria cells . In this sense , bacteria agglutination has been ascribed to unspecific adhesion through hydrophobic interactions , as observed for synthetic peptides derived from the parotid secretory protein [33] . Comparative analysis on those peptides highlighted the contributions of both hydrophobic and cationic residues in the agglutination activity [33] . These results suggest that some AMPs could exploit their intrinsic aggregation properties , by triggering bacteria agglutination as part of its mechanism of action as observed for a wealth source of AMPs in saliva , which provides a first barrier to bacteria adherence in the oral cavity [34] . Agglutinating activity has been reported crucial for the antimicrobial function of Eosinophil Cationic Protein ( ECP ) [35] , a small cationic protein specifically secreted by eosinophil granules during inflammation processes with diverse antipathogen activities [36]–[38] . ECP displays high antimicrobial action , with a specific bacteria agglutination activity reported for Gram-negative bacteria , at a concentration range close to the minimal inhibitory concentration , a behavior that may represent an effective bactericidal mechanism in vivo [39] . In order to characterize the relation between AMPs , bacteria agglutination and amyloid aggregation , we have used ECP as a model of study . We present here a detailed characterization of protein-mediated bacteria agglutination and prove the contribution of an aggregation prone domain to the protein antimicrobial action . Complementary studies on model membranes provide a further understanding of the membrane damage process promoted by protein aggregation .
To address the first question we compared the antimicrobial action of wild type ECP ( wtECP ) with the I13A mutant , previously described to be unable to form aggregates in vitro [28] . The antimicrobial assays reveal that , while wtECP has an average minimal inhibitory concentration ( MIC ) value around 0 . 5–1 µM , the I13A mutant is unable to kill bacteria even at 5 µM concentration ( Table 1 ) . To further correlate ECP antimicrobial and agglutination activities we studied bacteria cell cultures by confocal microscopy using the SYTO9/Propidium iodide nucleic acid fluorescent labels that allow registering both cell agglutination and viability over time . Interestingly , wtECP can agglutinate Gram-negative bacteria before a viability decrease is observed ( Figure 1A ) , however no cell agglutination takes place when bacteria are incubated with the I13A variant , even after 4 hours ( Supporting Information Figure S1 ) . These results are also supported by minimal agglutination concentrations ( MAC ) close to the MIC values ( Table 1 ) and by FACS experiments showing that wtECP but not I13A mutant is able to agglutinate E . coli cells ( Figure 1B ) . Thus , ECP antimicrobial activity on Gram-negative strains is strongly affected when abolishing the agglutination behavior ( Ile13 to Ala mutation ) . To further analyze the protein agglutination mechanism , we tested the wtECP and I13A mutant action on a simpler biophysical system such as phospholipid membranes where liposome agglutination is registered as a function of protein concentration . In contrast to wtECP , I13A mutant completely looses the ability to agglutinate membranes ( Figure 2A ) . In particular , when following wtECP agglutinating activity as a function of ionic strength , we observe that liposome agglutination is enhanced at high NaCl concentration ( Supporting Information , Figure S2 ) . These results suggest that vesicle agglutination is promoted by hydrophobic interactions . Even more , leakage activity in model membranes is also lost for I13A mutant ( Figure 2B ) , meaning that protein aggregation on the membrane surface is important not only for agglutination but also for later membrane permeabilization . These results are entirely consistent with those described above for bacteria cell cultures where the Ile to Ala mutation not only abolishes the cell agglutinating activity of ECP but also its bactericidal action . Next , to address the question whether cell agglutination is consistently driven by protein aggregation at the bacteria surface , we incubated bacteria cultures with ECP and visualized the samples using confocal microscopy . Our results show that wtECP binds to the bacteria surface and a strong protein signal is registered at the aggregation zones ( Figure 3A ) . On the contrary , though cell interaction is maintained for the I13A mutant , agglutination is observed neither in bacteria cell cultures nor in model membranes ( Figures 3A and 3B ) . As expected , for model membranes we show that only wtECP is able to promote agglutination ( Figure 3B ) . Therefore , we conclude that protein aggregation on the cellular surface is required for bacteria agglutination , which turns to be essential for the antimicrobial action . Agglutination is also observed in the presence of 20% plasma in a similar extent , suggesting that ECP agglutination is likely to take place in the physiological context ( Supplementary Information Figure S3 ) . As previously mentioned , ECP binding to bacteria is favored by interactions with the LPS outer membrane [35] , [41] , [47] . Consistently , we show here that LPS binding activity is lost for the I13A mutant , when compared with wtECP ( Supplementary Information Figure S4 ) . At this point however , the nature of the protein aggregates remained unknown . Thus , having previously shown that ECP is able to form amyloid-like aggregates in vitro , we decided to test if the observed aggregates have an amyloid-like structure using the amyloid-diagnostic dyes Thioflavin-T and Congo Red . When bacteria cultures are incubated with non-labeled wtECP , stained with ThT and visualized by total internal reflection fluorescence ( TIRF ) microscopy , we show that wtECP amyloid-like aggregates are located also at the cell surface ( Figure 4A ) similarly as what we observe for Alexa labeled wtECP ( Figure 3A ) . Consistently , no staining is observed for non-incubated cultures and for the I13A mutant ( Figure 4A ) . Moreover , upon bacteria incubation with wtECP , a red shift in the Congo Red spectrum is observed ( Supplementary Information Figure S5A ) , revealing that the protein amyloid-like aggregation is triggered upon incubation with bacteria cultures . Though ECP was previously shown to form amyloid-like aggregates in vitro only at low pH after a long incubation time ( 1–2 weeks ) , amyloid-like structures observed here are detected after only 4 hours of incubation . However , it is well known that some proteins can accelerate its aggregation kinetics in the presence of membrane-like environments [48]–[50] . Our results show that wtECP is able to form fibrillar-like aggregates on model membranes with an average size of 845±150 nm ( Figure 4B ) , comparable in size with the wtECP aggregates observed in vitro in the absence of lipid membranes ( ∼150 nm ) [28] . In fact , when tested for ThT binding , we observe aggregates with similar size ( Figure 4B ) . When wtECP is incubated with model membranes and tested for Congo Red binding , we obtain again a noticeable spectral shift ( Supplementary Information Figure S5B ) . To complete these results we have also performed all the experiments detailed above using the I13A mutant and found it to be unable to form amyloid-like aggregates ( Figure 4 ) . The results presented here for ECP reinforce the hypothesis that an amyloid-like aggregation process is taking place in the bacteria surface that drives bacteria cell agglutination , which is essential for the antimicrobial activity of the protein . In summary , after binding to the bacteria surface , a rearrangement of the protein could take place , exposing the hydrophobic N-terminal patch of the protein . Following , the aggregation process would start promoting the agglutination of the bacteria cells through the aggregation of the surface-attached protein molecules . The formation of aggregates on the bacteria surface will disrupt the lipopolysaccharide bilayer of Gram-negative cells exposing the internal cytoplasmatic membrane to the protein action , promoting the membrane disruption and eventually the bacteria killing . Cell agglutinating activity provides a particularly appealing feature that may contribute to the clearance of bacteria at the infectious focus . In this sense , bacteria agglutination would prepare the field before host phagocytic cells enter in the scene [33] . However , despite the interest in the pharmaceutical industry to identify the structural determinants for bacteria cell agglutination , bibliography on that subject is scarce and only few agglutinating antimicrobial proteins are described in the literature . Excitingly , there may be other proteins and peptides with similar characteristics that also follow the proposed model . Hence , the agglutinating mechanism may represent a more generalized process that may derivate in amyloid deposit formation at bacterial infection focuses . Besides , it has been reported that systematic exposure to inflammation may represent a risk factor on developing Alzheimer's disease [51] , [52] and other types of dementia [53] . Some studies have also demonstrated that the release of inflammatory mediators can also cause generalized cytotoxicity . In particular , ECP has been discovered to be cytotoxic [40] , [54] and neurotoxic , causing the Gordon phenomenon after injection intratechally in rabbits [55] . Therefore , our results suggest that the release of inflammatory mediators after infection ( like AMPs ) may either seed the aggregation processes in the brain and/or influence the membrane biophysical properties to trigger neurotoxicity and aggregation events .
Antimicrobial activity was expressed as the MIC100 , defined as the lowest protein concentration that completely inhibits microbial growth . MIC of each protein was determined from two independent experiments performed in triplicate for each concentration . Bacteria were incubated at 37°C overnight in Mueller-Hinton II ( MHII ) broth and diluted to give approximately 5·105 CFU/mL . Bacterial suspension was incubated with proteins at various concentrations ( 0 . 1–5 µM ) at 37°C for 4 h either in MHII or 10 mM sodium phosphate buffer , 100 mM NaCl , pH 7 . 4 . Samples were plated onto Petri dishes and incubated at 37°C overnight . For MAC determination , bacteria cells were grown at 37°C to mid-exponential phase ( OD600 = 0 . 6 ) , centrifuged at 5000×g for 2 min , and resuspended in 10 mM sodium phosphate buffer , 100 mM NaCl , pH 7 . 4 , in order to give an absorbance of 0 . 2 at 600 nm . A 200 µL aliquot of the bacterial suspension was incubated with proteins at various ( 0 . 1–10 µM ) concentrations at 25°C for 4 h . Aggregation behavior was observed by visual inspection and minimal agglutinating concentration expressed as previously described [42] . Bacteria cells were grown at 37°C to mid-exponential phase ( OD600 = 0 . 6 ) , centrifuged at 5000×g for 2 min , resuspended in 10 mM sodium phosphate buffer , 100 mM NaCl , pH 7 . 4 or the same buffer supplemented with 20% plasma to give a final OD600 = 0 . 2 and preincubated for 20 min . A 500 µL aliquot of the bacterial suspension was incubated with 5 µM of wtECP or I13A mutant during 4 h . After incubation , 25000 cells were subjected to FACS analysis using a FACSCalibur cytometer ( BD Biosciences , New Jersey ) and a dot-plot was generated by representing the low-angle forward scattering ( FSC-H ) in the x-axis and the side scattering ( SSC-H ) in the y-axis to analyze the size and complexity of the cell cultures . Results were analyzed using FlowJo ( Tree Star , Ashland , OR ) . Bacteria viability assays were performed as described before [39] . Briefly , bacteria were incubated in 10 mM sodium phosphate buffer , 100 mM NaCl , pH 7 . 4 with 5 µM of wtECP or I13A mutant and then stained using a syto 9/propidium iodide 1∶1 mixture . The viability kinetics were monitored using a Cary Eclipse Spectrofluorimeter ( Varian Inc . , Palo Alto , CA , USA ) . To calculate bacterial viability , the signal in the range 510–540 nm was integrated to obtain the syto 9 signal ( live bacteria ) and from 620–650 nm to obtain the propidium iodide signal ( dead bacteria ) . Then , the percentage of live bacteria was represented as a function of time . The ANTS/DPX liposome leakage fluorescence assay was performed as previously described [56] . Briefly , a unique population of LUVs of DOPC/DOPG ( 3∶2 molar ratio ) lipids was obtained containing 12 . 5 mM ANTS , 45 mM DPX , 20 mM NaCl , and 10 mM Tris/HCl , pH 7 . 4 . The ANTS/DPX liposome suspension was diluted to 30 µM concentration and incubated at 25°C in the presence of wtECP or I13A mutant . Leakage activity was followed by monitoring the increase of the fluorescence at 535 nm . For liposome agglutination , 200 µM LUV liposomes were incubated in 10 mM phosphate buffer , pH 7 . 4 , containing 5 to 100 mM NaCl , in the presence of 5 µM wtECP or I13A mutant and the scattering signal at 470 nm was collected at 90° from the beam source using a Cary Eclipse Spectrofluorimeter ( Varian Inc . , Palo Alto , CA , USA ) [57] . Experiments were carried out in 35 cm2 plates with a glass coverslip . For phospholipid membranes , 500 µl of 200 µM LUV liposomes ( prepared as described in Supplementary Information ) were incubated with 5 µM wtECP or I13A mutant for 4 h in 10 mM sodium phosphate buffer , 100 mM NaCl , pH 7 . 4 . For bacteria , 500 µl of E . coli cells ( OD600 = 0 . 2 ) were incubated with 5 µM wtECP or I13A mutant for 4 h in 10 mM sodium phosphate buffer , 100 mM NaCl , pH 7 . 4 . RNase A was used always as a negative control . Samples of both liposomes and bacteria were imaged using a laser scanning confocal microscope ( Olympus FluoView 1000 equipped with a UPlansApo 60× objective in 1 . 4 oil immersion objective , United Kingdom ) . wtECP and I13A mutant labeled with Alexa Fluor 488 were excited using a 488-nm argon laser ( 515–540 nm emission collected ) and Vibrant DiI was excited using an orange diode ( 588–715 nm emission collected ) . To study the interaction of proteins with lipid membranes , planar supported lipid bilayers were used ( Supplementary Information ) . When using bacteria , glass coverslips were previously treated with 0 . 1% poly-L-lysine to ensure that samples will adhere to the surface . 500 µl of E . coli cells ( OD600 = 0 . 2 ) were incubated with 5 µM wtECP or I13A mutant for 4 h and then transferred to poly-L-lysine treated microscopy plates and incubated for 15 minutes . To remove unattached cells , plates were washed twice with 10 mM sodium phosphate , 100 mM NaCl , pH 7 . 4 buffer . RNase A was used always as a negative control . Images were captured using a laser scanning confocal microscope ( Olympus FluoView 1000 equipped with a PlansApo 60× TIRF objective in 1 . 4 oil immersion objective , United Kingdom ) using the same conditions as described for confocal microscopy experiments . Thioflavin T ( ThT ) was used to detect amyloid aggregates . In this case , samples were incubated for 4 h with unlabeled proteins as described before and then incubated with ThT at 25 µM final concentration for 15 minutes . Then , plates were washed twice with 10 mM sodium phosphate , 100 mM NaCl buffer , pH 7 . 4 to remove unattached cells and ThT excess . | Microbial infections are reported among the worst human diseases and cause millions of deaths per year over the world . Antibiotics are used to treat infections and have saved more lives than any other drug in human history . However , due to extended use , many strains are becoming refractive to common antibiotics . In this light , new promising compounds , like antimicrobial proteins and peptides ( AMPs ) are being investigated . Some AMPs also show agglutinating activity; this is the ability to clump bacteria after treatment . This feature is particularly appealing because agglutinating peptides could be used to keep bacteria to the infection focus , helping microbe clearance by host immune cells . In this study , we propose a novel mechanism to explain agglutinating activity at a molecular level using Eosinophil Cationic Protein . We show that the agglutinating mechanism is driven by the protein amyloid-like aggregation at the bacteria cell surface . Accordingly , elimination of the amyloid behavior abolishes both the agglutinating and the antimicrobial activities . This study provides a new concept on how Nature could exploit amyloid-like aggregates to fight bacterial infections . Moreover , these results could also add new insights in understanding the relation between infection and inflammation with dementia and amyloid-related diseases like Alzheimer . | [
"Abstract",
"Introduction",
"Results/Discussion",
"Materials",
"and",
"Methods"
] | [
"macromolecular",
"assemblies",
"microbiology",
"chemical",
"biology",
"escherichia",
"coli",
"bacterial",
"pathogens",
"immune",
"system",
"proteins",
"proteins",
"microbial",
"pathogens",
"biology",
"pathogenesis",
"biophysics",
"drug",
"discovery",
"biochemistry",
"gram"... | 2012 | Exploring New Biological Functions of Amyloids: Bacteria Cell Agglutination Mediated by Host Protein Aggregation |
Hypoxia-induced cell injury has been related to multiple pathological conditions . In order to render hypoxia-sensitive cells and tissues resistant to low O2 environment , in this current study , we used Drosophila melanogaster as a model to dissect the mechanisms underlying hypoxia-tolerance . A D . melanogaster strain that lives perpetually in an extremely low-oxygen environment ( 4% O2 , an oxygen level that is equivalent to that over about 4 , 000 m above Mt . Everest ) was generated through laboratory selection pressure using a continuing reduction of O2 over many generations . This phenotype is genetically stable since selected flies , after several generations in room air , survive at this low O2 level . Gene expression profiling showed striking differences between tolerant and naïve flies , in larvae and adults , both quantitatively and qualitatively . Up-regulated genes in the tolerant flies included signal transduction pathways ( e . g . , Notch and Toll/Imd pathways ) , but metabolic genes were remarkably down-regulated in the larvae . Furthermore , a different allelic frequency and enzymatic activity of the triose phosphate isomerase ( TPI ) was present in the tolerant versus naïve flies . The transcriptional suppressor , hairy , was up-regulated in the microarrays and its binding elements were present in the regulatory region of the specifically down-regulated metabolic genes but not others , and mutations in hairy significantly reduced hypoxia tolerance . We conclude that , the hypoxia-selected flies: ( a ) altered their gene expression and genetic code , and ( b ) coordinated their metabolic suppression , especially during development , with hairy acting as a metabolic switch , thus playing a crucial role in hypoxia-tolerance .
Mammalian tissues experience a reduction in oxygen delivery at high altitude or during certain disease states , such as myocardial infarction and stroke . In order to survive , cells , tissues and organisms have developed various strategies to adapt to such O2 limited condition . There are indeed major differences between different organisms and cells in their ability to survive reduced environmental O2 . For example , turtle neurons are very tolerant to low oxygen and can survive without O2 for hours and days [1] , [2] . In contrast , mammalian neurons are very sensitive to reduced oxygen and cannot survive for even minutes under similar conditions . However , the mechanisms underlying survival in such extreme hypoxic conditions are not clear at present , in spite of the fact that there have been a number of interesting observations in this regard in the past few decades . For instance , it has been demonstrated that a number of hypoxia-tolerant animals ( e . g . Pseudemys scripta and Crucian Carp ) reduce their O2 consumption during hypoxia in such a way to minimize the mismatch between O2 supply and demand [3]–[5] . Similar phenomena was also observed in Drosophila melanogaster [6] , [7] and in newborn mammals [8] , [9] . Many questions , however , remain unsolved . For instance , we do not have an adequate understanding of the mechanisms that are responsible for reducing metabolic rate during low O2 conditions; and similarly , the mechanisms that are responsible for coordinating the suppression of these metabolic processes are still largely unknown . In the early 1990s , we discovered that the fruit fly , Drosophila melanogaster , is tolerant to acute anoxia ( zero mmHg O2 ) . Flies can sustain such environment for a few hours without any evidence of injury [6] . Since a ) Drosophila has been demonstrated to be a powerful genetic model for human diseases [10]–[14] and b ) many biochemical and genetic pathways are highly conserved between Drosophila and mammals , we used Drosophila in the current study to explore the mechanisms underlying tolerance to long-term hypoxia . We first generated a Drosophila melanogaster strain that can live perpetually ( i . e . , from generation to generation ) in severe , normally lethal , hypoxic conditions . To better understand the mechanisms underlying this remarkable hypoxia tolerance , we used cDNA microarrays containing 13 , 061 predicted or known genes ( ∼90% genes in the genome ) to examine the differences in gene expression profiles between the hypoxia-selected ( AF ) and naïve control ( NF ) flies [15]–[17] . We performed these studies in both larvae and adults to determine gene expression as a function of development . Furthermore , we used a combination of bioinformatic , molecular and genetic strategies to investigate the role of specific genes in hypoxia tolerance .
Twenty-seven wild-type isogenic lines constituted the parental population that we used for long-term experimental selection of a Drosophila melanogaster strain . At baseline , there was significant variability in hypoxia tolerance among these 27 lines , as determined by eclosion rate under 5% O2 and recovery time from anoxic stupor [17] . In order to determine the level of O2 at which to initiate the long-term selection experiment , we performed a pilot study by culturing F1 embryos of the parental flies under different levels of hypoxia ( e . g . , 8% , 6% or 4% O2 ) ( Figure 1A ) . We found that their survival rate was reduced at 6% O2 , and no adult flies were actually obtained at 4% O2 . At 8% O2 , however , the majority of embryos ( >80% ) completed their development and reached the adult stage . Therefore , hypoxia selection was initiated at 8% O2 , an O2 level in which flies can develop throughout their life cycle . The O2 concentration was then gradually decreased by ∼1% every 3 to 5 generations to maintain the selection pressure ( except for the transition between 5% and 4% which necessitated a much longer time ) . By the 13th generation , flies were able to complete development and perpetually live in 5% O2; and by the 32nd generation , the AF flies could even live perpetually under a severer level ( 4% of O2 ) , a lethal condition for NF . We hypothesized that this is due to , at least partially , newly occurring mutations or recombination and selection of favorable alleles in the AF population . To test this hypothesis , a subset of embryos obtained from selected flies were collected and cultured under normoxia for several consecutive generations . After 8 generations in normoxia , AF were re-introduced into the lethal hypoxic environment ( 4% O2 ) , and again , the majority ( >80% ) of the flies completed their development and could be maintained in this extreme condition perpetually . This result strongly suggested that the hypoxia-tolerance in the selected flies is a heritable trait . Several remarkable phenotypic changes were observed in the hypoxia-selected flies ( AF ) . As described previously [17] , the AF flies have a shortened recovery time from anoxia-induced stupor [6] , consume more oxygen in hypoxia , and show a significant reduction in body weight and size . We also demonstrated that this reduction in size is due to the reduction in both cell number and cell size [17] . In the hypoxia chambers , at 5% O2 levels , adult flies had significantly decreased body weight: the decrease in male body weight was about 25% and was further reduced to about 40% under 4% O2 ( Figure 1C and 1D ) . Interestingly , the reduction in body weight and size were reversed to normal when the AF embryos were grown under normoxia . Life span was also studied in our selected and naïve flies . Hypoxia-selection did not affect lifespan ( Figure 1B ) . Global gene expression profiles were examined in hypoxia-selected Drosophila melanogaster in the 3rd instar larval stage and in adults using cDNA microarrays that contained 13 , 061 known or predicted genes of the Drosophila melanogaster genome [16] , [18] . After analyzing the data sets with a significant cutoff of >1 . 5 fold difference and a false discovery rate ( FDR ) of <0 . 05 [19] , 2749 genes ( 1534 up- and 1215 down-regulated ) were significantly altered in the larval stage , but only 138 genes ( ∼20 times less than those in larvae ) met this criteria ( 95 up- and 43 down-regulated ) in the adult . The complete list of differentially regulated genes is detailed in Tables S1 and S2 . Among them , 51 genes were found to be altered in both larval and adult stages with 23 up-regulated and 7 down-regulated genes ( Figure 2A ) . Interestingly , most of the commonly up-regulated genes encode proteins that are related to immunity , and the majority of the commonly down-regulated genes encode proteins that are related to metabolism . The significantly altered genes were further analyzed , based upon the Gene Ontological categorizations ( GO ) [20] , by GenMAPP and MAPPFinder [20] , [21] . Less than half of the differentially expressed genes ( ∼40% ) encode for proteins whose functions have not been characterized; the remaining encode proteins that are involved in numerous biological functions , such as development , metabolism , defense mechanisms and signal transduction ( Tables S3 and S4 ) . More than 30 biological processes were found to be altered in both larvae and adult and these were mostly related to either defense ( especially immune responses , p<0 . 05 ) or metabolism ( especially carbohydrate and peptide metabolism , p<0 . 05 ) ( Figure 2B ) . The most affected processes in larvae were related to metabolism ( 1044 genes , especially carbohydrate metabolism , 135 genes , p<0 . 01 ) . In addition , multiple components of signal transduction pathways were identified to be significantly altered in larvae and these included EGF , insulin , Notch , and Toll/Imd signaling pathways ( Figure 3 , p<0 . 05 ) . To confirm the changes obtained from microarrays , 10 differentially expressed genes were randomly chosen and their expression levels were determined by real-time qRT-PCR using specific primers in larval samples ( Table S5 ) . The microarray results and the real-time PCR gave similar trends ( r = 0 . 85 , Figure S1 ) . Significant gene changes were identified in the family of genes regulating cellular respiration in AF , especially in larvae . The majority of these changes consisted of a down- rather than an up-regulation of gene expression ( Table S3 ) . Besides one pyruvate kinase isoform ( CG12229 ) , most of the genes encoding glycolytic enzymes were dramatically down-regulated ( Figures 4A and 4B ) . Similarly , suppression was also found in the TCA cycle ( Figures 4A and 4C ) , lipid β-oxidation , and respiratory chain complex genes in larvae ( Figure 5 ) . As shown in Figures 5B and 5C , among the 50 measured genes encoding components of the respiratory chain , 33 genes were down-regulated , and only 3 genes were up-regulated . Interestingly , the suppression of these metabolic genes occurred only in the larvae and not in the adult fly ( Figure 5C ) . Since many genes encoding metabolic enzymes were significantly down-regulated , we asked whether such down-regulation was coordinated at a transcriptional level . Therefore , the GenomatixSuite ( GEMS ) software was used to identify transcription factor binding elements in defined cis-regulatory regions of the TCA cycle , glycolysis and lipid β-oxidation genes . The TCA cycle related genes were separated into two groups , one containing the significantly down-regulated genes ( down-regulated group , 16 genes ) and the other containing those that were not significantly altered or up-regulated genes ( reference group , 8 genes ) ( Table S6 ) . The binding elements of the Drosophila transcriptional suppressor hairy were present in the regulatory region of the down-regulated genes ( 15 of 16 cis-regulatory regions , 0 . 88/kb ) but not in the reference group ( 1 out of 7 cis-regulatory regions , 0 . 15/kb ) ( p<0 . 0001 , CHI-Test ) ( Figure 6A ) . Of particular interest , the expression level of hairy was significantly up-regulated in AF ( Figure 6B , Table S1 ) . This result suggested that hairy , a key transcriptional suppressor , reduced the expression of the TCA cycle genes . No such specific transcription factor binding elements were found in genes encoding glycolysis or β-oxidation enzymes . To confirm that hairy directly binds to the cis-regulatory regions of these candidate TCA cycle genes , chromatin immunoprecipitation ( ChIP ) assay was performed using a specific antibody to hairy in Drosophila Kc cells . The candidate hairy binding targets were tested using specific primers in both hypoxia and normoxia ( Table S5 ) . For a negative control , we included a cis-regulatory region of another TCA cycle gene ( i . e . , CG6629 ) that was up-regulated in AF and had no hairy binding elements detected . We found that hairy did bind to the cis-regulatory region of gene l ( 1 ) G0030 in hypoxia but not in normoxia . Similarly , hairy was also found to bind to the cis-regulatory region of gene SdhB under both hypoxic and normoxic conditions , and its binding activity was significantly higher in hypoxia than in normoxia . Such hypoxia-induced increase in hairy binding did not occur in CG12344 , a hairy target gene , which encodes for a non-metabolic gene ( i . e . , an isoform of GABA-A receptor ) . This result demonstrates that the down-regulated TCA cycle genes are direct targets of hairy , and hairy specifically suppresses their expression under hypoxia . Furthermore , such hypoxia-induced suppression of TCA cycle genes was abolished in the hairy loss-of-function mutants , h1 or h1j3 ( Figure 7A ) . To further evaluate the role of hairy in hypoxia tolerance , we determined the survival rate of these two hairy loss-of-function mutants at 6% of O2 . This mild level of O2 was used since it is sufficient to show differences between the mutants and controls . As shown in Figure 7B , both hairy loss-of-function mutants exhibited much lower survival rate ( p<0 . 0001 , CHI-Test ) as compared to controls , proving the contribution of hairy to hypoxia tolerance in flies . Since some of the current experiments suggested that hypoxia tolerance was a heritable trait in the AF population , we studied these flies further to determine the genetic basis of this heritability . Two types of experiments were performed . First , we examined the activity levels of 2 key glycolytic enzymes , triose phosphate isomerase ( TPI ) and pyruvate kinase ( PK ) in NF and AF . We argued that , if there was a genetic basis for the change in enzyme activity , such activity level would be altered not only in the hypoxia-cultured flies but also in those cultured in normoxia , i . e . , the enzyme activity would be different from that in NF , whether in hypoxia or normoxia . As shown in Figures 8A and 8B , the activity of TPI in AF was indeed reduced when these flies were cultured in either normoxia or hypoxia . When PK activity was assessed , however , AF had a higher level activity under hypoxia and stayed at the same level in normoxia without a significant increase as compared to NF . Second , the genomic locus encoding TPI was sequenced to determine whether there were any differences between AF and NF . We chose to sequence the TPI locus , because , unlike PK , TPI is encoded by a single gene . As expected , a number of significant polymorphic differences were identified in the AF population that included 3 SNPs ( −77T/C , −53A/G , −51A/T ) and 1 indel ( −74 to −66 ) in the cis-regulatory region , 1 synonymous SNP ( 1051A/G ) in the coding region , 1 SNP ( 1480A/G ) in the 3′-untranslated region , and 1 SNP ( 1662C/T ) in the downstream region ( Figure 8C , Table S7 ) , demonstrating significant genetic differences in the AF population ( p<0 . 001 , CHI-Test ) .
The current laboratory selection experiments have shown that the hypoxia-selected flies have garnered , with ongoing generations , a spectacular ability to survive extremely low O2 conditions . To put this ability in perspective , these flies live perpetually now at an oxygen level that is equivalent to that above Mount Everest by ∼4 , 000 meters . This phenotypic breakthrough occurred after 32 generations of selection . One of the most profound phenotypic abnormalities is a reduction in body size in the AF flies . This substantial decrease in body size was presumably of physiologic relevance , which is likely related to shorten the O2 diffusion distances for improved survival . It is very interesting to note that , though we show evidence of heritability of the hypoxia survival trait in the selected flies , the change of body size and weight did not seem to be heritable , as body size was reverted to normal , even after a single generation in room air . The goal of the current study was to determine the genetic and molecular basis of adaptation to long-term ( i . e . , over generations ) hypoxic environments . Since survival under long term hypoxia is a complex trait , we expected this adaptation to be controlled by a number of pathways and genes . Indeed , our microarray and genetic analyses have shown that the survival of both larvae and adult flies was accompanied by alterations in a number of major signaling pathways , such as EGF , Insulin , Notch and Toll/Imd pathways . In this regard , there are a number of interesting observations made from our studies . First , the number of significantly altered genes was much larger in larvae than in the adults . More than 20% of all measured genes were significantly altered in the larval samples whereas only ∼1% changed in the adults , using the same statistical criteria . This major difference between larvae and adults may have a number of explanations , including the fact that larval cells undergo rapid growth , proliferation and differentiation as compared to adults . Second , although adaptive changes were found in both larvae and adult flies , it is possible that these changes were induced through both genetic and/or epigenetic physiologic regulation . Since we found that a ) the hypoxia-selected flies could live for a number of generations in normoxia and survive again when introduced back in 4% hypoxia , b ) some glycolytic enzymes maintained their activity in the AF population at a lower ( or higher ) level not only during hypoxia but also in normoxia ( e . g . , TPI and PK ) , and c ) a number of polymorphic changes existed in AF as compared to NF , our data provided direct evidence for actual genetic alterations that were responsible for , at least in part , tolerance to severe hypoxia . Third , several signal transduction pathways were altered in the selected flies , i . e . , EGF , Insulin , Notch and Toll/Imd pathway , which have also been found to be activated by hypoxic stimuli in mammalian tissues/cells [22]–[28] . For example , we and others have shown that the Notch receptor and its targets were altered in mouse heart [29] and cultured cells [30] , [31] following chronic hypoxia . Such similarities in hypoxia responses demonstrated that some of the mechanisms underlying hypoxia-tolerance are conserved across species . Fourth , previous studies have demonstrated an enhancement of glycolytic activity as a major metabolic response in cells/tissues following acute ( in minutes or hours ) hypoxia , in contrast , our present data showed that many genes encoding glycolytic enzymes were down-regulated in the AF . Moreover , no significant lactate accumulation was found in the AF flies ( unpublished observations from our laboratory ) . Finally , although hypoxia-tolerance is a complex trait , a single gene alteration , such as that of hairy , can remarkably make a significant difference in terms of survival . For more than three decades , physiologists have debated the importance of down-regulation of metabolism in hypoxia tolerance in some tolerant animals such as the turtle . For example , Hochachka and others have argued that anoxia-tolerant organisms depress their metabolism in order to minimize the mismatch between supply and demand [32]–[34] . While this idea , based on metabolic data , is intuitively appealing , there is no information about how various metabolic enzymes could be coordinated in order to survive severe long lasting hypoxia . Our current work provides the first evidence showing that such coordination can be achieved at a transcriptional level by a seemingly metabolic switch , i . e . , hairy , which is a transcriptional suppressor . This notion is supported by a number of observations from current study: a ) expression of hairy was up-regulated in AF , b ) hairy-binding region was presented in the cis-regulatory regions of the down-regulated genes but not others , c ) the functional ChIP analysis led us to believe that the down-regulation of these metabolic genes is based on the changes of hairy , especially that its binding was not increased for other , non-metabolic target genes , e . g . , CG12344 , and d ) hairy loss-of-function mutations abolished the suppression on the expression of TCA cycle genes and significantly reduced hypoxia survival in Drosophila . This decreased survival in flies carrying the hairy mutation is particularly important since these experiments strongly demonstrate that hairy contributes to hypoxia tolerance through the regulation of TCA cycle enzymes . While it is possible that the decreased survival is not related to the role that hairy plays in metabolic regulation but due to an inherent weakness of the hairy mutants , this is much less likely because a ) we have studied in the past different mutation alleles in flies and they do not necessarily behave differently in hypoxia [6] , and b ) we chose two different mutants of hairy and they have similar effects . Another interesting observation in this work is related to the enzymatic activities of triose-phosphate isomerase ( TPI ) and pyruvate kinase ( PK ) in AF . We hypothesized that , if there is a genetic basis for the change in enzyme activity , we should be able to detect the alterations in AF flies that were cultured in both hypoxia and normoxia , and the enzyme activity should maintain different from those in NF , whether in hypoxia or normoxia . Indeed , as expected , we found that the expression level and enzymatic activity of TPI were significantly lower in AF than those in NF in both hypoxia and normoxia . In addition , we identified 7 polymorphic differences in the genomic locus of the TPI gene in AF which demonstrated a genetic modification of this locus in AF strain . In contrast , the activity of PK in AF did not differ from that of NF in normoxia , although its activity was significantly higher in AF under hypoxia ( Figure 8B ) . This result demonstrated that there was an inhibition of this enzyme in NF under hypoxia , and such hypoxia-induced inhibition was abolished in AF . On the surface , this result might weaken our argument , a potential explanation for such a difference between PK and TPI is that the regulation of PK is much more complicated . Indeed , there are 7 genes encoding PK in D . melanogaster . In AF , we found that 1 gene ( i . e . , Pyk ) was down-regulated and another gene ( i . e . , CG12229 ) was up-regulated . Therefore , in addition to possible genetic rearrangements , other factors may play a role to keep higher activity of PK in AF under hypoxia and to minimize the difference between AF and NF in normoxia . If the changes that led to enhanced survival in extremely low O2 levels in the AF population are genetic in nature , as we demonstrated in this work , it is important to note that the breakthroughs in survival to lower and lower O2 levels took place after a relatively small number of generations in the Drosophila . This relatively small number of generations was also found when previous investigators were studying other phenomena such as geotaxis [15] . If translated into human years , the 32 generations that were needed to alter the phenotype and genotype in Drosophila can be approximated to about a thousand years . Although changes in the DNA code with selection pressure could take a much longer period of time ( potentially thousands and millions of years in “Darwinian time” ) , this work demonstrated that such changes in DNA and phenotype could proceed at a much faster rate , presumably because of the trait itself or the selection pressure applied . In summary , we have generated a Drosophila melanogaster strain that is very tolerant to severe hypoxic conditions through long-term experimental selection . Several adaptive changes have been identified in the hypoxia-selected AF flies that include up-regulation of multiple signal transduction pathways , modulation of cellular respiration enzymes , and polymorphic differences in metabolic enzymes such as TPI . While we believe that multiple pathways contribute to hypoxia-tolerant trait in this Drosophila strain , we demonstrate that hairy-mediated metabolic suppression is an important one . The adaptive mechanisms identified in this hypoxia-tolerant Drosophila model may also play a crucial role in protecting mammals from hypoxia injury .
Twenty-seven isogenic Drosophila melanogaster lines ( kindly provided by Dr . Andrew Davis ) were used as parental stocks for the long-term hypoxia-selection experiment as described previously [17] . Briefly , embryos collected from the parental population were divided into 6 groups , 3 groups of them were subjected to long-term hypoxia-selection , and 3 other groups were maintained under normoxia as controls . Both hypoxia-selection and control experiments were performed in specially designed population chambers ( 26 cm×16 cm×16 cm ) . These chambers were connected to either O2 at certain concentrations ( balanced with N2 , for the hypoxia-selection experiments ) or to room air ( 21% O2 , for the control experiments ) . The humidity in the chambers was maintained by passing the gas through water prior to going into the chambers . The flow speed was monitored by 565 Glass Tube Flowmeter ( Concoa , Virginia Beach , VA ) , and the O2 level within the chamber was monitored with Diamond General 733 Clark Style Electrode ( Diamond General Development Corp . , Ann Arbor , MI ) . The selection was started at 8% O2 and this concentration was gradually decreased by 1% each 3 to 5 generations to keep the selection pressure . Embryos , 3rd instar larvae and adult flies were collected from each generation and stored at −80°C for analyses . The results presented in the current study were derived from expression arrays using larval and adult samples . The body weight of hypoxia-selected flies was determined at each generation . Male flies ( n = 100 ) from hypoxia or control chambers were collected , weighed and used as the index of body weight . The hairy loss-of-function mutants ( h1 and h1j3 ) were obtained from Drosophila Stock Center ( Bloomington , IN ) . The survival rate of these stocks in hypoxia was determined by culturing them in a 6% O2 environment . After 3 weeks in culture , the number of live adult flies and pupae were counted . The ratio between live adult flies to the total number of pupae was calculated and presented as survival rate . The statistical significance of survival between hairy mutants and controls was calculated by CHI-test . cDNA microarrays containing 13 , 061 known or predicted genes of the D . melanogaster genome were processed according to previous descriptions [16] , [35] . Nine larval samples from AF or NF , 6 adult samples from AF or NF were included in this analysis . Each larval sample contained a pool of 25 3rd instar larvae , and each adult sample contained 25 male and 25 female adult flies from each individual population . Total RNA was extracted from the samples using TRIzol ( Invitrogen , Carlsbad , CA ) followed by a clean-up with RNeasy kit ( Qiagen , Valencia , CA ) . Three µg of total RNA from each sample was amplified with an in vitro transcription-based strategy [13] . A common reference design was applied for the hybridizations , and the reference RNA sample was created using a balanced pool of 3rd instar larvae ( for the larval samples ) or adult flies ( for the adult samples ) from each parental line . This reference was chosen so that relative abundance of each transcript could be calculated individually , and the relative levels of each transcript among biological replicates could be compared . A total of 30 arrays were included in this analysis , and the hybridizations were done in different days using arrays printed from different batches . Microarray images were acquired by GenePix 4000 microarray scanner using GenePix Pro 3 microarray analysis software ( Axon Instruments , Sunnyvale , CA , USA ) . The statistical significance ( q-value , i . e . , false discover rate ( FDR ) ) and the ratio of the changes in expression was calculated using Significance Analysis of Microarray ( SAM ) software [19] following LOWESS normalization . The fold changes were presented as ratios , if up-regulated , or −1/ratio , if down-regulated . The gene ontology ( GO ) based analyses were performed using GenMAPP software [20] . The microarray data can be retrieved using access number GSE8803 in the GEO database at http://www . ncbi . nlm . nih . gov/geo . Semi-quantitative real-time RT-PCR was used to evaluate the result of microarrays and to determine the differences in expression levels of genes encoding TCA cycle enzymes between the hairy mutants and control . All specific primers were designed by Primer 3 software [36] and synthesized at ValueGene ( San Diego , CA ) ( Table S5 ) . First strand cDNA was synthesized using SuperScriptII reverse transcriptase and Oligo- ( dT ) primer . Real-time PCR amplification was performed using ABI Prism 7900HT Sequence Detection System ( Applied Biosystems , Foster City , CA ) . For each reaction , 10 µl of 2× SYBR green PCR master mix ( Applied Biosystems , Foster City , CA ) and 0 . 5 µM of both forward and reverse primers along with 100 ng of each appropriate cDNA samples were mixed ( total reaction volume: 20 µl ) . Melting curves were determined and the final products were isolated with 4% agarose gel to ensure specificity of the reaction . The relative expression level was calculated using 2−ΔΔCt method , as described previously [37]–[39] . Drosophila melanogaster β-actin was used as internal control . The final results were presented as fold change of AF over NF or hairy mutants over yw control . All experiments were done in triplicate . The enzymatic activity of Triose Phosphate Isomerase ( TPI ) and Pyruvate Kinase ( PK ) were determined as previously described with modifications [40] . Briefly , the assay samples were extracted from 3rd instar larvae cultured under either normoxic or hypoxic condition ( 4% O2 ) . 0 . 2 ml ( 100 mg wet weight/ml ) of isolation medium ( 0 . 25 M sucrose , 1 mM EDTA-K , 5 mM HEPES-Tris , pH 7 . 4 , with protease inhibitors ) was added and the suspension of larval tissue in isolation medium was transferred into a 2 ml all glass homogenizer . 10% ( v/v ) Triton X-100 was added to this suspension making the final concentration of Triton X-100 0 . 5% ( v/v ) . Then the tissue was homogenized with the B ( i . e . , tight ) pestle . Aliquots of this homogenate were employed for enzymatic activity measurements . The enzymatic activities of TPI or PK were determined using 10 , 20 , or 30 µl of the homogenate with proper substrates . The dynamic of color formation was recorded using Beckman Coulter DU800 Spectrophotometer at selected wave length for each substrate . The kinetic parameters of the reaction were calculated by curve fitting . The common transcription factor binding sites in the defined promoter regions of candidate genes were analyzed using GenomatixSuite ( Genomatix Software GmbH , Germany ) . The genes encoding TCA cycle or β-oxidation enzymes were separated into down-regulated or reference groups . The genome DNA sequences from 2000 bp upstream of the first transcription starting site ( TSS ) to 500 bp downstream of the last TSS of each gene was downloaded and used as cis-regulatory region of the gene ( Drosophila genome R5 . 2 ) [41] . These sequences were subjected to GEMS analysis to identify common transcription factor binding elements [42] . The statistical significance of the transcription factor binding element frequency in the AF and NF population was analyzed by CHI-test . Chromatin immunoprecipitation and PCR detection of hairy binding in the cis-regulatory regions of genes encoding TCA cycle enzymes were performed in cultured Drosophila Kc cells ( obtained from Dr . Amy Kiger , UCSD ) [43] , [44] . The Kc cells were treated with 0 . 5% O2 for 4 hours . About 106 cells were used in each ChIP experiment . Chromatin immunoprecipitation was carried out using ChIP assay kit ( Upstate , Temecula , CA ) according to the manufacture's instructions . The immunoprecipitation was performed overnight at 4°C with 2 µg of hairy antibody ( Abcam , Cambridge , MA ) . DNA fragments were purified with phenol:chloroform ( Invitrogen , Carlsbad , CA ) . For PCR , 2 µl of a 25 µl DNA extraction was amplified with specific primers ( Table S5 ) . Genomic DNA was extracted from 15 male AF or NF adult flies . The samples were ground in 400 µl of homogenate buffer ( 100 mM Tris/HCl , 100 mM EDTA , 100 mM NaCl and 0 . 5% SDS , pH7 . 5 ) and incubated at 65°C for 30 min . Genomic DNA was extracted by adding in 800 µl extraction solution ( 1 2 . 5 ( v/v ) of 5 M KAc and 6 M LiCl ) . After 15 min of centrifugation at 13 , 000 rpm , the supernatant was transferred into a new tube and the genomic DNA was precipitated by adding 600 µl of isopropanol . The DNA pellet was washed with 70% of ethanol and dissolved in TE buffer . The genomic locus encoding TPI gene was amplified by PCR using specific primers ( forward primer: GTTTAAGGTCCGCAGAGGTG , and reverse primer: ATTTTGGCAAGCCTGTTGAT ) . All the coding exons and intronic flanking regions were amplified by polymerase chain reaction ( PCR ) using the high fidelity proofreading DNA polymerase , Platinum Pfx DNA Polymerase ( Invitrogen , Carlsbad , CA ) . The PCR products were purified and cloned into pCR4-TOPO plasmid to create the plasmid library for TPI alleles of the AF and NF population . Cycle sequencing was performed on an ABI automated sequencer ( Applied Biosystems , Foster City , CA ) by Eton Biosciences , Inc . ( San Diego , CA ) . Ten clones from AF or NF TPI allele library were sequenced and compared by using ClustalW2 software [45] ( http://www . ebi . ac . uk/Tools/clustalw2/ ) and DnaSP 4 . 0 [46] . The statistical power for the comparison was calculated by GPower software 3 . 0 . 3 [47] . | Hypoxia-induced injury has been related to multiple pathological conditions . In order to render mammalian cells and tissues resistant to low O2 environment , we wished to first understand the mechanisms underlying hypoxia-tolerance in resistant animals . Therefore , we generated a D . melanogaster strain that is tolerant to severe hypoxic conditions through long-term experimental selection . Several adaptive changes were identified in the hypoxia-selected flies that included up-regulation of multiple signal transduction pathways ( such as Notch pathway , Insulin pathway , EGF receptor pathway , and Toll/Imd pathway ) , modulation of cellular respiration enzymes , and polymorphic differences in metabolic enzymes ( such as TPI ) . While we believe that multiple pathways contribute to the hypoxia-tolerant trait in this Drosophila strain , we demonstrate that hairy-mediated metabolic suppression is a critical mechanism for reducing the mismatch between supply and demand of O2 . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"genetics",
"and",
"genomics/gene",
"expression",
"genetics",
"and",
"genomics/functional",
"genomics",
"genetics",
"and",
"genomics/physiogenomics",
"genetics",
"and",
"genomics/gene",
"function",
"genetics",
"and",
"genomics"
] | 2008 | Mechanisms Underlying Hypoxia Tolerance in Drosophila melanogaster: hairy as a Metabolic Switch |
We describe a comprehensive and general approach for mapping centromeres and present a detailed characterization of two maize centromeres . Centromeres are difficult to map and analyze because they consist primarily of repetitive DNA sequences , which in maize are the tandem satellite repeat CentC and interspersed centromeric retrotransposons of maize ( CRM ) . Centromeres are defined epigenetically by the centromeric histone H3 variant , CENH3 . Using novel markers derived from centromere repeats , we have mapped all ten centromeres onto the physical and genetic maps of maize . We were able to completely traverse centromeres 2 and 5 , confirm physical maps by fluorescence in situ hybridization ( FISH ) , and delineate their functional regions by chromatin immunoprecipitation ( ChIP ) with anti-CENH3 antibody followed by pyrosequencing . These two centromeres differ substantially in size , apparent CENH3 density , and arrangement of centromeric repeats; and they are larger than the rice centromeres characterized to date . Furthermore , centromere 5 consists of two distinct CENH3 domains that are separated by several megabases . Succession of centromere repeat classes is evidenced by the fact that elements belonging to the recently active recombinant subgroups of CRM1 colonize the present day centromeres , while elements of the ancestral subgroups are also found in the flanking regions . Using abundant CRM and non-CRM retrotransposons that inserted in and near these two centromeres to create a historical record of centromere location , we show that maize centromeres are fluid genomic regions whose borders are heavily influenced by the interplay of retrotransposons and epigenetic marks . Furthermore , we propose that CRMs may be involved in removal of centromeric DNA ( specifically CentC ) , invasion of centromeres by non-CRM retrotransposons , and local repositioning of the CENH3 .
LTR retrotransposons are useful tools for understanding genome evolution because of their target site specificity and our ability to estimate their insertion times based on sequence divergence of their LTRs [1] . Retrotransposons account for >75% of the maize genome sequence [2] and are responsible for much of the genome expansion that has taken place since the allotetraploidization event that gave rise to present day maize [3] , [4] . Centromeric retrotransposons ( CR ) were initially discovered as centromere-specific sequences in the grasses [5] , [6] . The CRs of maize ( CRM ) and rice ( CRR ) belong to distinct subfamilies [7]–[9] , which have been grouped most recently into four orthologous subfamilies [9] . One of these subfamilies , CRM1 , has proliferated extensively in the past 3–4 million years by generating at least 5 recombinant subgroups from two parental variants thought to have been combined in the maize genome during allotetraploidization [10] . No full-length element of the CRM1-orthologous rice subfamily ( CRR3 ) is found in the O . sativa ssp . japonica genome , raising doubt as to whether CR elements in general , and CRM1 in particular , are required for centromere function . With the exception of members of the recently discovered CRM4 subfamily , all known CRM elements localize almost exclusively to centromere regions as determined by fluorescence in situ hybridization [11] , [12] , and physical mapping [2] . The mechanism of centromere localization is as yet unknown . Like the centromeres of most eukaryotes , plant centromeres also contain tandem satellite repeats [7] , [13]–[16] . Tandemly arranged CentC repeats ( monomer length ≈156 nt ) and interspersed CRM are the major DNA components of maize centromeres [7] , [15] , [17] , but their role in centromere function is unclear . The satellite sequences of corn and rice , which diverged from a common ancestor approximately 50 MYA [18] , [19] , exhibit regions of high sequence similarity [20] and are clearly homologous . Functional centromeres of all eukaryotes examined to date are marked epigenetically by a centromeric histone H3 ( CENH3 ) , which replaces the canonical histone H3 in centromeric nucleosomes [21] . A key question in centromere biology is how deposition of CENH3 in centromere regions is controlled . Chromatin immunoprecipitation ( ChIP ) experiments with anti-CENH3 antibodies is an effective method for isolating centromeric chromatin [15] , and has been used previously to perform a comparative study of rice centromeric satellite sequences [20] and to precisely delineate the borders of several rice centromeres [22] . Excellent cytogenetic and genetic resources , including oat-maize addition lines that carry a single maize chromosome in an oat background [23] , together with the recently published reference genome [2] of the maize inbred B73 ( ZmB73v1 ) , make maize a good model for studying centromeres . Here we present the physical maps of maize centromeres 2 and 5 , on which the functional centromeres have been precisely delineated using anti-CENH3 ChIP sequences . The highly active retrotransposon population of maize provides a detailed record of centromere evolution that is unattainable from smaller genomes with fewer or less active retrotransposons .
Two methods were employed to identify molecular markers that can be used to genetically map maize centromeres , which consist largely of repetitive sequences . We used both the repeat junction method [24] and transposon display [25] , [26] with CRM2 to generate a total of 54 centromere-derived polymorphic markers ( Tables S1 , S2 ) that could be placed onto the maize genetic map using a mapping population [27] derived from inbreds B73 x Mo17 . This simultaneously anchored centromeric BACs to their respective chromosomes ( Tables S1 , S2 ) and provided the genetic map positions for all ten centromeres ( Table 1 ) . Using BAC sequence data from the Maize Genome Project [2] , fingerprinted contigs ( FPC ) data from the Arizona Genomics Institute [28] ( ftp://ftp:agiftpguest@ftp . genome . arizona . edu/pub/fpc/maize/ ) , and the centromeric markers described above , we were able to construct physical maps traversing the entire centromere on chromosomes 2 and 5 . Our BAC-based physical maps for these two centromeres are largely in agreement with the reference chromosomes presented of the B73 reference genome ZmB73v1 [2] , thus reference chromosome coordinates are provided for the features we describe here . The main difference between these maps is the closure of a gap on centromere 5 ( position 105 , 074 , 634 ) using the CentC-rich singleton BAC ZMMBBb0271K07 , which has not yet been incorporated into reference chromosome 5 . Even excluding this BAC , the CentC content of centromere 5 is about 3 times higher than that of centromere 2 . Centromeres 2 and 5 of B73 contain very little CentC as compared to the other eight centromeres [29] . Fiber FISH using CRM and CentC probes on B73 oat-maize addition lines carrying a single maize chromosome ( 2 or 5 ) , indicate that CentC repeats are confined to a few small blocks interspersed with CRM in both of these centromeres ( Figure S1 and Figure 1 , respectively ) . Measurements of the stretched chromosomes show that these CentC blocks of centromeres 2 and 5 span approximately 196 kb ( Figure S1 ) and 192 kb ( Figure 1 ) , respectively . The physical map of centromere 2 ( for a graphical representation of the entire region please see [30] ) is in good agreement with the FISH data as it contains a number of short CentC repeat clusters totaling about 31 kb and ranging in size from about 1 kb to 15 kb . These clusters span an approximately 130 kb region near the center of the functional centromere [30] , which is close to the fiber FISH estimate . The difference between the two maps is most likely due to the fact that the physical map still contains numerous gaps and consists of relatively small sequence fragments of unknown order and orientation . Similarly , the physical map of centromere 5 shows one major region of CentC spanning 246 kb [30] , and a repeat arrangement similar to that shown by fiber FISH ( Figure 1 ) , i . e . distinct CentC- and CRM1-rich regions . CRM1 and CRM2 constitute the majority of centromeric repeats ( CRM and CentC ) present in these two centromeres ( 94% and 80% for centromeres 2 and 5 , respectively ) , but the ratio of CRM1 to CRM2 in centromere 5 is about double that of centromere 2 ( Table 2 ) . We used ChIP with anti-CENH3 antibody followed by pyrosequencing to generate 149 , 756 mostly centromere-derived DNA sequences of maize inbred B73 with an average high quality read length of 165 nt and totaling 24 , 729 , 204 nt . The availability of high quality sequence covering all regions of the maize genome represented in FPC contigs of the AGI physical map [31] allowed us to map the immunoprecipitated sequences onto the physical map using MUMmer and BLAST [2] , thereby delineating the functional centromeres on all ten reference chromosomes ( Figure 2A , Figure 3A , Figure S2 ) . MUMmer , which was used to map reads to the genome at 100% identity over 100% of the read length , allowed us to anchor 44 , 897 ChIP sequences . Of the remaining sequences , 59 , 913 were mapped by BLAST using cutoffs of 96% identity over 96% of the ChIP read length . The reads that could not be mapped using these BLAST parameters likely represent centromeric regions that are missing in the ZmB73v1 reference genome assembly , which contains only an estimated 54% of the genome's total CentC content [2] . The BLAST and MUMmer reads are graphed as moving averages onto the reference chromosomes – both peak at the regions of highest centromere repeat density on all chromosomes ( Figure 2A and 2B , Figure 3A and 3B , Figure S2 ) . On chromosome 2 , the arms exhibit a background signal of about 2 . 1 reads per 100 kb window , which is approximately 30 times lower than the read count of the centromeric peak ( Table S3 ) . This background signal is likely due to co-purification of non-centromeric chromatin during the initial chromatin immunoprecipitation with anti-CENH3 antibody , as reflected by the small amount of background signal visible on the chromosome arms in FISH performed with the ChIP fraction ( Figure S3 ) . The major FISH signal corresponds to the ten centromeres , indicating significant enrichment of the CENH3 chromatin fraction . Several smaller peaks formed by reads with less than 100% identity are found in euchromatic regions of several chromosomes and correspond to knob repeats or plastid sequences . Most chromosomes contain a single CENH3 peak that correlates with a high centromere repeat density ( Figure S2 ) . For the chromosomes containing more than one centromere peak , we were able to identify the correct centromere position using the genetically mapped centromeric markers ( Tables S1 , S2 ) . About 13 . 6% ( 20 , 441 ) of the ChIP reads were not mapped because they did not meet the minimum BLAST length or identity requirements . Many of these are likely to be centromeric as they were classified as CRM ( 1 , 936 ) or CentC ( 766 ) based on cross_match ( http://phrap . org ) of 100% of the read . Another 24 , 505 reads ( 7 , 330 CRM and 758 CentC ) mapped to multiple regions with identical bitscores and are also not graphed on the reference chromosomes . This illustrates the problem that , although we were able to reliably classify ChIP reads as CRM or CentC , mapping a read to a single location in the genome is possible only for those reads containing a unique SNP . As a result , it is difficult to determine if any given CRM element is associated with CENH3 nucleosomes , especially if it has inserted recently . The functional centromere 2 is defined by a single CENH3 binding domain of 1 . 8 Mb ( Figure 2A and 2C; Table S3 ) . This is relatively large compared to the size of the four best-sequenced rice centromeres , which span 420–820 kb [22] but move chromosomes that are on average about five times smaller than the maize chromosomes ( 41 Mb vs 200 Mb ) . On centromere 5 the mapped ChIP reads reveal two distinct CENH3-containing regions of sizes 3 . 2 Mb ( “L” = left ) and 1 . 0 Mb ( “R” = right ) , separated by a circa 2 . 8 Mb interstitial ( “I” ) region exhibiting near background ChIP levels and discernable even at the whole chromosome level ( Figure 3A and 3C; Table S4 ) . Both of these blocks are anchored to centromere 5 by a number of markers , including repeat junction , transposon display , oat-maize addition line and genetic markers ( Figure 3D , Table S7 , Table S8 ) , which provide a high confidence level of the accuracy of the physical map . Thus we are confident of the location of the “R” region even though this CENH3-rich region is virtually devoid of centromeric repeats , making it difficult to detect and verify by FISH . Note that a complete physical map traversing an entire centromere is required to detect multiple CENH3 domains , which may exist in some of the other eight centromeres for which the physical maps are not yet completely assembled . As detailed above , the “L” and “R” blocks of centromere 5 together are 2 . 3 times larger than the entire functional centromere 2 . Remarkably , the smaller centromere 2 CENH3 region contains a higher density of CENH3 ChIP reads , such that the total number of reads mapped to each centromere is 1 , 130 for centromere 2 ( 1 . 8 Mb ) and 1 , 562 ( 1 , 247 in “L” plus 315 in “R” ) for centromere 5 ( 4 . 2 Mb ) . Note that the number of reads mapped to each centromere is only an approximate and indirect estimation of the number of CENH3 nucleosomes , and that this number is heavily influenced by the number of unique targets available in each region to which the reads can be mapped . Nevertheless , it appears as though the difference in centromere size is compensated somewhat by the density of CENH3 nucleosomes , measured indirectly as 628 ChIP reads/Mb for centromere 2 , and 390 and 315 ChIP reads/Mb for centromere 5 “L” and 5 “R” , respectively . Figure 2B and 2E and Figure 3B and 3E illustrate the centromeric repeat content ( CRM and CentC ) of centromeres 2 and 5 in non-overlapping 100 kb segments . These repeats reach local maxima of up to 91% per 100 kb window ( chromosome 9 in Table S5 ) . For centromere 2 , repeat content of these windows correlates well with the CENH3 content ( Figure 2A and 2C ) . The central CentC region is flanked on both sides by CRM1 and CRM2 elements . CRM1 sequence is present at slightly lower levels than CRM2 throughout the functional centromere ( Table S3; Figure 2E ) and is found in small amounts in the flanking regions up to 2 Mb away . In addition to consisting of two distinct CENH3 domains , centromere 5 differs from centromere 2 in that the centromeric repeats are not distributed evenly . A small amount ( 17 kb ) of CentC lies outside of the functional centromere at 100 . 7 Mb . The larger CENH3 block ( “L” ) contains predominantly CRM2 and a smaller amount of CRM1 ( Table S4; Figure 3E ) . Unlike centromere 2 , the largest block of CentC in centromere 5 lies at the right edge of this block ( 105 Mb ) , the “L”/”I” border . A number of CRM elements have inserted into this CentC cluster , which is flanked on both sides by large amounts of CRM1 . This has resulted in a skewed CRM1/CRM2 distribution on centromere 5 , with a CRM2-rich region in the left half of “L” and a CRM1-rich region at the “L”/”I” border that extends about halfway into “L” on one side , and into the CENH3-poor interstitial region on the other . The second , smaller CENH3 block of centromere 5 ( “R” ) contains very little centromeric repeat . As was expected from published FISH experiments , CRM elements belonging to the CRM1 , CRM2 and CRM3 subfamilies are localized primarily to centromeres ( Figure 2B , Figure 3B , Figure S2 ) . However , the physical maps do reveal small amounts of CRM1 and CRM2 sequences on most chromosome arms that would be difficult to detect by FISH . In some cases , these sequences represent a single element that may have inserted aberrantly . For example , element CRM1_18 near the telomere of 5L ( position 213 , 233 , 223 ) , which encodes an otherwise functional polyprotein , contains a mutation in the conserved chromodomain that might have impaired target-specific integration of this element . Its 5′ and 3′ LTRs are identical , indicating that this element inserted within the past 150 , 000 years . Other CRMs may have been translocated to chromosome arms from an initially centromeric position as part of another retrotransposon or helitron , though we have found no evidence for this to date . While mindful of these exceptions , we postulate that CRM elements predominantly target functional centromeres , and that the CRM insertions dated by the method of San Miguel et al . [1] therefore represent a historical record of centromere location over evolutionary time . This is supported by the fact that virtually all CRM elements with identical LTRs ( κ = 0 ) are located within the current CENH3 region as delineated by the ChIP reads . We were able to date the insertion time of a large number of retroelements that inserted in or near the functional centromeres 2 ( 128 elements ) and 5 ( 246 elements ) . The locations and dates of these insertions provide a powerful tool for elucidating centromere dynamics over evolutionary time . In general , recently ( κ ≤0 . 01 ) inserted CRM elements are located within the CENH3 regions , while non-CRM retrotransposons that inserted during the same period tend to be present in higher numbers outside of the centromeres ( Figure 2C , Figure 3C ) . In accordance with the CRM1 element evolution described by Sharma et al . [10] , the youngest CRM1 element insertions that are located in the centromere 2 CENH3 region and centromere 5 “L” region consist exclusively of the most recently formed recombinants R4 and R5 , while the older CRM1 elements lie closer to the border or outside of the current CENH3 region on both centromeres and belong to the older recombinant ( R3 , R2 , R1 ) or parental ( A and B ) types ( Figure 4 ) . The fact that recent CRM1 insertions are located almost exclusively in the current CENH3-containing region while older CRM1 elements are located both within that region as well as in nearby chromatin , suggests that the CENH3-containing region , and thus the functional centromere , can shift locally over time . The centromere 5 picture is complex: a large number of non-CRM retrotransposons appear to have inserted into both the CENH3-rich and the surrounding regions . Also , a large number of CRM1 elements have inserted near the major CentC cluster on the border of the “L” and “I” regions ( Figure 3E , Figure 4B ) . Recent CRM insertions are located exclusively within the left functional domain , while the CRM1 elements in the “I” region have inserted at progressively older times the farther they are located from the CentC cluster ( see trend lines in Figure 4B ) . Conversely , the youngest non-CRM elements have inserted predominantly in the interstitial and pericentromeric regions . In other words , within the functional domain “L” it is the CRM elements that have inserted after the non-CRM elements , whereas in the interstitial region the CRM1 elements have inserted before the other types of elements . Finally , individual CRM1 elements of similar ( old ) age vary in the number of ChIP reads mapped to each element in accordance with their chromosomal location: those elements located within the “L” or “R” block exhibit a higher number of ChIP reads than elements of the same age located in the interstitial region or on the long arm of chromosome 5 ( Figure S4 ) . Taken together these observations indicate that CRM1 elements do not cause the formation of functional centromere chromatin but simply possess an extremely efficient mechanism that targets their insertion into CENH3-containing chromatin . A combination of four repetitive element probes allows identification of all B73 chromosomes in FISH experiments ( Figure 5 ) . Novel CRM1- and CRM2-specific probes were used to assess the distribution and arrangement of these two CRM subfamilies on metaphase chromosomes . While all centromeres contain visible amounts of both CRM1 and CRM2 , centromere 9 appears to contain relatively little CRM2 , which may explain our inability to derive CRM2 transposon display markers for this centromere . Figure 5 and Figure 6 further demonstrate that CRM1 and CRM2 elements are distributed in overlapping but somewhat distinct positions on the metaphase chromosomes . In general , CRM2 appears to localize to the exterior centromere face of chromosome 5 and other chromosomes , while CRM1 appears to be more prominent in the sister chromatid cohesion region , but with overlap clearly observed between the two probes ( Figure 5 , Figure 6 ) . Thus the FISH data are consistent with the CENH3 distribution inferred from our ChIP mapping , and we now have three lines of evidence suggesting that CRM2 is more closely associated with CENH3 than CRM1 . First , at the genomic scale , enrichment of CRM2 in CENH3 ChIP is about three times higher than CRM1 enrichment [2] . Second , the CENH3-containing regions of both centromeres 2 and 5 contain more CRM2 than CRM1 ( Tables S3 , S4 ) . Finally , FISH experiments with subfamily-specific FISH probes for CRM1 and CRM2 indicate that CRM2 appears to be preferentially localized to the exterior face of the centromere , whereas CRM1 localizes predominantly to the inter-chromatid region . However , although this tighter association of CRM2 with CENH3 is true when considering the distribution of CRM1 and CRM2 on a genomic scale , the physical maps of centromeres 2 and 5 clearly illustrate that the young subgroups of CRM1 ( R4 and R5 ) mirror the location of CRM2 ( Figure 4 ) , and that the likelihood of an element being located within the CENH3 region appears to be a function of the time at which a given CRM element has inserted rather than the subfamily to which it belongs .
We used two novel but generally applicable methods for deriving centromere markers that were critical for anchoring the centromeres to both the physical and genetic maps . First , we used a modified transposon display method [25] , [26] to screen a large number of potential centromeric markers for polymorphisms between the two parents of the IBM mapping population . In essence this is a centromere-specific AFLP screen that utilized the LTR of the abundant centromere-specific CRM2 retrotransposon as one of the priming sites . Polymorphic AFLP bands were mapped onto the IBM population and subsequently cloned , sequenced , and mapped onto the BAC sequences provided by the Maize Genome Sequencing Consortium [2] . The second method is based on the use of PCR primers derived from repeat junctions identified on centromeric , i . e . CentC- or CRM-containing , BAC clones [24] . JunctionViewer software [30] was used to identify repeat junctions located within 2 . 5 kb of each other ( e . g . resulting from nested insertions ) . Primers were subsequently designed on these junctions and tested for polymorphism between the mapping parents . Finally , polymorphic markers were mapped onto the genetic map using the IBM population . In contrast to the transposon display method , the junction method utilizes junctions between all types of centromeric repeats , and thus provided a complementary marker set , particularly for centromeres containing relatively little CRM2 ( e . g . centromere 9 ) . However , the large number of potential markers that had to be screened individually , as well as the very precise PCR reaction conditions required to produce differential amplification in the two mapping parents , made this a very labor-intensive method for finding centromere-specific markers . Both repeat junction and CRM2 display markers are dominant markers that were effective in anchoring centromeres to genetic and physical maps . The identification of novel centromeric markers using the repeat junction and transposon display methods , in combination with anti-CENH3 ChIP followed by pyrosequencing allowed us to precisely delineate the edges of the functional centromeres on all chromosomes . On most chromosomes the CENH3 nucleosomes map to a single region , but on several chromosomes additional peaks are observed . In some cases the additional peaks are caused by underlying knob repeats . However , unlike what is observed at centromere peaks , few if any reads that map to these knob repeats are 100% identical to their target . Therefore we believe that these peaks are generated by reads that originate from the estimated >90% of knob repeats that are absent from the maize reference genome ZmB73v1 [2] mapping to the best heterologous location available on the reference chromosomes . However , we cannot completely exclude the possibility that some knob repeats are associated with CENH3 in the other eight centromeres that have not yet been completely assembled . We were able to construct physical maps that traverse the entire B73 centromere region for two chromosomes , allowing us for the first time to analyze the repeat content and arrangement in the context of a complete maize centromere . These two centromeres are unusual in that they contain small amounts of CentC satellite as confirmed by FISH experiments [29] . However , other maize inbreds do contain large amounts of CentC in centromeres 2 ( B37 , KYS , W22 ) and 5 ( K10 , Stock6 ) [29] . The presence of the related CentO satellite in all rice centromeres , the fact that the related Tripsacum has high levels of CentC at all centromeres but much lower and highly variable levels of CRM on different centromeres [12] , and the fact that rice contains few CR elements compared to maize [9] lead us to believe that the CentC satellite represents an ancient form of centromere repeat and that the low CentC-containing centromeres 2 and 5 of B73 represent relatively recent changes . The restriction of recent CRM1 insertions ( κ ≤ 0 . 01 = <750 , 000 years ago ) on centromeres 2 and 5 to the current CENH3 domains indicates that these elements are equipped with an effective targeting mechanism that directs the majority of these elements into active centromeres . Chromatin components are thought to play a role in directing the yeast Ty elements to their chromosomal targets [32] . Furthermore , Lamb et al . [33] discovered that maize retrotransposon families are enriched in distinct patterns on maize chromosomes and noticed a correlation between insertion patterns of opie and prem2/ji with the modified histone H3K4me2 . Finally , the chromodomain of the fungal chromovirus MAGGY integrase protein has been shown to interact with a certain methylated histone H3 variant and direct integration of heterologous retroelements to chromosomal regions containing these variants [34] . Thus , although the exact targeting mechanism for CRM elements remains to be determined , CENH3 [8] or centromere-specific histone H3 methylation variants [35] , [36] represent plausible candidates for directing these elements to centromeres . Regardless of the targeting mechanism , CRM elements provide a record of the centromere location over evolutionary time and can be used to recreate centromere evolution . This is illustrated particularly well by the major CRM1 cluster of centromere 5 flanking the “L”/”I” border region and the major CentC cluster ( Figure 3C ) : for the CRM1 ( as well as the much less numerous non-autonomous CentA ) elements located in the interstitial region , there is a direct correlation between the element's distance from the “L”/”I” border and its insertion time , presumably because they were pushed away from the active centromere region by subsequent CRM1 insertions into the CENH3 region when it was centered on the CentC cluster . These CRM1 insertions , in turn , may pave the way for the insertion of other retrotransposons that lack the ability to insert into functional centromere regions , thus further increasing the distance between the old CRM1 insertions and the present day functional centromere , which essentially consists of a CRM region flanked by CentC satellite . Similar dynamics can be observed on centromere 2: the partial CRM1 elements at 92 . 5 and 92 . 7 Mb are older than those in or near the present-day functional centromere , from which these elements are separated by a number of more recently inserted non-CRM elements . In contrast to the centromere 5 CENH3 regions , which contain no ( “R” ) or only very recent CRM insertions ( left half of “L” ) , centromere 2 contains a continuous record of CRM element insertions at its present location ( Figure 2C ) , and therefore appears to have existed in this location for the past 3–4 million years . In contrast , centromere 5 seems to have undergone a significant lateral shift during this time period that appears to have contributed to its larger size and apparent lower CENH3 density . By extrapolating this process , i . e . alternating CRM and non-CRM element insertions leading to changes in centromere size and location , to the more distant past for which we lack a good retrotransposon insertion record ( because older insertions have been removed from the genome ) , the remodeling can be extended to the entire centromere 5 region as follows: the original CentC-rich centromere may have been invaded by an ancient CRM subfamily ( possibly CRM4 ) that split the CentC cluster in two and expanded the “L” region by making it accessible to non-CRM retrotransposons that make up the bulk of “L” . As a result , the left CentC cluster ( at 100 . 7 Mb ) is no longer associated with CENH3 . This was followed by insertions of CRM1 elements in the major remaining CentC cluster at the “L”/”I” border , which caused the separation of the L and R domains . This wave of CRM1 insertions may have also deleted CENH3-containing chromatin ( possibly CentC ) , which in turn may have caused the CENH3 domain to expand into the “L” region , opening this region to CRM1 and CRM2 elements while preventing non-CRM elements from inserting . Only these most recent waves of CRM1/2 insertions can be reconstructed , as older retroelement insertions are more likely to be partially or completely removed from the genome . The small number of retrotransposon insertions into the “R” region makes it difficult to reconstruct its history . One explanation for this dearth of CRM insertions is that the “R” region has formed relatively recently in response to the changes described above for the “L”/”I” regions . A more likely explanation may be that the lower apparent CENH3 density makes this region a less attractive target for CRM insertion than the “L” region . Centromere 5 dramatically illustrates the centromere's ability to move locally in response to retrotransposon insertions . The left half of the current centromere 5 “L” block appears to have acquired CENH3 only during the very recent past – the density of CRM1 elements around the CentC cluster located at the “L”/”I” border suggests that prior to this the centromere was located between 103 . 1 and 107 . 4 Mb ( boxes A/A' and B/B' in Figure 4 ) . That centromere would have looked very similar to today's centromere 2 , i . e . a central CentC cluster surrounded by CRM1 and CRM2 ( Figure 4B ) . Notably the regression lines of the CRM1B/R1 elements are quite similar for centromeres 2 and 5 “L” ( −2 Mb and −2 . 75 Mb per 1 . 5 million years , respectively ) , indicating that during the B/R1 period of activity , both centromeres 2 and 5 “L” shifted towards the short arm as a result of retrotransposon insertions at the centromere/long arm border . In the case of centromere 5 this has resulted in a gradual increase of the CRM-free region separating the “L” and “R” blocks as illustrated by the increasing distance between A/A' , B/B' and C/C' ( Figure 4 ) . Although the newly formed interstitial region is relatively small ( ∼2 Mb ) , a number of fascinating questions arise from this separation , including whether the spindle binds to the “R” region , how this “pseudodicentric” chromosome is oriented and whether the two CENH3 regions of a single chromatid could bind microtubules from opposite poles , which histone variants are present in the “I” region , why there are so few CRM insertions into “R” , whether “R” would be able to function as the sole centromere of chromosome 5 and what the ultimate fate of “L” and “R” might be . Plant genomes have the ability to purge LTR retrotransposons , and the half-life of rice retrotransposons has been estimated to be less than 6 million years [37] . The vast majority of elements available for this analysis have LTRs with κ<0 . 1 , indicating they inserted in the past 7 . 7 million years . Nevertheless , this evidence shows that CRM element insertion can be followed by non-CRM insertion in the same genomic region , and vice versa . Thus it appears that centromeres , as defined by CENH3 nucleosomes , are fluid , and it is conceivable that CENH3 nucleosomes can move from adjacent sites into previously canonical chromatin . Once this occurs , CRM elements target and invade this newly formed centromere region . However , following extensive insertion of CRM elements that may initially be colonized by canonical nucleosomes , the probability of non-CRM elements inserting increases . The sum total of these interactions is illustrated by the chromosomal views of 2 and 5 ( Figure 2A , Figure 3A ) : older CRM4 elements cluster within 30–40 Mb of the peak marking the present day functional centromere located at 90 Mb in centromere 2 and 105 Mb in centromere 5 . These CRM4 elements may represent vestiges of an ancient centromere that have been pushed out of the centromere by consecutive retroelement insertions such as the ones we have documented for CRM1 elements for the past 4 million years . Alternatively , CRM4 elements may lack the centromere targeting exhibited by their cousins ( CRM1 , 2 , 3 ) and instead preferentially target the pericentromeric heterochromatin . These CRM4 clusters are distinct from those located around 155 Mb of chromosome 2 , which may be the remnant of an ancient centromere that was inactivated during the course of the corn genome consolidation following the allotetraploidization event , or alternatively , represent misassembly of this reference chromosome , which shows a break in rice/sorghum synteny in this region [2] . Due to the high sequence identity between elements of a particular subfamily it is difficult to determine from our pyrosequencing data whether any given recently inserted element is associated with CENH3 . About half ( 7 , 330/14 , 598 ) of all CENH3 reads that had been classified as “CRM” mapped to more than one location with equal bitscores . The 18-fold enrichment of CRM1 elements in the ChIP data indicates that many CRM1 elements are associated with CENH3 , but the overall 3-fold lower enrichment of CRM1 in comparison to CRM2 elements implies that the older CRM1 elements that now lie outside of the functional centromere ( e . g . “I” ) are indeed devoid of CENH3 nucleosomes . This is borne out by a comparison of CRM1 elements that inserted at similar times ( κ = 0 . 026–0 . 035 ) on various regions of chromosome 5: elements that inserted within the “L” or “R” regions contain numerous perfect matches to anti-CENH3 ChIP reads , while those elements that inserted within the “I” region or on the long arm have fewer such matches ( Figure S4 ) . CENH3 loading in Arabidopsis has been shown to occur during the G2 phase of the cell cycle [38] , while canonical nucleosomes are loaded during S phase . Like the centromeres of human and Drosophila [39] , [40] , rice [41] and corn [36] centromeres contain both CENH3 and canonical nucleosomes . CRM elements may be populated initially by canonical nucleosomes following integration . The subsequent replacement of canonical by CENH3 nucleosomes in some CRM elements may be dependent on their location relative to the center of the functional centromere , i . e . be more likely if the element has inserted into a CENH3-rich region . This could be mediated by a CENH3 loading mechanism that targets CENH3-rich regions . In other words , CRM elements appear to be associated with centromeres not because they hold an intrinsic attraction for CENH3 nucleosomes , but because they are more likely to be loaded with these nucleosomes as a result of inserting into active centromeres . The high density of CRM elements in centromeres has been postulated to be conducive to intra-strand recombination between adjacent elements [10] . We suspect that such recombination between adjacent CRM elements inserted into CentC clusters will remove intervening CentC repeats , leading to the significantly reduced CentC content observed in present day B73 centromeres 2 and 5 . It is noteworthy that both of these centromeres still do contain some CentC , which raises the intriguing question of whether centromeres lacking all CentC are viable , and whether a mechanism exists to restore the CentC content of CentC-depleted centromeres . Note that the CentC cluster to the left of centromere 5 block “L” is no longer associated with CENH3 . CentC has successfully weathered sequential CRM invasions since the divergence from the maize/rice common ancestor 50 million years ago – yet CRM2 ( and possibly the new CRM1 recombinants R4 and R5 ) seems to be more tightly associated with CENH3 at the present time . The work described here demonstrates the extraordinary value of the high quality maize genome sequence for the study of plant centromere evolution . The insights gained here could not have been provided by analysis of the smaller “model organism” genomes , rice and Arabidopsis , or by whole genome shotgun sequence that cannot easily be assembled in highly repetitive regions such as centromeres . The large genome of maize , which is more representative of a typical plant genome than those of the other model plants , has accumulated many relatively recent retrotransposon insertions that both shape and document its genome evolution . The fact that the maize genome has been sequenced using a minimum tiling path of all FPC contigs makes this sequence particularly amenable for repeat analysis . In summary , we have developed a generally applicable set of methods to map and analyze centromere regions of any organism . Our approach is dependent on the availability of good genetic maps and mapping populations , identification of centromere-specific markers , a high quality genome sequence with a good physical map , anti-CENH3 ChIP followed by pyrosequencing , and FISH to support physical mapping data . The convergence of these techniques in the economically important , large-genome crop plant corn has enabled us to document the unexpected fluidity of its centromeres .
Maize BAC sequences that were generated as part of the Maize Genome Project [2] and contained CentC/CRM based on BLAST homology to GenBank accessions AY321491 . 1 and AY129008 . 1 were used to develop repeat junction markers by the method of Luce et al . [24] . JunctionViewer software [30] was developed to screen the sequenced BAC reads or sequence contigs for the presence of repeats junctions between centromeric repeats ( CentC , CRMs ) and/or repeats from the TIGR Zea Repeats v3 . 0 database . The precise coordinates of the repeat junctions were determined based on BLAST homology to other Zea mays sequences in the high throughput genomic sequences ( HTGS ) database of GenBank , and primers spanning the junctions were designed manually . The junction markers were tested by PCR for polymorphism between inbreds B73 and Mo17; PCR conditions were optimized when amplification differed in intensity between the two parents . A total of 57 polymorphic markers were obtained by screening 791 repeat junction primers , and thirty-five of these were mapped using the IBM population . Transposon display was carried out as described [25] , [26] with the following modifications . The full-length sequence of CRM2 ( AY129008 ) was obtained from NCBI . Primers were designed to specifically amplify the flanking sequences of CRM2 but not other CRM families . Genomic DNA was digested using BfaI and PCR-amplified by pairing CRM2 primers with an adapter primer that hybridizes to the BfaI site . The primers for primary amplification were CRM2_R1 ( 5′- GAGGTGGTGTATCGGTTGCT ) and BfaI +0 ( 5′- GACGATGAGTCCTGAGTAG ) . For selective amplification the primers were P33-labeled CRM2_R2 ( 5′- CTACAGCCTTCCAAAGACGC ) and BfaI +3 selective bases ( where different bases were added to the Bfa +0 primer ) . The final annealing temperature for selective amplification was 58°C . The PCR products were electrophoresed on 6% polyacrylamide gels , and the bands cut out for re-amplification . The re-amplified bands were either cloned and sequenced or directly sequenced from the PCR products . The genotypes of representative centromere repeat junction or transposon display markers were determined in 94 IBM [27] plants from a B73 x Mo17 cross . A representative centromere marker for each chromosome ( Table 1 ) was mapped against a framework of ∼700 SSR and 700 SNP markers or the IBM population by Mike McMullen . The genetic locations from this data set were used to infer genetic position on the IBM2 2008 Neighbors map ( www . maizegdb . org ) . Complete mapping data are available at www . maizegdb . org . Sequences were mapped to chromosome 5 using an oat-maize addition ( OMA ) line for chromosome 5 . PCR primers were designed on genic , non-genic single-copy or non-genic low-copy sequences , or on infrequently repeated sequences as long as the product was expected to be unique . Gene homologous sequences were identified by WU-BLAST ( http://blast . wustl . edu ) nucleotide homology alignments between BAC sequences and the Rice Annotation Project Release 5 . 0 Oryza genic sequences ( ftp://ftp . plantbiology . msu . edu/pub/data/Eukaryotic_Projects/o_sativa/annotation_dbs/pseudomolecules/version_5 . 0/all . chrs/all . cds ) using JunctionViewer . Repeat homology was identified by NCBI BLAST sequence alignments between target BAC or reference chromosome sequences and the HTGS database . Two PCR reactions were performed for each primer pair – one using DNA from the B73 chromosome 5 oat-maize addition line ( obtained from H . W . Rines , U . of Minnesota ) and another with B73 genomic DNA as template . The annealing temperature for reactions was 60°C . Primer pairs resulting in PCR products with strong single bands were sequenced and those sequences were compared to their expected product sequence to confirm unique amplification . Of the 20 PCR primer pairs , 15 produced unique amplicons that were identical in sequence between B73 and OMA line for chromosome 5 . ChIP was performed as previously described [16] . Approximately 50 g of leaf tissue harvested from seedlings of maize inbred B73 were used in ChIP using the maize anti-CENH3 antibody [15] . We obtained approximately 3 µg of immunoprecipitated DNA for pyrosequencing ( GenBank Sequence Read Archive SRA009397 ) . A small amount of the ChIPed DNA was used for FISH analysis to confirm the enrichment of the ChIPed DNA in the centromeres ( Figure S3 ) . CRM/CentC sequence coverage was identified by competitive WU-BLAST as described [2] . Full-length CRM elements were identified as described by Sharma and Presting [9] . Other types of retrotransposons present in the centromere 2 and 5 regions were identified using the maize retrotransposons and LTRs from the TEnest database ( http://www . public . iastate . edu/~imagefpc/Subpages/te_nest . html ) downloaded on 16 May 2009 , as well as JunctionViewer annotations . Complete elements were identified using the reference chromosomes as a BLAST database and the complete retroelements as the query with a word size of 20 and an e-value of 1e-50 . Locations within the centromere were extracted , extended to the full length of the retrotransposon plus an additional 2000 nucleotides on each side . Elements were grouped by family and aligned with the TEnest query using ClustalW [42] . LTRs and TSD were identified visually . The TEnest LTR database and consensus LTRs of CRM were used to identify fragmented elements ( due to sequence assembly errors , nested retrotransposon insertions or deletions ) that could not be aligned or identified with the full-length retrotransposon . 5′ and 3′ LTRs were identified using the reference chromosomes as a BLAST database and the LTRs as the query with a word size of 20 and an e-value of 1e-50 . Locations within the centromere were extracted , extended to the length of the LTR plus an additional 200 nucleotides on each side . LTRs were grouped into separate files based on subfamily and aligned with ClustalW . LTRs were sorted by location and TSDs were compared . Two LTRs of the same element type , located within 200 kb of each other and containing nearly identical TSD ( i . e . the 5′ TSD of one LTR matching at least four of the five nucleotides of the 3′ TSD of the other LTR ) were considered to belong to the same retrotransposon . Insertion times for these fragmented LTRs were dated based on sequence divergence using the method of San Miguel et al . [1] . Evolutionary distances ( κ = estimated number of nucleotide substitutions per site ) between LTR pairs with TSDs were calculated using the K2P model in MEGA version 4 . 0 [43] . One CRM1 retrotransposon from chromosome 5 ( 109 . 7/109 . 8 Mb ) was dated without verifying TSDs , as this element ( CRM , κ = 0 . 029 ) contains an insertion in its LTR – only 249 nt of its LTRs were used to calculate κ . Fourteen other elements with TSDs ( 4 CRM4 , 2 CRM2 and 8 non-CRM ) that contained insertions or gaps were manually truncated based on their alignment prior to estimating κ . To detect CRM1 and CRM2 subfamilies , primers specific to each subfamily were designed in the 5′ LTR , 5′ UTR-polyprotein ( Plyp1 ) and polyprotein regions ( Plyp2 ) ( Table 1 ) . The sequence diversity of CRM1 elements necessitated the design of multiple LTR ( A , B/R1 , R2 , R3 , R4/R5 ) and polyprotein ( A , B ) primers . CRM1 and CRM2 specific regions were amplified using Zea mays inbred B73 genomic DNA by 40 cycles of polymerase chain reaction ( 94° for 40 sec , 60° for 30 sec , and 72° for 1 min ) and subsequently cloned in StrataClone PCR cloning vector pSC-A-amp-kan ( GenBank accessions GQ345011-GQ345022 ) . CRM1 and CRM2 specific FISH probe cocktails were generated by pooling their respective amplicons in equimolar amounts . Metaphase FISH was performed as described by Kato et al . [29] . Fiber-FISH procedures were performed according to Jackson et al . [44] with some modifications . For three-color detection , the biotin-labeled probe ( CentC ) , dig-labeled probe ( CRM2 ) and DNP-labeled probe ( CRM1 ) were detected with far red , red and green , respectively , using three successive layers of antibodies as follows: Layer 1: rabbit anti-DNP + streptavidin 647 in TNB ( 0 . 1 M Tris-HCl pH 7 . 5 , 0 . 15 M NaCl , 0 . 5% blocking reagent ) ; Layer 2: biotinylated anti-streptavidin + chicken anti-rabbit 488+ mouse anti-dig in TNB; Layer 3: streptavidin 647+ rabbit anti-mouse 568 1∶200 in TNB . All antibody incubations were at 37°C; the first layer was for 1 h and the last two for 45 min each . All antibody washes were for three times of 5 min at RT using TNT ( 0 . 1 M Tris-HCl , 0 . 15 M NaCl , 0 . 05% Tween 20 , pH 7 . 5 ) . A final wash in PBS ( 0 . 14 M NaCl , 8 mM Na2HPO4 , 1 . 8 mM KH2PO4 , 2 . 7 mM KCl , pH 7 . 4 ) was performed , and the slides were drained and mounted in Vectashield without counterstain . The fluorescence signals were detected using a Hamamatsu CCD camera . The images were processed using Meta Imaging Series 7 . 5 software using an Olympus BX51 epifluorescence microscope equipped with FITC-Cy3-Cy5-DAPI four-way filter sets ( Olympus ) . A conversion factor of 3 kb/µm ( derived from [44] , [45] ) was used to approximate the physical DNA distance from the micrographs . | Centromeres tend to be the last regions to be assembled in genome projects , as their mapping is hampered by their characteristically high repeat DNA content and lack of genetic recombination . Using unique markers derived from these repeat-rich regions , we were able to generate and annotate physical maps of two maize centromeres . Functional centromeres are defined not so much by their primary DNA sequence as by the presence of CENH3 , a special histone that replaces canonical histone H3 in centromeric nucleosomes . Little is known about how deposition of CENH3 is regulated , or about the interplay between centromeric repeats and CENH3 . By graphing the density of CENH3 nucleosomes onto the physical map , we delineated the functional centromeres in today's maize genome . We then used the large number of LTR retrotransposon insertions , for which the corn genome is well known , as “archeological evidence” to reconstruct the historic centromere boundaries . This was possible because i ) some retrotransposon families of maize ( CRM ) appear to possess a unique ability to preferentially target centromeres during integration and ii ) insertion times of individual retrotransposons can be calculated . Here we show that the centromere boundaries in maize have changed over time and are heavily influenced by centromeric and non-centromeric repeats . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [
"plant",
"biology/plant",
"genomes",
"and",
"evolution",
"genetics",
"and",
"genomics/comparative",
"genomics",
"evolutionary",
"biology/plant",
"genomes",
"and",
"evolution",
"molecular",
"biology/centromeres",
"evolutionary",
"biology/genomics",
"genetics",
"and",
"genomics... | 2009 | Maize Centromere Structure and Evolution: Sequence Analysis of Centromeres 2 and 5 Reveals Dynamic Loci Shaped Primarily by Retrotransposons |
With few studies conducted to date , very little is known about the epidemiology of rickettsioses in Bhutan . Due to two previous outbreaks and increasing clinical cases , scrub typhus is better recognized than other rickettsial infections and Q fever . A descriptive cross-sectional serosurvey was conducted from January to March 2015 in eight districts of Bhutan . Participants were 864 healthy individuals from an urban ( 30% ) and a rural ( 70% ) sampling unit in each of the eight districts . Serum samples were tested by microimmunofluorescence assay for rickettsial antibodies at the Australian Rickettsial Reference Laboratory . Of the 864 participants , 345 ( 39 . 9% ) were males and the mean age of participants was 41 . 1 ( range 13–98 ) years . An overall seroprevalence of 49% against rickettsioses was detected . Seroprevalence was highest against scrub typhus group ( STG ) ( 22 . 6% ) followed by spotted fever group ( SFG ) rickettsia ( 15 . 7% ) , Q fever ( QF ) ( 6 . 9% ) and typhus group ( TG ) rickettsia ( 3 . 5% ) . Evidence of exposure to multiple agents was also noted; the commonest being dual exposure to STG and SFG at 5% . A person’s likelihood of exposure to STG and SFG rickettsia significantly increased with age and farmers were twice as likely to have evidence of STG exposure as other occupations . Trongsa district appeared to be a hotspot for STG exposure while Punakha district had the lowest STG exposure risk . Zhemgang had the lowest exposure risk to SFG rickettsia compared to other districts . People living at altitudes above 2000 meters were relatively protected from STG infections but this was not observed for SFG , TG or QF exposure . This seroprevalence study highlights the endemicity of STG and SFG rickettsia in Bhutan . The high seroprevalence warrants appropriate public health interventions , such as diagnostic improvements and clinical treatment guidelines . Future studies should focus on vector profiles , geospatial , bio-social and environmental risk assessment and preventive and control strategies .
Rickettsial infections including scrub typhus and Q fever are usually referred to as rickettsiosis [1] . Rickettsioses are zoonotic infections transmitted to humans through bites of infected ticks , fleas , lice and mites or through aerosols generated during exposure to infected placentas and birth fluids of mammals in the case of QF [2] . The family Rickettsiaceae includes two genera , Rickettsia and Orientia , which include many human pathogens some of which cause lethal infections with up to 30% mortality without treatment [3 , 4] . The genus Rickettsia has more than 20 species making up several groups among which the spotted fever group ( SFG ) and typhus group ( TG ) are established human pathogens [4 , 5] . The SFG rickettsia includes the etiologic agents of Rocky Mountain spotted fever ( R . rickettsii ) and Mediterranean spotted fever ( R . conorii ) and many others . The TG rickettsia include agents of epidemic ( R . prowazekii ) and endemic ( R . typhi ) typhus [4] . Orientia has two species; O . tsutsugamushi and O . chuto [6] , together forming the scrub typhus group ( STG ) . Coxiella burnetii is the causal agent of Q fever ( QF ) . Of all the methods to detect rickettsial infections , antibody detection by serology is the most commonly used , microimmunofluorescence assay ( IFA ) being the currently accepted gold standard [7] . After an infection , IgM can be detectable for months and IgG for years [7 , 8] . SFG and TG rickettsia occur worldwide and are a significant cause of morbidity in south-east Asia [9] . STG was originally thought to be confined to the Asia-Pacific region but now has been reported from the Middle East [6] , Africa [10 , 11] and South America [12] . Q fever has a worldwide distribution [13] except New Zealand [14] although fears of its introduction have been raised [15] . Rickettsioses are both emerging and re-emerging infections [16 , 17] . Despite being endemic in Asia and causing significant burden to public health , true prevalence studies of these infections are limited . In India , rickettsial diseases including scrub typhus have been documented in several states from all parts of the country [1] . A seroepidemiology study in northeast India , in areas bordering Bhutan reported a sero-prevalence of 30 . 8% , 13 . 8% and 4 . 2% against STG , SFG and TG respectively [18] . In Darjeeling , another Indian district near Bhutan , a 2005 study reported an overall incidence of STG at 34 cases/100 , 000 population/pa , varying from 2 cases/100 , 000 population in July to 20/100 , 000 population in September and decreasing to zero in December [19] . Q fever has been under-reported from India and recent data are lacking [20] . A Chinse study reports an overall Q fever prevalence of 10% and highlights it as an under-reported and underdiagnosed illness [21] . Although situated in the endemic Asia Pacific region , Bhutan has reported scrub typhus cases only since 2009 [22] and SFG , TG and QF have not been reported to date . Rickettsial diseases ( excluding Q fever ) have been included in the national notifiable diseases since 2010 with increasing reports , mostly scrub typhus , from 118 cases in 2011 to 605 cases in 2015 . Despite the increasing notifications and improving awareness , there are currently no clinical guidelines on management of rickettsial infections in Bhutan , and awareness needs improving . There are no reports of Q fever in Bhutan owing to the lack of diagnostic facility both in the human and animal sector at present . Therefore , a serological investigation was undertaken to determine the seroprevalence of rickettsial infections including QF in Bhutan .
Bhutan is composed of 20 districts and 205 subdistricts with an estimated population of 770 , 000 in 2016 [23] . The Bhutan national census in 2005 reported on 1044 rural villages/chiwogs and 311 urban towns as primary sampling units ( PSUs ) . Population density in different districts vary between 9–64 people/km2 [23] . For this study , the 20 districts were stratified into four regions; eastern ( 5 districts ) , central ( 4 districts ) , western ( 5 districts ) and southern ( 6 districts ) as defined by the Bhutan National Statistical Bureau ( NSB ) [23] for their national surveys . From each region , two districts were selected with a probability proportionate to size ( PPS ) method , selecting eight of twenty districts for the study ( Fig 1 ) . A rural and an urban PSU were selected from each district by the same PPS method . To assess the influence of altitude on exposure , altitude of places were arbitrarily grouped into low ( <1000 meters ) , medium ( 1000–2000 meters ) and high altitude ( >2000 meters ) . This descriptive cross-sectional sero-survey was carried out from January to March 2015 , during the dry winter and early spring season . The sample size was calculated using a multi-stage cluster sampling method . Persons <13 years were excluded due to the possible risk of complications during blood sampling in remote areas . The sample size needed to estimate the number of households to be surveyed with a 95% confidence interval and other assumptions ( 50% prevalence rate , 0 . 05 margin of error , a design effect of 2 and an expected rate of participation of 90% ) was calculated to be 864 . Based on Bhutan’s urban-rural population proportion of 30:70 [23] , 30% of the participants were taken from urban and the remaining from rural settings; therefore , of the 864 households selected , 256 were from urban and 608 from rural settings . Each of the eight selected districts contributed 108 households ( 76 rural and 32 urban households ) . The households were taken from the household list with unique identification numbers developed during the previous national surveys ( National Health Survey 2012 and NCD STEPS Survey 2014 ) . When a selected PSU had a lesser number of households than required , a nearby PSU was added . After selection of the household , all eligible members ( ≥13 years ) present in the house were listed and one member was selected for the study through a lottery system . After selection , written consent was obtained; demographic details and environmental exposure history were taken by trained laboratory personnel through a face-to-face interview and blood samples were collected . Serum was extracted and stored at -70°C until shipment to Australia . Serum samples were shipped at room temperature to the Australian Rickettsial Reference Laboratory ( ARRL ) [24] , a nationally accredited laboratory for rickettsial testing , where serological testing was carried out by indirect microimmunofluorescence assay ( IFA ) [25] . Antibodies against SFG rickettsia were individually tested using R . australis , R . honei , R . conorii , R . africae , R . rickettsii and R . felis antigens; TG rickettsia using R . prowazekii and R . typhi antigens; STG using O . tsutsugamushi ( Gilliam , Karp and Kato strains ) and O . chuto antigens , and QF using C . burnetii phase I and phase II antigens . Samples were initially screened at low dilutions and titrated to end-point ( titre ) when positive . With slight modification from the usual ARRL interpretation criteria [24 , 25] , antibody titres of ≥1:256 for IgG and/or ≥1:1024 for IgM against any of the SFG , TG and STG antigens were considered positive for the rickettsial group agents . Similarly an antibody titre of ≥1:50 for IgG or IgA and ≥1:100 against IgM against C . burnetii phase I or II or both were considered positive for Q fever . Positive and negative control wells were included in each slide during testing . Data were entered into an Excel spreadsheet and analysed using STATA software version 14 . Chi-squared or Fischer’s exact test was used to explore the association between seropositivity and study variables considering p values of ≤0 . 05 significant . Univariate logistic regression was used to determine crude odds ratio ( COR ) and p values . All variables with p values 0 . 2 or less in the univariate analysis were taken for multivariate logistic regression to determine adjusted OR ( AOR ) and corresponding p values of <0 . 05 considered significant . The study was approved by the Bhutan Research Ethics Board of Health ( REBH ) ( Ref: REBH/Approval/2014/019 ) and the Human Research Ethics Committee ( HREC ) , University of Newcastle , Australia ( Ref: H-2016-0085 ) . All individuals or parent/guardian provided written consent before participation .
A total of 864 participants were enrolled from the eight districts and all selected candidates consented to the study . There were 345 ( 39 . 9% ) males and the mean age of participants was 41 . 1 ( range 13–98 ) years . Most participants belonged to the age group of 26–40 years . Farmers 414 ( 47 . 9% ) were the highest group by occupation ( Table 1 ) . In seropositive participants , most had IgG antibody titres of 1:256 or 1:512 against STG , SFG and TG rickettsia and titres of 1:100 or 1:200 against Coxiella phase II IgG , IgA or IgM . A very small number of high antibody titres of up to 1: 2048 in STG ( ≈3% ) and SFG ( ≈ 0 . 1% ) were seen in participants . Overall , the most prevalent rickettsial infection was STG ( 22 . 6% ) and the least prevalent was TG rickettsia ( 3 . 5% ) . Evidence of past exposure to multiple agents was also seen; the commonest being dual exposure to SFG and STG ( 5% ) ( Fig 2 ) . Seropositivity rates were not significantly different between males and female for all four infectious agents . The prevalence of each infection , especially STG and SFG , appeared to increase with age and farmers exhibited the highest seropositivity rates . Thirty percent ( 256 ) of the participants were from urban areas . Among all participants , 550 ( 63 . 7% ) reported having animal contact and almost half ( 426 ) had pets at home . In addition , 620 ( 71 . 8% ) reported contact with vegetation and forest during their daily activities , 205 ( 24 . 3 ) recollected suffering from febrile illness in the past , 337 ( 40 . 8% ) had past tick bites , 153 ( 18 . 0% ) had an eschar in the past and 202 ( 23 . 6% ) had past flea bites . Many of the demographic and environmental variables showed significant baseline correlation with seropositivity against each infection in Chi-squared or Fisher’s exact test ( Table 2 ) and a few were statistically significant in the logistic regression analysis . The comparative seropositivity in the urban and rural sampling units of the eight districts and the overall national prevalence of all four infections are shown in Fig 3 and the estimated proportion of each infection at different sampling units ( urban and rural ) of the eight districts presented in Fig 4 . Significant epidemiological factors and seropositivity are shown for STG ( Table 3 ) and SFG ( Table 4 ) . No factors showed association with QF or TG rickettsia seropositivity , likely due to the small number of seropositives . The prevalence of STG seropositivity increased with age . The odds of exposure to STG infection was significantly higher in farmers compared to other occupations . Punakha district had the lowest risk of exposure to STG infections while people living in Trongsa district were three times more likely to be infected as those in other districts . Contact with domestic animals more than doubled the odds of exposure to STG . People residing at high altitude had 89% lower odds of being exposed to STG compared to those residing at lower altitude ( AOR 0 . 11 , p 0 . 002 , 95% CI 0 . 03 , 0 . 44 ) , ( Table 3 ) . SFG rickettsial seropositivity prevalence also increased with age . A person over 55 years of age was three times more likely to have been exposed to SFG than the younger age groups . Compared to other districts , Zhemgang district had a significantly lower odds of exposure to SFG rickettsia ( AOR 0 . 34 , p value = 0 . 010 , 95% CI 0 . 15 , 0 . 77 ) . Altitude did not affect the prevalence of SFG ( Table 4 ) .
This study revealed an overall seroprevalence of 48 . 7% against rickettsioses in Bhutan with the highest prevalence to scrub typhus ( 22 . 6% ) followed by SFG rickettsia ( 15 . 7% ) , Q fever ( 6 . 9% ) and TG rickettsia ( 3 . 5% ) . Evidence of past exposure to two or more rickettsial agents was seen in 11 . 1% of the participants depicting probable dual or multiple infections in an endemic setting or possibly cross-reacting antibodies . This is the first seroprevalence study on rickettsioses in Bhutan and may be used as baseline data for subsequent studies in this country although it is recognised that prevalence estimates may vary when measured at different times of the year . The limitations of the findings from the exclusion of children ( <13 years old ) should be borne in mind . This was required to avoid risks of complications during blood sampling especially in remote areas where medical assistance is hard to obtain . Unintended but unavoidable exclusion of potential participants could have also occurred due to a member of the household being away from home during sampling . Information on past fevers , tick bites and eschar might have had drawbacks due to participants failing to comprehend technical terms . In addition , inter-district , urban versus rural as well as high versus low altitudes comparisons was not precise due to the highly variable landscape within and between districts . Unavailability of adequate local data on climatic conditions , environmental and geospatial information at primary sampling unit ( urban and rural ) level made it impossible to explain inter-district differences of exposure to the infections . The findings in this study were similar to a seroprevalence study in north-east India , that reported the highest seroprevalence against STG ( 30 . 8% ) followed by SFG ( 13 . 8% ) and TG ( 4 . 2% ) [18] . The similarity is noteworthy due to the proximity of these areas to Bhutan . Similar occurrences of these infections in neighbouring countries may benefit from coordinated cross-border prevention and control activities . The odds of exposure sequentially increased with increasing age of participants in case of STG and SFG rickettsiae . This mirrors the situation in endemic areas where increasing number of people would be exposed to the infections as they advance through life leading to an accumulation of older seropositive people in the community . In south-east Asia , murine typhus was reported more in urban dwellers , while STG and SFG were more prevalent in rural dwellers [9] . However , in Bhutan , this study did not find any significant differences in any of the four infections between urban and rural residents probably reflecting similar environmental conditions between the two populations . This is supported by finding no significant differences between occupational groups for all infections , with the exception of STG where farmers had higher seropositivity rates compared to other occupations . There were differences between a few districts for STG and SFG infections , highlighting hot spots for these two infections . Trongsa district in central Bhutan appeared to be a hotspot for STG infection and Punakha district exhibited significantly low odds of exposure . STG exposures were significantly low amongst participants in high altitude areas . This may be explained by cold weather at high altitude areas not favouring mite survival . Of all the districts , Zhemgang in south-central Bhutan had the lowest odds of exposure to SFG rickettsia . Unlike STG infections , altitude had no effect on SFG , TG and QF exposure . This could be due to different tick species at different altitudes transmitting different infections . Expanding primary and secondary clusters of STG infections were also reported in China [26] . Such clusters or hotspots would benefit from focused public health interventions especially where resources are limited as in the case of Bhutan . Targeting prevention and control activities in hotspot areas could be effective and cost saving . Antibody titres of 1:256 or 1:512 were the commonest observed antibody levels amongst the participants . A small number of participants with higher antibody titres of 1:1024 or 1:2048 may have been due to recent infections ( symptomatic or asymptomatic ) or due to recurrent subacute exposures stimulating antibody production . Cross-reactions between antibodies within the rickettsial species , especially between SFG and TG rickettsia , are known to occur . Therefore , persons with mixed antibodies may not necessarily have had multiple infections but may be due to cross-reacting antibodies resulting from the one infection . This could also explain some of the observed multi-species exposures . Background antibody levels in endemic situations are known to interfere with serological diagnosis of acute infections especially with rapid point-of-care diagnostics [17] . This is worthy of note in the Bhutanese setting where only rapid point-of-care diagnostics are available currently . There is urgent need to improve diagnostic facilities in Bhutan to provide more specific assays such as the microimmunofluorescence assay and molecular diagnostics , especially in the main centres . Point-of-care diagnostics could still be useful in the smaller districts and remote health centres for ease of use . Rickettsioses have been associated with poor maternal and neonatal outcomes including stillbirth and low birth weights [27 , 28] in endemic situations . The role of rickettsioses in the high maternal and neonatal mortality and morbidities in Bhutan deserves to be studied . Scrub typhus has also been known to involve the central nervous system manifesting as meningoencephalitis [29–31] and STG has been recently reported as a significant cause of encephalitis in northeast India [32] . This is important in the Bhutanese context in light of establishing the causes of acute encephalitis syndromes including meningococcal infection , Japanese encephalitis and other viral meningitis syndromes which are poorly understood at present . Documented deaths in Bhutan due to proven scrub typhus had resulted from meningoencephalitis and gastrointestinal perforation [22] , which are known to be severe complications of scrub typhus . Understanding these occurrences in endemic areas could be helpful in averting preventable deaths from such complications . Rickettsioses are important causes of illnesses in international travellers [33] . In a study on the spectrum of illness amongst ill returned travellers from six GeoSentinel sites , rickettsial infections were significant causes with 17 and 32 patients/1000 cases returning from south central and south-east Asia respectively [34] . Bhutan is an emerging destination for international travellers involving in activities like camping , trekking , cultural and rural home-stays . The high prevalence of rickettsioses could potentially expose travellers to these infections . Therefore travellers should be aware of the risk and become educated on preventive measures . In addition , educating travellers would keep them vigilant for any febrile illnesses during travel or upon returning to their home countries , enabling them to provide a detailed travel history and for their treating doctor to include rickettsial infections in their differential diagnoses . There are limited prevalence studies on rickettsioses in south-east Asia [9] . Studies are even scantier in south and central Asia including Bhutan where most published studies were focused on clinical cases and acute febrile patients . Therefore , a prevalence study of these neglected but re-emerging infections in these endemic areas should be carried out with active regional collaborations and participations . This first seroprevalence study in Bhutan highlighted the endemicity of rickettsioses especially STG and SFG rickettsia . Findings on TG rickettsia and Q fever should be interpreted with caution due to the detection of fewer positive cases . This high rickettsial seroprevalence needs attention from the Bhutan Ministry of Health such as appropriate public health interventions , diagnostic improvement and clear clinical treatment guidelines . Future studies should focus on vector profiles , geospatial , bio-social and environmental risk assessment and preventive and control strategies formulation . | Rickettsial infections including scrub typhus and Q fever are not widely recognised in Bhutan although the country is situated in an endemic Asian region . With two recorded outbreaks , scrub typhus is known to occur but other rickettsial infections and Q fever have not been recorded . In this first seroprevalence study of rickettsial infections , an overall seroprevalence of 49% was detected against rickettsioses amongst 864 participants . Seroprevalence was highest against scrub typhus group ( STG ) ( 22 . 6% ) followed by spotted fever group ( SFG ) rickettsia ( 15 . 7% ) , Q fever ( QF ) ( 6 . 9% ) and typhus group ( TG ) rickettsia ( 3 . 5% ) . Evidence of exposure to multiple agents were also noted; the commonest being dual exposure to STG and SFG at 5% . A person’s likelihood of exposure to STG and SFG significantly increased with age and farmers were twice as likely to have evidence of STG exposure as other occupations . Trongsa district in central Bhutan appeared to be a hotspot for STG exposure . People living at altitudes above 2000 meters were relatively protected from STG infections but this was not observed for SFG , TG and QF exposure . The findings from this study may direct future research on rickettsioses in Bhutan . These neglected tropical diseases warrant specific public health interventions in Bhutan . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussions"
] | [
"bhutan",
"medicine",
"and",
"health",
"sciences",
"typhus",
"immune",
"physiology",
"pathology",
"and",
"laboratory",
"medicine",
"pathogens",
"immunology",
"geographical",
"locations",
"microbiology",
"rickettsia",
"bacterial",
"diseases",
"age",
"groups",
"antibodies",... | 2017 | Seroprevalence of rickettsial infections and Q fever in Bhutan |
Primary infection of Toxoplasma gondii during pregnancy can be transmitted to the unborn child and may have serious consequences , including retinochoroiditis , hydrocephaly , cerebral calcifications , encephalitis , splenomegaly , hearing loss , blindness , and death . Austria , a country with moderate seroprevalence , instituted mandatory prenatal screening for toxoplasma infection to minimize the effects of congenital transmission . This work compares the societal costs of congenital toxoplasmosis under the Austrian national prenatal screening program with the societal costs that would have occurred in a No-Screening scenario . We retrospectively investigated data from the Austrian Toxoplasmosis Register for birth cohorts from 1992 to 2008 , including pediatric long-term follow-up until May 2013 . We constructed a decision-analytic model to compare lifetime societal costs of prenatal screening with lifetime societal costs estimated in a No-Screening scenario . We included costs of treatment , lifetime care , accommodation of injuries , loss of life , and lost earnings that would have occurred in a No-Screening scenario and compared them with the actual costs of screening , treatment , lifetime care , accommodation , loss of life , and lost earnings . We replicated that analysis excluding loss of life and lost earnings to estimate the budgetary impact alone . Our model calculated total lifetime costs of €103 per birth under prenatal screening as carried out in Austria , saving €323 per birth compared with No-Screening . Without screening and treatment , lifetime societal costs for all affected children would have been €35 million per year; the implementation costs of the Austrian program are less than €2 million per year . Calculating only the budgetary impact , the national program was still cost-saving by more than €15 million per year and saved €258 million in 17 years . Cost savings under a national program of prenatal screening for toxoplasma infection and treatment are outstanding . Our results are of relevance for health care providers by supplying economic data based on a unique national dataset including long-term follow-up of affected infants .
Women with primary infection with T . gondii during pregnancy may exhibit no symptoms , but there is about a 50% risk of transmission to the fetus and the possibility of mild to profound injury to the unborn child in untreated women [1] . The risk of maternofetal transmission increases over the course of the pregnancy , from very low risk in the first trimester to nearly 100% in the final weeks of pregnancy . In the event of transmission , risk of injury to the fetus varies inversely with gestational age , with the risk of profound injury greatest in the first trimester and the possibility of mild disease or no recognized symptoms in later stages of gestation [1 , 6 , 13 , 14] . Consequences of CT can include retinochoroiditis , hydrocephaly , cerebral calcifications , splenomegaly , hearing loss , blindness , and death [1 , 6 , 15 , 16] . In countries with prenatal screening for primary infections and consequent pre- and postnatal treatment , rates of CT and severity of symptoms in infants are lower than in countries without screening programs or compared to historical data before screening was initiated [7 , 10 , 17 , 18] . In comparison , a recent study of children in the United States with CT who had no pre- or postnatal treatment found that 91% of the children referred had visual and/or mental impairment by age 12 [9] . The risk of CT is complicated , however , by the diversity of genotypes of T . gondii . Type II predominates in Europe and was thought to be the predominant genotype in North America [6 , 19–21] . Recent research has identified greater diversity in US wild and domestic animals than was previously thought [22–24] . Types I and III and atypical genotypes are more common in Central and South America [25–27] . These latter strains are more virulent and are associated with ocular disease even when acquired postnatally by immune-competent persons [28] . South American genotypes are also associated with more serious injuries in CT [19 , 20 , 28–30] . Prevalence of infection with T . gondii varies considerably in Europe , from 7% in Norway [31] , 10% in the United Kingdom [32] , 19% in Italy [33] , 32% in Spain [34] , 33% in Austria [31 , 35 , 36] , and 34% in Slovenia [37] , to 37 to 44% in France [7 , 38] ( all reported since 2000 ) . Over the past 20 years , prevalence has fallen rather dramatically in most of the high prevalence countries coincident with national education campaigns , which have perhaps led to changes in food preparation [7 , 31] . Systematic screening of pregnant women also plays an educational role in highlighting the importance of food safety and hygiene for the health of the unborn . Countries with high prevalence in the past similarly had high rates of primary infection in women during pregnancy . This may seem paradoxical since the higher the prevalence among women of child-bearing age , the higher will be the proportion of women entering pregnancy who are immune . Since prevalence , however , increases with age , the majority of young women are not immune and continue to be at risk , presumably with the same food preparation habits as before . The substantial drop in prevalence from the 1990s to the present was accompanied by a substantial drop in maternal incidence after an initial rise [7 , 17] . Austria in 1974 , France in 1992 , and Slovenia in 1995 initiated mandatory prenatal screening programs aimed at reducing maternofetal transmission as well as the severity of injury from CT . Numerous studies have reported that systematic prenatal screening and treatment were coincident with substantial reductions in maternofetal transmission and sequelae of CT [7 , 10 , 12 , 13 , 17 , 18 , 36 , 37 , 39–45] . No systematic economic evaluation of those programs , however , has been published . The countries with systematic prenatal screening and treatment programs face the paradox of successful prevention . Now there are so few children with serious , disabling symptoms of CT that it can appear that the risk of maternal infection does not warrant the expenditure for universal prenatal screening programs . Health budgets are under continual scrutiny . In some countries political currents have changed and the assumption of state responsibility for health is questioned . Moreover , there are diverse stakeholders in the decision to allocate funds to prenatal screening or to other national health needs: the Ministry of Health , insurance funds , the Ministry of Education , social security administrations , and families of affected children . The purpose of the current work is to compare the societal costs of CT under the Austrian national program of prenatal screening for primary toxoplasmosis with the societal costs that would have occurred in the absence of the screening program . In 1961 , Thalhammer revealed a rate of CT of 78 per 10 , 000 live births for the Austrian population [46] . In response , mandatory prenatal screening for toxoplasma infection for all pregnant women was implemented in 1974 under the auspices of the national health care system [46 , 47] . This prenatal screening is part of a national prevention program called “Mother-Child-Booklet-Program” for all pregnant women and their infants through early childhood . The costs are covered by the government and the local health insurance funds; the program is free of charge for families . The Austrian national program is described in detail in previous works [12 , 31 , 48] . Serological prenatal screening is performed ideally on a bimonthly schedule , at 8 , 16 , 24 , and 32 weeks of gestation as well as a maternal or neonatal test for women seronegative up to the time of birth and women who have not been tested during pregnancy . In women with proven seropositivity before the current pregnancy , no further toxoplasma testing is necessary . Women who are tested and found to have been seropositive before conception require only one test . Those with suspected primary infection during pregnancy are tested twice . In Austria during this screening program , the local laboratories used 9 different test methods for the detection of IgM Toxo antibodies , each performed according to manufacturer recommendations . In the case of primary infection in a pregnant woman or to clarify suspicious test results , blood samples were retested in the reference laboratory . The Toxoplasmosis Laboratory at the Medical University of Vienna routinely uses the in-house Sabin Feldman dye test , immunosorbent agglutination assay ( ISAGA ) -IgM ( bioMérieux , France ) , VIDAS Toxo IgG Avidity ( bioMérieux , Frankreich ) , and PCR diagnostics for the detection of toxoplasma infections in pregnant women and their children . In women with primary infection , amniocentesis and polymerase chain reaction of the amniotic fluid is recommended , but costs for those additional tests are not covered by the program . A positive result from amniocentesis identifies an affected fetus prenatally and influences the treatment during pregnancy . The routine PCR analysis used for the B1 gene after amniocentesis showed a sensitivity and specificity of 87 . 2% and 99 . 7% . Furthermore , the results revealed a positive predictive value and negative predictive value of 94 . 4% and 99 . 3% [48] . More recently , using the 529-bp PCR protocol improved sensitivity up to 100 . 0% [49] . Pregnant women are treated after the diagnosis of primary infection until birth , and infants with proven or suspected congenital infection are treated during the first year of life . In cases of CT , additional investigation , including cranial ultrasound , funduscopy , and complete blood count , is part of the program . The screening program reached 93% of pregnant women over the period covered by this analysis , although the ideal schedule was not achieved for most women [31] . The Austrian Toxoplasmosis Register records the serology history and birth outcomes for 1 , 387 , 680 pregnant women from 1992 to 2008 [12] . All cases of CT are recorded in the Register and thus it provides the basis for evaluating the costs of the program and pediatric outcomes over the 17-year period . In 10% of women no toxoplasma testing was necessary due to proven seropositivity before pregnancy . Screening confirmed additional infected women , resulting in seroprevalence of 34 . 4% used in the model [31] . The Register reported 70 women with primary infection of T . gondii and 8 cases of CT per year . The management of women and infants was stable , as was the rate of toxoplasma infection , during the observation period . Pediatric long-term follow-up revealed that 81% of infants with T . gondii infection did not show any clinical signs as of May 2013 . All clinical variables for infection , transmission , and outcomes in infants are shown in Table 1 .
The maternal screening study was approved by the local ethics committee at the Medical University of Vienna , Vienna , Austria ( 824/2009 ) . All adult subjects and parents of any child participants gave their informed consent orally in person or by telephone at the time of inclusion . The individuals were included in the nationwide toxoplasmosis register performed 1992‒2008 and their oral consent was documented in the register data file . Written consent could not be obtained , due to the fact that this was a nationwide study . The data were processed anonymously . The current economic study utilized anonymous data from the national screening program .
The TreeAge program calculates all of the costs that occur in each scenario—the counterfactual ( No screening ) compared to all actual lifetime costs in Austria resulting under the screening scenario . Thus the TreeAge program attributes costs to the Screening scenario that result from treating infants who are infected despite the program , including those whose mothers were not screened or were screened inadequately , with the lifetime costs of follow-up , accommodation , and parental work time lost . In Austria , if there were no screening program , one must assume that the state would provide health care for a child born with , or who later develops , CT symptoms . So the costs of diagnosing and caring for a symptomatic infected child are not really costs of the screening program itself . They would occur ( and in much larger numbers ) without the national screening . The €8 . 4 million a year under the Screening scenario represents the costs of the screening program plus the lifetime societal cost for the affected children born during the 17-year period . The screening program itself entails very little cost . It includes only testing all pregnant women ( except those already known to be seropositive ) and treating women with primary infection . It also would include the cost of treating the very few asymptomatic infected infants because without screening they would be missed , but with screening , they would be treated from birth . Under the screening program , there have been 70 incident infections in mothers per year . Without treatment , there would be a fetal infection rate of 0 . 508 [12] and a probability of asymptomatic CT of 0 . 06 [1] . Thus , there would be two asymptomatic infected newborns treated per year because of the screening program who would have been overlooked without screening ( 70 x 0 . 508 x 0 . 06 = 2 . 10 ) . Costs for each of those children would be: 5 infant IgG test , 5 infant IgM test , pediatric treatment , CBC , ECG , cranial ultrasound , and 17 funduscopies , which amount to €1 , 372 . The costs of the screening program , shown in Table 4 , total approximately €1 . 9 million per year for all pregnancies , or €25 per pregnancy . A new diagnostic appears likely with a test cost of about €4 . Recalculating with a test cost of €4 would reduce the total cost of prenatal screening and maternal treatment from about €1 . 9 to about €1 . 2 million ( calculation not shown ) . The costs of the screening program can be compared to the cost of caring for a child whose mother is not treated . The costs for individual services and productivity losses are listed in Table 2 , but each symptomatic child generates multiple kinds of costs . In the tree before rollback ( calculation ) , Fig 1 , all the costs for an individual child for each outcome are listed at the terminal node . For example , in the No-Screening scenario , a child with severe visual , cognitive , and hearing impairment ( Terminal node #14 in Fig 1 ) will incur the following costs ( assuming symptoms at birth that lead to testing , treatment , and follow-up care ) : 5 infant IgG tests , 5 infant IgM tests , pediatric treatment , CBC , ECG , cranial ultrasound , and 17 funduscopies , as well as the direct costs and productivity losses for child and parents associated with severe visual , cognitive , and hearing impairment , and special education costs . Fig 2 ( Terminal node #14 ) shows the sum of those costs . The lifetime cost for one child with severe visual , cognitive , and hearing impairment is €1 . 8 million ( €1 , 778 , 210 ) . Thus the costs of the entire screening program for one year are nearly the same as the potential costs for a single severely affected child whose mother was not treated . A child with only severe visual impairment generates costs of €482 , 811 ( at terminal node #9 ) . The costs for four such children exceed the annual cost of the screening program . Without prenatal treatment , more than 90% of infected children have been found to have some form of serious impairment [1 , 9 , 52 , 53] . Prenatal screening with pre- and postnatal treatment as needed prevents or mitigates most injuries . Austria has 70 primary infections per year [12] . If we assume 50% maternofetal transmission without prenatal treatment , as seen in Austrian women who were not treated [12] , that would be 35 cases of CT each year , rather than the 8 cases per year under the treatment program , with symptoms ranging from mild visual impairment to fetal death . Because the model calculates costs on a population basis , the cost of €426 in the tree is a cost per Austrian birth , which is multiplied by the number of births , resulting in potential costs of €35 million for the 35 children who would be infected under the No-Screening scenario . The screening program costs €1 . 9 million per year while the societal costs of the No-Screening scenario are €35 million per year . It is useful to see these costs in relation to overall Austrian government spending and Gross Domestic Product ( GDP ) . The annual cost of the screening and treatment program is 0 . 007% of total Austrian public spending on health and 0 . 003% of overall Austrian government spending . The annual cost of the program is 0 . 0006% of Austrian GDP ( Derived from data at www . focus-economics . com/country-indicator/austria/gdp-eur-bn; World Development Indicators , www . wdi . worldbank . org ) . Calculating just the impact on the Austrian public budget—that is , omitting the lifetime costs of lost earnings that fall on affected children , their families , and society , and VSL for fetal and infant deaths , we find that the maternal screening program is still cost-saving . As seen in Fig 3 , and summarized in Table 3 , expenditures by government and government-sponsored insurers , based on Austrian experience over the period 1992 to 2008 , cost €33 per birth compared to an estimated €219 per birth if the prenatal screening program had not been implemented in Austria . ( As explained above , this overstates the budgetary cost of the screening program itself because it includes diagnosis and care of children who would be cared for under the Austrian health care system even without a screening program . ) Even from the extremely narrow budgetary perspective , the Austrian national program has more than paid for itself in reducing the costs to the state and state-sponsored institutions of treating and educating children injured by CT by €186 per birth for 1 . 4 million births over the period . That amounts to a total budgetary saving of more than €258 million , or more than €15 million per year . Results of the sensitivity analysis show that the savings both to society and to the government budget are robust to variations in all costs . Varying costs by ±10% had a trivial effect on cost per birth in the No-Screening and Screening scenarios and consequently on the savings that result from screening , for both the full societal cost and for the public budget . Fig 4 shows an Incremental Tornado Analysis from the societal perspective . The x axis shows the difference in costs per birth between the No-Screening and Screening scenarios with an Expected Value ( EV ) of €323 . The horizontal bars show the full variation in the Expected Value ( savings per birth ) resulting from the range of values for each cost parameter . Both Fig 4 and Table 5 demonstrate the trivial impact on the large savings that result from screening . The variation in VSL had the greatest effect on costs , but even then the difference between low and high values for savings was only €56 and the savings from screening never fell below €275 per birth . Fig 5 shows the one-way sensitivity analysis on VSL in the societal model , which again demonstrates that whether one includes only the loss of earnings ( €800 , 000 ) or the upper bound of the OECD estimate for VSL ( €6 . 7 million ) , there is little impact on the savings derived from the screening program , showing the same minimum savings of €275 per birth seen in Fig 4 and Table 5 . Fig 6 shows the Incremental Tornado Analysis for the Budget impact . The Expected Value , that is savings per birth , is €186 . The variation in savings per birth never exceeds €17 and the minimum savings from the screening program for the budget is never less than €178 per birth , as seen also in Table 6 .
Successful screening and treatment programs , such as Austria’s , face two challenges , both of which derive from their success . As with other public health programs , the European prenatal screening programs and education campaigns confront the paradox of success . People do not see or hear about infants affected by CT as they did in the past when infant deaths or profound brain injuries and visual impairment of varying degrees were more common , due to high rates of CT . Prevention programs only seem expensive in the absence of disease . In the face of budget pressure , the absence of infants with injuries of CT can be misunderstood to mean there is no longer a risk . On the contrary , it has taken two decades of successful prenatal screening and treatment to make the risk invisible . Moreover , the success of education programs in reducing prevalence in the population , while it may protect women by making them more aware of the risk of eating undercooked meat and unwashed fruits and vegetables , actually creates a larger population of women still at risk of infection , and particularly so since even the water supply is a source of infection in some regions . The second challenge to the prenatal screening programs comes from the methodological debate over the validity of observational studies versus randomized controlled trials as the evidence base for interventions . Numerous authors have suggested that the question of efficacy of prenatal screening and treatment can only be adequately answered with randomized controlled trials ( RCTs ) [13 , 39 , 66 , 67] . RCTs , however , pose an insurmountable ethical problem in countries where prenatal screening has been associated with significant improvement in outcomes for infants whose mothers were treated prenatally . An RCT requires equipoise , which is lacking in countries with successful screening programs ( Austria , France , and Slovenia , for example ) and in countries with similar epidemiology and access to care . Without equipoise , it is doubtful that one could construct an ethical trial that would require random assignment of some pregnant women to denial of a treatment with demonstrated efficacy [6 , 7] . Blinding could be incompatible with informed consent . It is also unlikely that such trials would have sufficient power because , with informed consent , few parents would be likely to choose not to medicate . The resulting selection bias would also invalidate the results of the trial . This ethical question is not unique to prenatal screening programs for CT . Interventions to reduce smoking , for example , were implemented based on observational data . Any RCT assigning participants to smoking would not have passed ethical review . It has been impossible to construct valid RCTs for treating sexually transmitted diseases to reduce HIV incidence because observational studies and an earlier trial demonstrated that such treatment is beneficial [68] . Similarly , any other effective treatments for cofactor infections cannot ethically be withheld from controls [68 , 69] . Observational and historical data from Austria , France , and Slovenia , and perhaps even comparative data from the United States , have eliminated the equipoise necessary for an ethical RCT of prenatal screening and treatment for primary infection of T . gondii . The European screening programs for CT have had noteworthy success , reducing the number of deaths and profound injuries in affected infants . That success itself in reducing preventable suffering and death commends the programs for continuation . The cost savings for national health care systems and society at large reinforce the argument for continuation . CT is a health problem worldwide and it is not possible to eliminate all sources of infection for pregnant women , nor is a vaccine likely to be developed in the near future . There are , however , successful CT-prevention programs that are reducing clinical effects of CT and saving money for national health administrations and cost to society . Our results understate the benefits of following the Austrian national program because the costs associated with injuries to infants whose mothers were not tested in accordance with the protocol are attributed to the screening scenario [31] . If those mothers had been tested on schedule , the injuries in the infants would most likely have been fewer and less severe , as was the case for the infants tested on schedule . Another source of overstatement of costs of actual Austrian practice is that we show the direct costs of ideal compliance with the protocol in obstetric visits , including the cost for all susceptible women having five tests , whereas , in practice , 97% of women had fewer than three tests . With fewer tests , that also means shorter treatment and lower treatment costs than the ideal . The average time between tests was 14 weeks , rather than the prescribed eight weeks . For two women whose infants were profoundly affected , the time between tests was 19 weeks [31] . If Austrian practice were in full compliance with the protocol , actual direct costs of screening and prenatal treatment would have been slightly higher , but the costs of treatment and accommodation of infants injured by CT and the loss of their productivity and that of their parents would have been substantially lower because fewer infants would have slipped through the screening process . The costs of screening and preventive treatment are negligible compared to the costs of treatment and accommodation for infants whose injuries are not prevented . Net benefits strongly favor screening . As demonstrated by the Austrian national program , prenatal screening and treatment result in substantial cost saving , both from the conventional societal perspective and even from the narrow perspective of budgetary impact . Results in both cases are robust to wide variations in parameter values . Our data show the positive economic value of such a prevention measure . In summary , our findings of this economic analytic-decision model represent an important base for the discussion regarding implementation or continuation of prenatal screening for toxoplasma infection . | Toxoplasma gondii is a widespread parasitic disease . In the event of primary infection during pregnancy , this parasite can be transmitted from mother to unborn child . Clinical presentation of congenital toxoplasmosis varies from asymptomatic to life-threatening risk for the fetus and infant and in later life . Prevention programs and screening strategies of health care providers vary in different countries . Austria has implemented mandatory prenatal screening for toxoplasmosis for four decades . The screening is free of charge for families and costs are covered by national health care providers . Compliance with the national program is good and outcomes for infected pregnant women and their infants since 1992 are well documented . We compared lifetime costs of screening , treatment , and follow-up with costs in a No-Screening scenario in an economic decision-analytic model . Prenatal screening resulted in substantial cost savings due to reduction in congenital toxoplasmosis and consequent injuries in affected children . | [
"Abstract",
"Introduction",
"Method",
"Results",
"Discussion"
] | [
"children",
"cognitive",
"neurology",
"medicine",
"and",
"health",
"sciences",
"maternal",
"health",
"obstetrics",
"and",
"gynecology",
"toxoplasma",
"gondii",
"neuroscience",
"parasitic",
"protozoans",
"parasitic",
"diseases",
"pediatrics",
"age",
"groups",
"cognitive",
... | 2017 | Congenital toxoplasmosis in Austria: Prenatal screening for prevention is cost-saving |
Initial axial patterning of the neural tube into forebrain , midbrain , and hindbrain primordia occurs during gastrulation . After this patterning phase , further diversification within the brain is thought to proceed largely independently in the different primordia . However , mechanisms that maintain the demarcation of brain subdivisions at later stages are poorly understood . In the alar plate of the caudal forebrain there are two principal units , the thalamus and the pretectum , each of which is a developmental compartment . Here we show that proper neuronal differentiation of the thalamus requires Lhx2 and Lhx9 function . In Lhx2/Lhx9-deficient zebrafish embryos the differentiation process is blocked and the dorsally adjacent Wnt positive epithalamus expands into the thalamus . This leads to an upregulation of Wnt signaling in the caudal forebrain . Lack of Lhx2/Lhx9 function as well as increased Wnt signaling alter the expression of the thalamus specific cell adhesion factor pcdh10b and lead subsequently to a striking anterior-posterior disorganization of the caudal forebrain . We therefore suggest that after initial neural tube patterning , neurogenesis within a brain compartment influences the integrity of the neuronal progenitor pool and border formation of a neuromeric compartment .
Segmentation is a fundamental step during vertebrate brain development . It involves patterning of the cranial neural tube into distinct and segregated transverse units aligned serially along the longitudinal axis [1] . The most important prerequisite for segmentation are borders between the successive neuromeres to allow individual regionalization , growth , and acquisition of distinct functional identity . This process may be hindered in an embryonic brain by the fact that it rapidly increases in size and complexity . Molecular mechanisms underlying segmentation have been studied during development of the relatively simple hindbrain region [2] , [3] . Expression patterns of many regulatory genes also suggest a neuromeric organization of the embryonic forebrain [4] , [5] . Recent studies support a segmental forebrain bauplan with three prosomeres ( P1–P3 ) ( reviewed in [1] ) . Based on morphology and gene expression the alar plate of the diencephalon is divided into the prethalamus ( P3 ) , thalamus ( P2 ) , and pretectum ( P1 ) . The epithalamus including epiphysis and habenular nuclei are part of P2 . The border between prethalamus and thalamus is defined by compartment borders with the interposed narrow region known as the zona limitans intrathalamica ( ZLI ) . Extracellular cell adhesion proteins such as Tenascin within the ZLI have been suggested to mediate lineage restriction between the ZLI and the anteriorly adjacent prethalamus and posteriorly adjacent thalamus [6]–[8] . Similarly , the diencephalic-mesencephalic border ( DMB ) , at the posterior limit of the pretectum , has been identified as a compartment boundary where , in addition to Tenascin , an Eph-ephrin dependent mechanism has been suggested to maintain cell segregation [6] , [9] , [10] . Recent fate mapping studies suggest that the border between the thalamus and the pretectum may also be lineage restricted [11] . However , little is known about a possible mechanism leading to cell lineage restriction between these compartments . The embryonic thalamus ( P2 ) becomes subdivided into two molecularly distinct domains: the rostral thalamus ( rTh ) marked by expression of the proneural gene Ascl1 and the caudal thalamus ( cTh ) , which expresses Neurog1 [12]–[14] . In tetrapods , the rTh contributes to the majority of the GABAergic neurons in the thalamus including ventral lateral geniculate ( vLGN ) and intergeniculate leaflet ( IGL ) , whereas the caudal thalamus gives rise to predominately glutamatergic nuclei projecting to the pallium [15]–[17] . LIM homeobox ( Lhx ) genes regulate developmental processes at multiple levels including tissue patterning , cell fate specification , and growth [18] . These selector genes act as highly similar and highly conserved paralogs . They show a restricted expression pattern in the developing caudal forebrain in frog and mouse; Lhx1/Lhx5 mark the rTh and the pretectum , whereas expression of the Apterous group of Lhx2/Lhx9 is confined to the cTh [19]–[22] . In the mouse , Lhx2 function is required for the acquisition of neuronal identity in different regions such as the telencephalon and nasal placode [23] , [24] . In the cortex , Lhx2 is required to limit the adjacent cortical hem , which expresses BMP as well as canonical Wnts . Both signaling pathways orchestrate hippocampal development [25] , [26] . This suggests that Lhx2-mediated neurogenesis is involved in maintaining the integrity of cortex . In the diencephalon , the Lhx2/Lhx9 positive cTh is also enriched in Wnt signaling pathway components in monkeys [27] . Correspondingly , this region is located next to sources of canonical Wnt ligands at the mid-diencephalic organizer ( MDO ) , the signal-generating population in the ZLI , and at the diencephalic roof plate [8] , [28] . Although the arrangement of these two Wnt positive organizers and the Lhx2/Lhx9 expression pattern in the adjacent Wnt receiving tissue is similar to that in the cortex , our knowledge on their function during diencephalon development is still lacking . During early patterning , Wnt signaling was suggested to have an influence on induction of the thalamus [29]–[31] , but the function of Wnts during regionalization remains unclear . After initial anterior-posterior patterning of the neural tube during gastrulation , it is believed that brain segments develop largely independently . Here we show that Lhx2 and Lhx9 are redundantly required to drive neurogenesis in the zebrafish thalamus . Furthermore , we show that neuronal differentiation mediated by Lhx2/Lhx9 has an impact on maintenance of the thalamus boundaries . Lhx2/Lhx9 restrict the expression of the cell adhesion factor Pcdh10b to the thalamus and therefore sustain the thalamus as a true developmental compartment . Thus , Lhx2/Lhx9 is required for proper development of the thalamus , the core relay station in the brain , and for the integrity of the entire caudal forebrain .
To explore neuronal differentiation in the thalamus , we examined the expression dynamics of lhx2 and lhx9 at early stages of caudal forebrain development ( Figures 1 and S1 ) . We detect expression of lhx9 in the diencephalon first at 30 hpf ( primordial stage 15; Figure 1a , asterisk ) , while at 42 hpf ( high-pec stage ) , the lhx9 expression domain broadens and an overlapping domain of lhx2 expression becomes apparent ( Figure 1b ) . At 48 hpf ( long-pec stage ) , lhx2 and lhx9 are co-expressed in the thalamus ( Figure 1c , asterisk ) . This expression is maintained at later stages ( Figure S1 ) . A cross-section validates the overlap of Lhx2 and Lhx9 positive cells , predominantly laterally in thalamic neuroepithilium ( Figure 1c′ ) . At 48 hpf , lhx9 expression is in proximity to , but with a distinct separation from , the Shh-positive MDO and basal plate ( Figure 1d , d′ ) . In order to determine the fate of cells in this shh and lhx9 negative domain , we cloned the zebrafish homolog of the hey-like transcription factor ( helt ) . Helt has been described as a specific marker of the prospective GABA interneurons of the rostral thalamus ( rTh ) , pretectum , and midbrain [34] , [35] and is required for the formation of these interneurons in the mouse mesencephalon [36] . The expression domain of helt abuts the rostral , ventral , and caudal extent of the lhx9 expression domain ( Figure 1e , e′ ) . Complementary to the helt expression , we find an overlap with glutamatergic neurons marked by vglut2 . 2 at 3 dpf ( Figure S1 ) . This suggests that lhx9 marks the caudal thalamus ( cTh ) and is absent in the GABAergic rTh and pretectum in zebrafish . The ßHLH factor neurogenin1 is strongly expressed in an intermediate layer of the neuroepithelium of the cTh , most likely the subventricular zone ( Figure 1f , f′ ) . Expression of neurog1 abuts the expression of lhx9 in the cTh . The medial part of the lhx9 expression domain overlaps with the expression of the differentiation marker id2a ( Figure 1g , g′ ) . The expression domain of the thalamus-specific post-mitotic neuronal marker lef1 [16] , [37] overlaps entirely with lhx9 ( Figure 1h , h′ ) . The dorsal limit of the Lhx9 domain is adjacent to that of Wnt3a , a marker of the central epithalamus ( Figure 1i , i′ ) . Nevertheless , the lhx9 expression domain overlaps with the expression of the Wnt target axin2 in the diencephalic alar plate ( Figure S1 ) , suggesting that Wnt expression at the epithalamus/MDO might be required to activate the Wnt signaling cascade in the thalamic territory . Thus , we can define Lhx2/Lhx9 as a marker for post-mitotic neurons of the thalamic mantle zone in zebrafish at 48 hpf . At 48 hpf key markers for neurogenesis in the zebrafish brain are expressed in a pattern representing best comparability with amniote brains [38] . Therefore , we chose this stage for the following analyses . To address the function of Lhx2 and Lhx9 in the developing caudal thalamus , we used an antisense Morpholino-based knock-down strategy ( Figure S2 ) . Neither lhx2−/− zebrafish mutant embryos ( beltv24 ) [39] ( n = 13 ) nor single morphant embryos for either lhx2 or lhx9 ( n = 29 ) are visibly distinguishable from uninjected wild type embryos ( Figure S2 ) similar to the situation in the Lhx2 knock-out mouse . However , lhx2/lhx9 double morphant embryos showed significant disruption of thalamic structure ( Figure 2 ) . This is consistent with their overlapping expression domains in the diencephalon ( Figure 1 ) and suggests a functional redundancy within the Apterous group during caudal thalamus development . Therefore , we focused on an approach to reduce both Lhx2 and Lhx9 messages simultaneously by generating double morphant embryos . In addition , we analyzed the lhx9 knock-down morphant in the zebrafish lhx2 mutant background . To define the step in thalamic neuronal differentiation that is dependent on Lhx2/Lhx9 function , we analyzed the expression of the following set of thalamus-specific markers: the neurogenic marker deltaA [40] , the ßHLH factor neurog1 , marking early thalamic progenitors [41] , a regulator of neuronal differentiation id2a , and a marker for mature thalamic neurons lef1 [16] , [42] , the caudal thalamus-specific homeobox gene gbx2 [43] , [44] , and the pan-neuronal marker elav-like 3 ( formerly Hu antigen C ) [45] . These markers can be allocated to three layers in a neuroepithelium in zebrafish: the ventricular proliferation zone ( VZ ) is positive for deltaA , the intermediate or subventricular zone ( SVZ ) zone is marked by neurog1 , and the post-mitotic mantle zone ( MZ ) by elavl3 [38] . At 48 hpf , we observe a lateral expansion of the deltaA positive ventricular zone in lhx2/lhx9 morphant embryos ( 36/54; Figure 2a–b′ ) . Likewise , the expression of the proneural factor neurog1 ( n = 18 ) in the subventricular zone expands laterally ( Figure 2c–d′ ) . Consequently , the expression of the post-mitotic thalamic neuronal markers id2a ( 19/31 ) and lef1 ( 13/20 ) is significantly reduced ( Figure 2e–h ) . Interestingly , the Shh-dependent homeobox transcription factor gbx2 ( n = 25 ) as well as the Wnt mediator tcf7l2 show no alteration in compound morphant embryos ( Figures 2i , j′ , S3 ) . The pan-neuronal marker elavl3 is decreased in the mantle zone ( 3/5; Figure 2k , l ) . This suggests that DeltaA and Neurog1 positive thalamic progenitors need Lhx2/Lhx9 function to proceed with neuronal differentiation ( Figure 2m , n ) . To validate our knock-down strategy and to restrict our analysis temporally and spatially to the thalamus after 24 hpf , we adapted the electroporation technique to the zebrafish system . We were thereby able to deliver DNA unilaterally into the neural tube by pulsed electric stimulation at 24 hpf ( Figure 3a ) and analyze the thalamus at 48 hpf ( Figure 3b ) . Electroporation of EGFP DNA leads to neither molecular nor morphological alteration of the forebrain/midbrain area ( Figure 3c , d; n = 15 ) . Based on previous experiments , we asked if Lhx2 function is sufficient for the induction of post-mitotic thalamic neurons in the Lhx2/Lhx9-double-deficient embryos . Therefore , we re-introduced Lhx2 function unilaterally in the thalamus of Lhx2/lhx9 morphant embryos at 24 hpf corresponding to the endogenous onset of Lhx2 expression ( Figure 1 ) . At 48 hpf , the loss of id2a ( 7/19 ) , lef1 ( 3/15 ) , and Elavl3:GFP ( 8/15 ) expression within the thalamus of lhx2/lhx9 morphant embryos was restored in the electroporated hemisphere at 48 hpf ( Figure 3f , h , j ) . It seems that the laterally expanded epithalamus of morphant embryos can be restored in the electroporated hemisphere ( arrowheads ) . Therefore , we conclude that Lhx2/Lhx9 function is crucial for neurogenesis in the caudal thalamus . Furthermore , Lhx2 alone can compensate for the loss of Lhx2 and Lhx9 , suggesting a redundant function between these paralogs during thalamic neurogenesis . Finally , local electroporation is a valid tool to validate the specificity of a knock-down approach in zebrafish . In the next set of experiments we analyzed the consequence of Lhx2/Lhx9 deficiency on adjacent tissues: the mid-diencephalic organizer ( MDO ) and the embryonic epithalamus ( ETh ) . We find that in morphant embryos the expression domain of lmx1b . 1 , a marker for the MDO and the Eth , expands ventro-posteriorly into the thalamus at 36 hpf ( 31/36; Figure 4a–b′ ) . Similarly , the expression domains of wnt3a ( 89/141 ) and wnt1 ( 8/11 ) also expand ( Figures 4c–d′ , S3 ) . A cross-section reveals that the wnt3a expression is induced ectopically lateral to the habenula , presumably in the thalamic territory ( Figure 4d′ , arrow ) although the forming habenula remains wnt3a negative [46] . To test whether the expanded Wnt expression affects thalamic development , we first monitored Wnt activity in the diencephalon . Here , we analyzed the expression pattern of the pan-canonical Wnt target gene axin2 at 24 hpf , 48 hpf , and at 72 hpf . As expected , we were not able to detect expansion of axin2 expression prior to onset of Lhx2/Lhx9 expression in the thalamus ( Figure S3 ) . From 48hpf , axin2 expression is progressively increased in the thalamus of Lhx2/Lhx9-deficient embryos ( 35/53; Figures 4e–f′ , S3 ) . We confirmed these results using a Wnt reporter zebrafish line 7×TCFsiam:GFP , which expresses GFP under the control of seven repetitive TCF-responsive elements driving a minimal promoter . The GFP expression is detectable around known canonical Wnt sources in the diencephalon—that is , the MDO/ETh area ( Figure 4g , h ) . Lhx2/Lhx9 morphant embryos show expanded GFP expression in the thalamus ( 23/35; Figure 4g′ , h′ ) . In summary , we find that the knock-down of Lhx2/Lhx9 in zebrafish embryos results in an expansion of the epithalamic expression domain of Wnt ligands . This leads to an enhancement of Wnt signaling in the diencephalon , predominantly in the subjacent thalamus . To address the consequences of the loss of Lhx2/Lhx9 and the subsequent upregulation of Wnt signaling on the integrity of the caudal diencephalon , we analyzed the expression pattern of regionally expressed cell adhesion factors in the caudal forebrain . We find that the expression of the cell adhesion molecule , protcadherin10b ( pcdh10b ) , starts in the cTh during late somitogenesis ( Figure S4 ) . At 48 hpf , pcdh10b is predominantly expressed in the progenitor layer , non-overlapping with the post-mitotic lhx2/lhx9 positive neurons ( Figure 5a , a′ ) . The expression domain of pcdh10b abuts dorsally the expression domain of the epithalamus including the wnt3a expression domain ( Figure 5b , b′ ) and posteriorly with the domain of the pretectal marker gsx1 ( Figure 5c , c′ ) . Thus , pcdh10b marks specifically caudal thalamic progenitors at 48 hpf . To investigate the functional interaction between Lhx2/Lhx9 and Pcdh10b , we electroporated lhx2 DNA unilaterally into the caudal diencephalon . Overexpression of Lhx2 proved to be sufficient to inhibit pcdh10b expression in the ventricular zone of the thalamus ( 16/36; Figure 5c , c′ ) . Furthermore , the thalamic expression domain of pcdh10b in lhx2/lhx9-deficient embryos expands into the mantle zone of the cTh ( 17/23 , Figures 5d′ , e′ , S4 ) . This suggests a repressor function of Lhx2 on pcdh10b expression . Interestingly , and beyond a direct repressor effect in situ , pcdh10b also expanded posteriorly into the normally Lhx2/Lhx9 negative pretectum ( Figure 5d , e ) . How do we explain this non-autonomous expansion of pcdh10b following knock-down of Lhx2/Lhx9 ? We wondered whether this could be linked to increased Wnt signaling in the diencephalon of Lhx2/Lhx9-depleted embryos . Therefore , we altered canonical Wnt signaling by treating embryos with small molecule effectors of the Wnt signaling pathway such as the activator , BIO ( a GSK3ß inhibitor ) [47] . To mimic the situation in lhx2/lhx9 morphant embryos , and to avoid gross malformation due to altered patterning during gastrulation , we started ectopic activation of Wnt signaling at 16 hpf and treated the embryos up to 48 hpf . In treated embryos we see ectopic induction of axin2 expression at 48 hpf ( Figure S4 ) , an expansion of pcdh10b expression into the pretectum ( 30/36; Figure 5f , f′ ) similar to the outcome from Lhx2/Lhx9 depletion . In BIO treated embryos , the expression pattern of the principal signal of the MDO , shh , and the patterning marker pax6a are unaltered excluding pleomorphic effects of the treatment ( Figure S4 ) . Following these results , we analyzed the expression of pcdh10b in embryos carrying a mutation in the Wnt pathway inhibitor Axin1 [48] . Although axin1 mutants lack most of the telencephalon and the eyes ( Figure S4 ) , we find an enlarged expression domain of pcdh10b in the cTh at 48 hpf ( Figure 5g , g′ ) . Accordingly , we treated embryos with the Wnt signaling antagonist IWR-1 ( a tankyrase inhibitor , Figure S4 ) [49] from 16 hpf to 48 hpf . Inhibition of Wnt signaling exhibits a decrease of pcdh10b expression ( 55/58; Figure 5h , h′ ) . To validate these results , we used a heatshock inducible transgenic fish line to overexpress the canonical Wnt antagonist Dickkopf1 , Dkk1 ( Figure S4 ) [50] , [51] at 10 hpf . Indeed , we find a similar decrease of pcdh10b expression ( Figure 5i , i′ ) . This effect is seen before , but not after , endogenous pcdh10b induction , suggesting that Wnt signaling is required for induction of pcdh10b but not for its maintenance ( Figure S4 ) . To dissect the regulatory contribution of Lhx2/Lhx9 and Wnt signaling to pcdh10b expression , we reduced Wnt3a function in Lhx2/Lhx9-deficient embryos ( Figure 5j , j′ ) . Interestingly , here we do not find the posterior expansion of the pcdh10b expression domain into the pretectum ( 40/84; Figure 5i ) . However , we still observe the expansion of pcdh10b into the neuronal layer ( 40/84; Figure 5i′ ) . In summary , these data suggest that Wnt signaling , most likely by Wnt3a , induces expression of pcdh10b in the caudal thalamus and Lhx2/Lhx9 are able to limit pcdh10b expression to the progenitor zone ( Figure 5k ) . Furthermore , ectopic upregulation of Wnt signaling is able to induce pcdh10b expression also in the ventricular zone of the pretectum . To study the consequences of altered Pcdh10b levels in the developing caudal forebrain , we analyzed the maintenance of the border zone between thalamus and pretectum in Lhx2/Lhx9 morphant embryos and Pcdh10b-deficient embryos ( Figure 6 and Figure S5 ) . We used five different sequential approaches from the onset of neuronal differentiation at 42 hpf to the formation of a mature thalamus at 4 dpf . Firstly , we analyzed thalamus-specific GFP expression in the Gbx2:GFP transgenic zebrafish line ( Figure 6a–c′ ) [52] . In embryos deficient for Lhx2/Lhx9 , we observe that GFP-positive cells in the ventricular zone of the pretectum become detached from the Gbx2:GFP positive thalamus ( 8/14; Figure 6b , b′ , white arrow ) , suggesting the loss of lineage restriction at the thalamus/prectectum boundary and the spread of thalamic cells into the pretectum . Assuming this to be the case , we next asked if different levels of pcdh10b are required to maintain lineage restriction at this border . Therefore , we interfered with Pcdh10b function by using a Morpholino antisense approach for Pcdh10b [53] . In pcdh10b morphant embryos we find Gbx2:GFP positive cells ectopically in the pretectal progenitor layer ( 18/25; Figure 6c , c′ , white arrows ) . Secondly , we examined the separation of thalamic and pretectal domains by the regional expression of the transcription factors lhx9 and gsx1 ( Figure 6d–f′ ) . Knock-down of Lhx2/Lhx9 ( 11/16 ) or Pcdh10b ( 46/73 ) leads to significant intermingling of lhx9 positive thalamic cells and gsx1 positive pretectal cells ( Figure 6e–f′ , white arrows ) . Thirdly , considering the relay thalamus being mainly glutamatergic whereas the central pretectum remains mainly GABAergic , we looked at the localization of the ßHLH factors Tal1 and Neurog1 . Tal1 marks the inhibitory neurons of the rTh and pretectum , whereas glutamatergic progenitors express Neurog1 [13] . To achieve single-cell resolution , we analyzed the offspring of a Tal1-GFP transgenic line crossed to a Neurog1-RFP transgenic line . We find the specification of ectopic Tal1 positive neurons in the territory of the caudal thalamus in Lhx2/Lhx9 double morphant embryos as well as in Pcdh10b-deficient embryos ( Figure 6h–i′ ) . Fourthly , we analyzed the expression of Gad1 , a marker of inhibitory GABAergic neurons by fluorescent ISH at 3 dpf ( Figure 6j ) . In both Lhx2/Lhx9-deficient embryos ( 4/8 ) and pcdh10b morphant embryos ( 6/10 ) gad1 positive cells are mis-located within the glutamatergic caudal thalamic domain ( Figure 6k–l′; white arrows , Figure S5 ) . Fifthly , we studied the anatomy of the caudal forebrain by analyzing areas of clustered cell nuclei at 4 dpf . In wild type embryos , we observe demarcations between prethalamus and thalamus ( the ZLI ) , between the thalamus and the pretectum , and between the pretectum and the midbrain ( the diencephlic-mesencephlic border; DMB ) ( Figure 6m ) . The observed anatomical compartition correlates with the described genetic profile of these territories ( Figure S5 ) . In lhx2/lhx9 morphant embryos , the demarcation between the thalamus and pretectum is not detectable , although the ZLI and the DMB are unaltered ( Figure 6n ) . In pcdh10b morphant embryos , we are not able to identify the boundary between pretectum and thalamus ( Figure 6o ) , while the ZLI and DMB are still visible . We hypothesize that similar adhesive properties in the thalamus and in the pretectum lead to a loss of separation of these brain parts . Thus , we conclude that a Pcdh10b positive thalamus and a Pcdh10b negative pretectum are required to establish a border between these compartments .
The molecular mechanisms that control the orderly series of developmental steps leading to mature thalamic neurons are poorly understood . Although numerous transcription factors are specifically expressed in the thalamus [14] , only a few have been functionally characterized such as Gbx2 , Neurog2 , and Her6 . Gbx2 knock-out mice show disrupted differentiation of the thalamus by the absence of thalamus-specific post-mitotic neuronal markers Id4 and Lef1 , and subsequently lack cortical innervation by thalamic axons [44] . Although Neurog2-knock-out mice show a similarly severe failure in neuronal connectivity to the cortex , the expression of Lhx2 , Id2 , and Gbx2 is unchanged in these mice , suggesting that in the absence of Neurog2 thalamic neurons are not re-specified at the molecular level [54] . In contrast , Her6 regulates the thalamic neurotransmitter phenotype by repressing neurog1 function and subsequently the glutamatergic lineage . By contrast , Her6 function is a prerequisite for Ascl1a-positive interneuron development in the GABAergic rostral thalamus [13] . Here , we investigate the function of conserved Lhx2 and Lhx9 expression during thalamic development . Lim-HD genes form paralogs such as Lhx1 and Lhx5 , and Lhx2 and Lhx9 [18] . These pairs have been implicated in various aspects of forebrain development . Lhx1/Lhx5 influence Wnt activity by promoting the expression of the Wnt inhibitors sFRPs . This local Lhx-mediated Wnt inhibition is required in the extra embryonic tissue for proper head formation [55] and establishment of the prethalamus [31] . The Apterous group , Lhx2 and Lhx9 , is required for multiple steps during neuronal development . Lhx2 is required in mouse for maintenance of cortical identity and to confine the cortical hem , allowing proper hippocampus formation in the adjacent pallium [26] , [56] . However , Lhx2 function during diencephalic development is still under debate . Although the Apterous genes are already present in the nervous system of the cephalochordate Amphioxus—that is , AmphiLhx2/9 [57]—and co-expression of Lhx2 and Lhx9 has been documented in the diencephalon of vertebrates , such as zebrafish ( here ) , Xenopus [20] , [22] , and mouse [21] , their function in the thalamus has remained unclear . Recent studies of Lhx2 mutant mice showed no alteration during thalamic neuronal regionalization [58] . Furthermore , the function of Lhx9 has not been described , but the expression pattern suggests a role during forebrain development and possibly in parcellation of the thalamus [21] . Here , we show that single knock-down of Lhx2 or Lhx9 has no diencephalic phenotype with the markers analyzed ( Figure S2 ) , comparable to the Lhx2 knock-out mouse , but that simultaneous knock-down of both Lhx2 and Lhx9 leads to stalling of thalamic neurogenesis at the late progenitor stage ( Figure 2 ) . Furthermore , the activation of Lhx2 alone is sufficient to compensate for the loss of both Lhx2 and Lhx9 ( Figure 3 ) . Our results suggest that Lhx2 is functionally redundant to Lhx9 to ensure proper thalamic development . In contrast to other vertebrates , zebrafish embryos show co-expression of Lhx2 and Lhx9 in the telencephalon until 48 hpf ( Figure 1 ) , which could again suggest redundancy [32] . Indeed the pallium is less affected in the lhx2−/− mutant fish compared to loss of the neocortex in Lhx2−/− mutant mice [39] , [59] . Furthermore , in the Lhx9 negative nasal placode , the knock-out of Lhx2 has been shown to lead to a similar neuronal arrest [24] , [60] . In the thalamus , Lhx2/Lhx9 may regulate genes that are essential to complete neuronal development , such that neurons do not reach the terminal neuronal stage . In Lhx2/Lhx9 morphant embryos , we find that the expression of deltaA , neurog1 , as well as pcdh10b is increased . During neuronal development in fish , Neurog1 has been shown to activate delta genes directly by binding several E-box motives in the delta promoter region [40] . This suggests that in Lhx2/Lhx9 morphant embryos , neuronal progenitor development is arrested at the level of deltaA/neurog1 expression . Consistently , terminal thalamic neuronal markers such as Id2a and Lef1 are absent in Lhx2/Lhx9 morphant embryos . Interestingly , both of these markers have been shown to be activated by Wnt signaling [61] , [62] . Although local Wnt activity is upregulated locally in the lhx2/lhx9 morphant embryos , these target genes are not transcribed , suggesting that Lhx2/Lhx9 thalamic neuronal differentiation is coupled to a second competence phase for Wnt signaling . Also , the late and restricted onset of Lhx2/Lhx9 expression in the thalamus and their requirement for Id2a and Lef1 expression may explain the thalamic neuronal specificity of the Wnt target lef1 . Thus , we propose that Lhx2/Lhx9 are essential determinants for cells to reach the late stage of thalamic neuronal development . In the spinal cord , Lim HD factors together with ßHLH factors have been shown to be required for cell cycle exit [63] . The Lim containing factor Isl-1 and Lhx3 together with the ßHLH factors Neurog2 and NeuroM act in a combinatorial manner to directly trigger motor neuron differentiation . In the thalamus , we find a similar process: Lhx2/Lhx9 inhibit the expression of progenitor markers such as pcdh10b and activate the expression of postmitotic differentiation markers such as id2a , lef1 , and elavl3 . Interestingly , proper differentiation of thalamic neurons is required to restrict the MDO and dorsal roof plate ( Figure 7 ) , a finding that reflects the conversion of neocortex in Lhx2 knock-out mice . Here , the Gdf7 positive cortical hem expands at the expense of the neocortex [23] . This supports the hypothesis that proper neuronal differentiation is required to maintain brain compartments and their borders . In the mid-diencephalon , the central source of patterning cues is the MDO . Here , three different signaling pathways merge: Shh , Fgf , and Wnt [64] . Shh signaling has been shown to induce proneural genes such as Ascl1 in the rostral thalamus and Neurog1 in the caudal thalamus ( cTh ) [12] , [13] , [65] and a set of transcription factors assigning specific properties to the developing thalamic cells [14] , [21] , [66]–[68] . Furthermore , Fgf signaling influences the development of the rTh [69] and parts of cTh , the motor learning area [70] . Interestingly , although the mid-diencephalon expresses a set of canonical and non-canonical Wnt ligands and receptors [27] , [28] , the function of Wnt signaling is not clear . Wnt signaling seems to be required for mediating thalamic identity in chick embryonic explants [29] and mutation of the Wnt co-receptor Lrp6 leads to a severe reduction of thalamic tissue in mice [30] . Here , we show that Wnt signaling from the MDO and the roof plate influence compartition of the caudal diencephalon . The canonical Wnt signaling pathway plays a pivotal role in mediating adhesiveness and the key effector of the Wnt pathway , β-catenin , was initially discovered for its role in cell adhesion [71] , [72]: it promotes adhesiveness by binding to the transmembrane , Ca2+-dependent homotypic adhesion molecule cadherin , and links cadherin to the intracellular actin cytoskeleton . Although several classes of molecules are involved in morphogenetic events , cadherins appear to be the major group of adhesion molecules mediating formation of boundaries in the developing CNS [73] . After a phase of ubiquitous expression , cadherins display a very distinct expression pattern in the neural tube [74] . In the developing diencephalon , classical cadherins , such as Chd2 , Chd6b , and Chd7 , mark presumptive nuclear gray matter structures within developmental compartments [75] . Still , these studies so far are not able to explain the different compartition in the caudal forebrain . Here , we describe the expression pattern of the non-clustered protocadherin , pcdh10b , in the developing diencephalon and show that it marks the ventricular zone of the thalamus at mid-somitogenesis ( Figure S5 ) . During somitogenesis , pcdh10b modulates cell adhesion and regulates movement of the paraxial mesoderm and somite segmentation [53] . We find that the border of pcdh10b expression co-localizes with the border between thalamus and pretectum during diencephalic regionalization ( Figure 5 ) . Furthermore , we could link Pcdh10b expression to canonical Wnt signaling . In chick , some hallmarks of lineage restriction for the border between thalamus and pretectum have been observed previously; for example , vimentin and chondroitin sulfate proteglycans are strongly enriched at this border . Similar to the anatomical observation in fish ( Figure 6j–l ) , the chick neural tube shows a morphological ridge where interkinetic movement is disrupted [6] . However , there are conflicting data from direct analyses of cell lineages in the caudal chick forebrain regarding cell compartment borders between thalamus and pretectum [6] , [76] . This may be explained by the different stages of analysis . In other vertebrate models , Pcdh10 expression has been reported only at later stages in development , in chicken HH28 , and in mouse E15 [77] , [78] , arguing against a comparable role in these model organisms . However , Pcdh10 together with Pcdh8 , 12 , 17 , 18 , and 19 belong to a structurally related subfamily , the non-clustered δ2 protocadherins , and several members indeed show an expression pattern during somitogenesis in mouse [79] . Although we have not carried out direct lineage restriction experiments by tracing small cell clones at the border , we suggest that the thalamic area intermingles with the pretectum when both areas express similar levels of this adhesion molecule ( Figure 7 ) . Our data are supported by the fact that pcdh10b knock-down or overexpression also lead to a similar phenotype in somite development [53] . Similarly in Gbx2 knock-out mice , thalamus cells start to intermingle with pretectum cells [11] . Interestingly , these authors observe a non-cell autonomous function for this transcription factor and claim a restriction mechanism mediated by an unknown cell adhesion factor . We suggest that , as for Lhx2/Lhx9 , Gbx2 is required for the acquisition of proper neuronal identity and the lack of Gbx2 may lead to a similar sequence of events—that is , expansion of the Wnt-positive roof plate and alteration in pcdh10b expression . This hypothesis should be tested in the Gbx2 knock-out mouse . Notably , as pcdh10b is also expressed in hindbrain rhombomeres [80] its function should be determined during differentiation in this well-studied segmented part of the neural tube; should compartment formation in the caudal forebrain and hindbrain turn out to involve similar molecular effectors , we may reach a unifying mechanism for compartition of the neuraxis—whether it be in the generation of single units ( thalamus , pretectum ) or iterated modules ( rhombomeres ) . Thus , we suggest that Lhx2/Lhx9 is required for neurogenesis within the thalamus and is important to maintain longitudinal axis patterning of the CNS also at later stages . Alteration of neurogenesis in a brain part affects the development of the neighboring parts and thus leads to loss of the integrity over compartment boundaries .
Breeding zebrafish ( Danio rerio ) were maintained at 28°C on a 14 h light/10 h dark cycle [81] . To prevent pigment formation , embryos were raised in 0 . 2 mM 1-phenyl-2-thiourea ( PTU , Sigma ) after 24 hpf . The data we present in this study were acquired from analysis of wild-type zebrafish of KCL ( KWT ) and of the ITG ( AB2O2 ) as well as the transgenic zebrafish lines; tal1:GFP [82] , hs-dkk1:GFP [51] , elavl3:GFP [83] , GA079:RFP [84] , shh:RFP , neurog1:RFP [41] , gbx2:GFP [52] , and the belladonna zebrafish mutant line with a loss of lhx2 [39] and masterblind mutant line carrying a mutation in axin1 [48] . In bel/lhx2 mutants , a 22 bp deletion in the third exon causes a frame-shift and therefore a stop codon after the second LIM domain . Embryos were staged [85] and ages are listed as hours post fertilization ( hpf ) . Transient knock-down of gene expression was performed as described in [13] . We used the following Morpholino-antisense oligomeres ( MO , Gene Tools ) at a concentration of 0 . 5 mM: lhx2 MO ( 5′-GCT TTT CTC CTA CCG TCT CTG TTT C-3′ ) , lhx9 MO ( 5′-AGG TGT TCT GAC CTG CTG GAG CCG T-3′ ) , wnt3a MO [86] , and pcdh10b MO [53] . The injection of MO oligomers was performed into the yolk cell close to blastomeres at one-cell or two-cell stage . For electroporation , embryos were manually dechorionated and mounted laterally in 1 . 5% low melting-point agarose at 24 hpf . We locally injected 0 . 5 µg/µl GAP43-GFP DNA solution or 1 µg/µl pCS2+lhx2 DNA [32] solution in the III brain ventricle . The positive charged anode was positioned on top of the diencephalon , whereas the negative cathode was positioned underneath the diencephalon ( Figure 3 ) . For electroporation , we used a platinum/iridium wire with a 0 . 102 mm diameter ( WPI Inc . ) . During the electroporation procedure the embryo was kept in 1× Ringer as conductive fluid . We used the stimulator CUY21 ( Nepa Gene Ltd . ) with the following stimulation parameters: 24 V voltage square wave pulse , 4 ms pulse length , 2 ms pulse interval , delivered three times . Settings are based on the published electroporation approaches in [87] . To manipulate Wnt signaling in vivo , we used BIO [47] ( ( 2′Z , 3′E ) -6-Bromo-indirubin-3′-oxime , TOCRIS Bioscience or IWR-1 [49]; SIGMA ) as pharmacological agonist and antagonist of the Wnt signaling pathway . For Wnt signaling analyses , embryos were dechorionated at 16 hpf ( 15–17-somite stage ) and incubated with 4 µM of BIO in 1% DMSO , 40 µM IWR-1 in 0 . 2% DMSO , or with 1% DMSO only . Prior to staining , embryos were fixed in 4% paraformaldehyde/PBS at 4°C overnight for further analysis . Whole-mount mRNA in situ hybridizations ( ISH ) were performed as described in [88] . Antisense probes were generated from RT-PCR products for the following probes with primer pairs ( forward/reverse ) : lhx2b , 5′-AGT GCG TCT CAC GGA AAT CT-3′/5′-GCA TCC ATG ATC GGT CTT CT-3′; lhx9 , 5′-CGT TGG AGA AAG TGG ACT GG-3′/5′-TGG TGA AGA ATT CCG ATC AA-3′; sema3d , 5′-GCT GCA GAA ATC TCC TCG TC-3′/5′-ATT TTG CAC AAG TGG GCA TT-3′; helt , 5′-CCA AAA AGC TCG CCT TTA ATC-3′/5′-AAC ATA TTA AGA CGT ATT TAC AGA GCA-3′; lmx1b . 1 , 5′-GAC AAC AGC CGG GAT AAA AA-3′/5′-CCA TCC GAT TGG ACA TTA CC-3′ . The expression pattern and/or antisene RNA probes have been described for shha ( formerly known as shh; [89] ) , gsx1 [90] , pax6a [91] , gbx2 [92] , axin2 [46] , lef1 [93] , wnt3a [94] , dla [95] , id2a [96] , lmx1b . 1 [97] , pcdh10b [53] , gad1 ( gad67 ) [17] , and vglut2 . 2 [98] . Post-ISH , embryos were re-fixed in 4% paraformaldehyde/PBS at 4°C overnight and transferred to 15% sucrose/PBS and kept for 8 h at 4°C . For embedding , embryos were transferred to a mould filled with 15% sucrose/7 . 5% gelatine/PBS at 42°C for 10 min . The moulds were kept overnight at 4°C , frozen in liquid nitrogen on the following day , and stored at −80°C until required . Frozen blocks were sectioned coronal with 16 µm thickness on the cryostat . To reveal neurons that have initiated axogenesis , we used a monoclonal antibody against acetylated tubulin ( Sigma , T-6793 ) in a concentration of 1∶20 as described in [88] . For visualizing cell nuclei , embryos were fixed in 4% paraformaldehyde/PBS at room temperature for 2 h and transferred in 1× PBS . Fixed brains were hemisected and incubated in 25 µM SYTOX nucleic acid stain ( Invitrogen ) overnight . After washing in 1× PBS brains were mounted laterally for confocal imaging analysis . Prior to imaging , embryos were deyolked , dissected , and mounted in 70% ( v/v ) glycerol/PBS on slides with cover slips . Images were taken on Olympus SZX16 microscope equipped with a DP71 digital camera by using the imaging software Cell A . For confocal analysis , embryos were embedded for live imaging in 1 . 5% low-melting-point agarose ( Sigma-Aldrich ) dissolved in 1× Ringer's solution containing 0 . 016% tricaine at 48 hpf . Confocal image stacks were obtained using the Leica TCS SP5 X confocal laser-scanning microscope . We collected a series of optical planes ( z-stacks ) to reconstruct the imaged area . Rendering the volume in three dimensions provided a view of the image stack at different angles . The step size for the z-stack was usually 1–2 µm and was chosen upon calculation of the theoretical z-resolution of the 40× objective . Images were further processed using Imaris 4 . 1 . 3 ( Bitplane AG ) . | The thalamus is the interface between the body and the brain . It connects sensory organs with higher brain areas and modulates processes such as sleep , alertness , and consciousness . Our knowledge about the embryonic development of this central relay station is still fragmented . Here , we show that the transcription factors Lhx2 and Lhx9 are essential for the development of the relay thalamus . Zebrafish embryos lacking Lhx2/Lhx9 have stalled neurogenesis - neuronal progenitor cells accumulate but do not complete their differentiation into thalamic neurons . In addition , we find that the neighboring Wnt-expressing epithalamus expands into the space containing mis-specified thalamus in these embryos . We identified a thalamus-specific cell adhesion modulator , Pcdh10b , which is controlled by canonical Wnt signaling . Altered Wnt-dependent Pcdh10b function in Lhx2/Lhx9-deficient embryos leads to intermingling of the thalamus and adjacent brain compartments and consequently regionalization within the caudal forebrain is lost . Organization of the developing CNS into molecularly distinct but transient segments and the implications for regional differentiation are well established for the developing hindbrain . We conclude that this applies to caudal forebrain too: Lhx2 and Lhx9 emerge as crucial factors driving neurogenesis and maintaining the regional integrity of the caudal forebrain . These are two prerequisites for the formation of this important relay station in the brain . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"molecular",
"neuroscience",
"neurogenesis",
"gene",
"regulation",
"cadherins",
"neuroscience",
"cell",
"differentiation",
"gene",
"function",
"animal",
"models",
"developmental",
"biology",
"model",
"organisms",
"molecular",
"development",
"molecular",
"genetics",
"signali... | 2011 | Lhx2 and Lhx9 Determine Neuronal Differentiation and Compartition in the Caudal Forebrain by Regulating Wnt Signaling |
Leukotriene B4 ( LTB4 ) is secreted by chemotactic neutrophils , forming a secondary gradient that amplifies the reach of primary chemoattractants . This strategy increases the recruitment range for neutrophils and is important during inflammation . Here , we show that LTB4 and its synthesizing enzymes localize to intracellular multivesicular bodies that , upon stimulation , release their content as exosomes . Purified exosomes can activate resting neutrophils and elicit chemotactic activity in a LTB4 receptor-dependent manner . Inhibition of exosome release leads to loss of directional motility with concomitant loss of LTB4 release . Our findings establish that the exosomal pool of LTB4 acts in an autocrine fashion to sensitize neutrophils towards the primary chemoattractant , and in a paracrine fashion to mediate the recruitment of neighboring neutrophils in trans . We envision that this mechanism is used by other signals to foster communication between cells in harsh extracellular environments .
Chemotaxis , the directed movement of cells in response to external chemical gradients , is essential to a wide array of biological processes ranging from developmental processes , wound healing , angiogenesis , and immune responses and is implicated in pathological conditions such as chronic inflammatory diseases and metastasis [1] . Upon exposure to endpoint primary chemoattractants , cells secrete secondary chemoattractants that serve to maintain the robustness and sensitivity to the primary chemoattractant signals [2] . Once secreted , these secondary chemoattractants form a gradient to recruit cells that are farther away , thereby dramatically increasing the range and persistence of detection [3] . Intercellular communication through the release of secondary chemoattractants may be homotypic , where the primary and secondary chemoattractant are the same , or it may be heterotypic , where the secondary chemoattractant is different from the primary chemoattractant and is released following stimulation by primary attractants . Homotypic intercellular communication is remarkably exhibited in the social amoebae Dictyostelium discoideum , where collective chemotaxis of cells towards cyclic adenosine monophosphate ( cAMP ) is regulated by the release of cAMP and the formation of characteristic chains of cells called streams [4] . Unlike cAMP in Dictyostelium , the release of the secondary chemoattractants CCL3 and CXCL8 by monocytes and dendritic cells in response to the primary chemoattractant Serum Amyloid A represents an example of heterotypic intercellular communication [5] . This ability of chemotaxing cells to transduce autocrine and paracrine chemical signals in response to primary signals has been termed signal relay . Heterotypic signal relay occurs in neutrophils migrating towards primary chemoattractant gradients through the release of leukotriene B4 ( LTB4 ) [6] . Following stimulation , cytoplasmic phospholipase A2α translocates to the nuclear envelope , where it is poised to hydrolyze membrane bound lipids to form arachidonic acid ( AA ) [7] . Simultaneously , 5-lipoxygenase ( 5-LO ) is mobilized to the nuclear envelope where it associates with the 5-LO activating protein ( FLAP ) and acts on AA to generate leukotriene A4 ( LTA4 ) . LTA4 , by means of LTA4 hydrolase ( LTA4H ) , is finally converted into LTB4 [7] , which is then secreted from cells by an unknown mechanism . Initially thought to be released at the site of infection to mediate neutrophil recruitment and proinflammatory processes [8 , 9] , LTB4 was later shown to be a central secondary chemoattractant in vivo [10] . More recently , we showed that LTB4 preferentially mediates neutrophil swarming towards tissue injury sites [11] and established that the secreted LTB4 in physiological gradients of primary chemoattractants acts as an amplifier of neutrophil chemotaxis and mediates signal relay between migrating neutrophils [6] . In order for secondary chemoattractants to act as bona fide signal relay molecules , they must be released in a form that enables the generation of stable gradients during chemotaxis . It has been established that the release and subsequent diffusion of LTB4 creates extremely transient gradients due to the small size of the molecule [12] . The self-diffusion coefficient of a typical formyl chemoattractant is ~ 10−5 cm2/sec , which is a log order of magnitude higher than typical unsaturated fatty acids [13] . This implies extremely shallow LTB4 concentration profiles . Indeed , assessment of the diffusive properties of AA , the structurally similar precursor of LTB4 , shows shallow and transient gradients compared to a primary chemoattractant such as N-formylMethionyl-Leucyl-Phenylalanine ( fMLP ) [12] . As a plausible mechanism to generate stable secondary gradients , one may argue in favor of a carrier-based mechanism for passive LTB4 transport , such as binding to serum albumin , and this may indeed be true for systemic transport [14] . This , however , would not account for short-range gradients in migrating cells owing to protracted and uncontrolled LTB4 release . Moreover , this mechanism would not explain how hydrophobic molecules such as LTB4 are protected from the aqueous environment of the extracellular milieu . The mechanisms by which LTB4 is secreted and how LTB4 gradients are formed or mediate signal relay of primary chemotactic signals have yet to be determined . However , studies on gradient formation of lipid-modified Drosophila morphogens [15] , or the formation of palmityolated-Wnt gradients during Drosophila embryogenesis [16] and cAMP gradient propagation in Dictyostelium [17] , point towards vesicular packaging as an effective way of signal dissemination in the extracellular milieu . In the present study , we investigated whether a similar vesicular packaging of LTB4 is involved in the creation of a stable gradient during neutrophil chemotaxis . To do so , we assessed whether LTB4 is secreted though extracellular vesicles ( EVs ) and if its synthesis and transport involve intracellular vesicular trafficking . Most importantly , we also determined if vesicles released during chemotaxis are indeed capable of mediating the LTB4-dependent signal relay response during neutrophil chemotaxis .
To begin identifying the mechanisms that underlie LTB4 secretion , we measured LTB4 content as well as the distribution of 5-LO in resting and activated neutrophils . We fractionated unstimulated and fMLP-stimulated primary human neutrophils using nitrogen cavitation , differential centrifugation , and separation on iodixanol density gradients ( Fig 1A ) . In resting neutrophils , LTB4 content was primarily low across the different fractions with a small peak observed in fraction 3 ( Fig 1B ) . On the other hand , in fMLP-stimulated neutrophils , a nonuniform asymmetric increase of LTB4 levels was observed , where LTB4 levels were elevated in both low- ( fractions 1–5; density ~1 . 05–1 . 08 g/ml ) and high-density fractions ( fractions 10–12; density ~1 . 17–1 . 19 g/ml ) , but not in intermediate density fractions ( fraction 6–9; ~1 . 09–1 . 11 g/ml ) that contained the cis-Golgi markers GM130 ( Fig 1B and 1C ) . This fMLP-induced asymmetric partitioning of LTB4 across different densities was unlike other canonical secretory proteins such as myeloperoxidase ( MPO—a marker for the azurophilic granules ) and matrix metallopeptidase 9 ( MMP9—a marker for gelatinase granules ) [18] , although total MPO and MMP9 levels appeared to increase poststimulation ( Fig 1C ) . These findings suggest that canonical secretory pathways used by MMP9 and MPO are not involved in LTB4 trafficking . On the other hand , fMLP stimulation did induce the redistribution of the tetraspannin CD63 to higher density fractions . As CD63 is known to traffic between late endosomal and secretory compartments such as multivesicular bodies ( MVBs ) [19] , we reasoned that CD63 redistributes to MVBs upon fMLP addition . This was confirmed by the presence of LAMP1 , a high-density lysosomal marker shown to be present in MVBs of neutrophils [20] , in fractions 9–12 ( Fig 1C ) . Remarkably , 5-LO was also found in CD63- and LAMP1-positive high-density fractions upon fMLP stimulation , and the distribution pattern of LTB4 was similar to that of 5-LO in both resting and activated conditions , suggesting active 5-LO in this cellular compartment . Together , the presence of LTB4 and 5-LO in CD63- and LAMP1-positive fractions suggest the involvement of MVBs in LTB4/5-LO transport . To confirm whether the 5-LO-containing vesicles indeed reside in MVBs , we performed transmission electron microscopy ( EM ) on neutrophils chemotaxing towards fMLP . As previously reported using immunofluorescence [21] , we observed substantial 5-LO labeling on the nuclear envelope ( Fig 2D , 2E and 2G ) compared to antibody control ( Fig 2A ) . Consistent with our fractionation findings , we also detected 5-LO on MVBs—both on the membrane and on intraluminal vesicles ( ILVs ) ( Fig 2B–2E , 2G and 2I ) . Indeed , quantitation of immunogold staining of different vesicular compartments showed that the majority of 5-LO localized to MVBs compared to non-MVB vesicles ( Fig 2F ) , with 30%–35% of all MVBs staining positive for 5-LO . In addition , 5-LO was observed on the plasma membrane ( Fig 2B and 2C ) , which may be due to the transfer of 5-LO-containing outer MVB membranes to the plasma membrane during fusion events ( Fig 2Giii ) . Most interestingly , we found that a substantial number of 5-LO-containing MVBs ( 25%–30% ) were tightly associated with the nuclear envelope ( Fig 2E , 2G and 2Hi ) . These MVB-associated regions of the nuclear envelope also showed specific enrichment for 5-LO . MVBs typically fuse to the plasma membrane and release their ILVs , termed exosomes , into the extracellular milieu [22] . In this context , we readily observed 5-LO-containing ILVs released as exosomes from the trailing edge of neutrophils ( Fig 2H ) and the presence of 5-LO-containing vesicles at the synapse between two migrating neutrophils ( Fig 2I ) . As CD63 is highly enriched in exosomes [23] , we next assessed the cellular distribution of 5-LO and CD63 in live chemotaxing cells . We expressed mCherry-tagged 5-LO ( mCherry-5LO ) and/or GFP-tagged CD63 ( CD63-GFP ) in the pluripotent hematopoietic cell line PLB-985 , which can be differentiated into neutrophil-like cells [24] . We again found that 5-LO localizes to the nucleus in resting mCherry-5LO cells ( S1A Fig and S1 Movie ) . However , upon uniform fMLP addition , we readily observed the redistribution of mCherry-5LO to the cytoplasm in a punctate pattern ( Fig 3A and S2 Movie ) . This vesicular localization was even more apparent in cells chemotaxing in a gradient of fMLP and was found to be more prominent at the trailing edge compared to the leading edge of chemotaxing cells ( Fig 3B and S3 Movie ) . The trailing edge localization of 5-LO-containing vesicles was further substantiated through phalloidin counterstaining , which marks the leading edge F-actin ( S1B Fig ) . To determine the nature of the 5-LO positive vesicles , we visualized the dynamic distribution of CD63-GFP . As expected , we observed a punctate localization of the MVB marker in both resting ( S1A Fig ) and stimulated ( S1C Fig ) cells expressing CD63-GFP . Remarkably , in mCherry-5LO/CD63-GFP coexpressing cells chemotaxing towards fMLP , we observed a clear colocalization of both proteins ( Fig 3C , S4 Movie ) , which again occurred mainly at the trailing edge of the chemotaxing cells . Together , these findings establish that 5-LO is present on MVBs that fuse with the plasma membrane and release their exosomal content during chemotaxis . Having established the presence of 5-LO in MVBs , we set out to determine whether neutrophils secrete exosomes that contain LTB4 and LTB4 synthesizing enzymes . To this end , supernatants from fMLP-stimulated neutrophils were filtered through 0 . 45 μm pore-sized membranes and centrifuged to remove larger vesicles and cell fragments . Pellets recovered from ultracentrifugation of the filtrate contained a heterogeneous mixture of EVs ( unpurified EV ) ranging in size from 50–500 nm , as ascertained from by EM ( Fig 3D ) . Crude EVs were further fractionated on a discontinuous gradient of iodixanol to yield a vesicle population with a density of ~1 . 09 g/ml and vesicle size between 50–120 nm ( median vesicle size = 80 nm; Fig 3D ) —a characteristic size for exosomes [25] . We found that the purified vesicles were enriched in the tetraspannins CD63 and CD81 , also reported in exosomes [23] , and had very low amounts of the endoplasmic reticulum ( ER ) integral protein Calnexin or the ER lumen marker GRP94 ( Fig 3E ) . Furthermore , the possibility of contamination from ectosomes and other plasma membrane-derived vesicles was excluded by the absence of the ectosome marker CD11b in purified vesicles ( S2A Fig ) [26] . We also found that the release of exosomes from neutrophils increased upon fMLP addition , as the detection of the exosomal markers CD63 and CD81 increased with fMLP treatment ( Fig 3F ) . This increase was dependent on the concentration of fMLP added to cells ( S2B Fig ) . Moreover , the release of exosomes from cells was also observed following ionomycin addition but not with granulocyte macrophage colony-stimulating factor ( GM-CSF ) treatment ( S2C Fig ) . Most importantly , we determined that the enzymes responsible for the synthesis of LTB4 , namely 5-LO , FLAP , and LTA4H , were all present in fractions containing HSC70 ( another known exosome marker [27] ) ( Fig 3G ) and EM analyses revealed that both 5-LO and CD63 specifically decorate exosomes isolated in these fractions ( Fig 3H ) . The exosomes isolated from fMLP-stimulated neutrophils showed a high LTB4 content , which was dramatically reduced by pretreating neutrophils with the FLAP inhibitor MK886 ( Fig 3I ) . Together , these findings show that upon fMLP addition , neutrophils release exosomes containing LTB4 and the enzymes required for its synthesis . We next studied the extent by which exosomes mediate the effects of LTB4 on neutrophil function . We first purified exosomes from mCherry-5LO and CD63-GFP expressing cells following fMLP treatment in the presence and absence of MK886 and measured their LTB4 content . We found that the exogenous expression of either mCherry-5LO or CD63-GFP significantly increases exosomal LTB4 ( Fig 4A ) and , as we observed with neutrophils ( Fig 3I ) , LTB4 content was dramatically inhibited by pretreatment with MK886 ( Fig 4A ) . Quantification of the released exosomes showed that the decrease in LTB4 content in MK886-treated cells was not a result of a decrease in exosome release by these cells ( Fig 4B ) . More importantly , the addition of exosomes derived from mCherry-5LO or CD63-GFP expressing cells to neutrophils rapidly induced cellular polarization and adhesion , indicating that the exosomes readily activate resting neutrophils ( Fig 4C ) . These observations were further strengthened by the increase of both pErk1/2 and pAkt levels upon the exogenous addition of exosomes from mCherry-5LO or mCherry expressing cells to resting neutrophils ( Fig 4D and S3A Fig ) . Of note , the extent by which the exosome preparations increased pErk1/2 and pAkt levels was greater in cells exposed to mCherry-5LO exosomes versus mCherry exosomes ( S3A Fig ) , which could reflect the higher LTB4 content of exosomes derived from mCherry-5LO expressing cells , compared to mCherry expressing cells ( Fig 4A ) . The EZ-Taxiscan microfluidic device was used to assess whether exosomes are capable of inducing a chemotactic response . As seen in Fig 4E and S5 Movie , neutrophils were able to migrate towards exosomes derived from mCherry-5LO expressing cells with speeds and chemotactic indices ( CIs ) comparable to those observed in the presence of LTB4 alone . To determine whether the chemotactic response was mediated by LTB4 , we exposed neutrophils treated with the LTB4 receptor-1 antagonist LY223982 to exosomes . We found that LY223982 treatment dramatically reduced the chemotactic response of neutrophils to both LTB4 and exosomes derived from mCherry-5LO-expressing cells ( Fig 4F and S6 Movie ) . The antagonist-treated cells , however , did not show migration defects to saturating concentrations of fMLP , showing that LY223982 treatment did not impede chemotaxis due to nonspecific effects ( S3B Fig ) . We also observed that exosomes derived from MK886-treated mCherry-5LO cells displayed weak chemotactic activity compared to exosomes from control treated cells ( S7 Movie ) , further showing that the LTB4 present in exosomes is responsible for the chemotactic behavior . These results were confirmed biochemically by assessing the effects of the exogenous addition of LTB4 or exosomes on pAkt in neutrophils pretreated with LY223982 . As shown in Fig 4G , LTB4 or exosome addition gave rise to equal levels of pAkt for each exosome preparation and LTB4 amounts used and , most importantly , the pAkt response was blocked by pretreating the cells with LY223982 ( see S3C Fig , for quantification ) . Together , these findings establish the central role of LTB4 in exosome-mediated neutrophil activation . To specifically assess the role of exosome formation and release for LTB4 secretion , we knocked down Rab27a or neutral sphingomyelinase 2 ( nSmase2; SMPD2 gene ) using small hairpin RNAs ( shRNA ) in PLB-985 cells . Rab27a is critical in MVB docking to the plasma membrane , and its depletion was shown to reduce exosome secretion [28 , 29] . nSmase2 is important in exosome secretion by mediating the budding of exosomes into MVBs [30] . We achieved an efficient knockdown ( KD ) of Rab27a and SMPD2 expression in both undifferentiated and differentiated PLB-985 , and the KD of one gene did not alter the expression of the other ( S4A Fig ) . Of the six shRNA sequences screened for each gene , we selected Rab27a shRNA 1 ( sh1 ) and Rab27a shRNA 3 ( sh3 ) for further studies ( presenting 75% ± 4% and 70% ± 8% reduction of protein levels , respectively ) , whereas shRNA 2 ( sh2 ) and shRNA 4 ( sh4 ) were selected for SMPD2 ( presenting 82% ± 8% and 85% ± 6% reduction in protein levels , respectively ) . Using CD63 and CD81 as markers , we found that both SMPD2 and Rab27a KD cells show reduced exosome production upon treatment with 2 nM fMLP compared to control nonspecific shRNA ( NSshRNA ) cells ( Fig 5A and S4B Fig ) . The purity of the exosome preparations was assessed using the ectosome marker CD11b . Similarly , LTB4 content of purified exosomes from both KD cell lines was markedly lower than in control cell lines ( Fig 5B ) , although LTB4 levels across the different cell types were not different when normalized to the total exosomal protein content ( Fig 5C ) . These findings indicate that depletion of either nSmase2 or Rab27a does not affect LTB4 synthesis . When we measured the total amount of LTB4 secreted from each cell line in response to 1 nM fMLP , a physiological relevant concentration , we also observed a 75%–85% reduction in KD cells compared to control ( Fig 5D ) . Interestingly , stimulating cells with a saturating concentration of fMLP ( 1 μM ) reduced total LTB4 secretion by 40%–50% . Although this could result from residual Rab27a or nSmase2 activity in the KD cell lines , nonexosomal sources of LTB4 could become dominant under bulk activation conditions . We next set out to assess the chemotactic behavior of the KD cell lines . We found that KD of either SMPD2 or Rab27a specifically reduced the directional motility ( or CI ) of cells towards a subsaturating concentration of fMLP ( effective concentration 1 nM ) ( Fig 5E and S8 Movie ) , as the speed of migration remained unchanged for both KD cell lines ( Fig 5E ) . To further investigate whether exosomes play a role in the relay of primary chemotactic signals , we measured myosin light chain II ( MLCII ) phosphorylation following a subsaturating fMLP stimulation , a process acutely affected by the disruption of LTB4 signaling in neutrophils [6] . As we previously reported , we measured an increase in pMLCII levels in cells expressing NSshRNA in response to 2 nM fMLP ( Fig 5F ) . We also found that the increase of pMLCII levels was further accentuated in PLB-985 cells over expressing the receptor for LTB4 ( LTB4R1 ) , suggesting higher sensitivity of LTB4R1 expressing cells towards signal relay processes and the pivotal role that LTB4 plays in this response ( Fig 5F ) . In sharp contrast , and as observed in neutrophils where LTB4 synthesis is inhibited [6] , no fMLP-mediated pMLCII increase was measured in either Rab27a or SMPD2 KD cells ( Fig 5F and S4C Fig ) . Together , these findings show that exosome release regulates LTB4 secretion and signal relay during neutrophil chemotaxis . Importantly , the defects of the KD cells appeared to be highly specific . Both SMPD2 and Rab27a KD cells did not show defects in their ability to adhere upon fMLP stimulation ( S5A Fig ) , nor did they show any defect in their ability to increase pERK1/2 upon fMLP stimulation ( S5B Fig ) . Furthermore , as we found with MK8886 treatment [6] , the defects in directional migration were absent when a saturating concentration of fMLP ( effective concentration 100 nM ) was used ( S5C Fig and S9 Movie ) . Together , these findings indicate that neither Rab27a nor nSmase2 regulate the ability of cells to respond to fMLP and rule out KD specific bystander effects . We next sought to determine whether exosomal LTB4 acts in an autocrine and/or paracrine fashion during neutrophil chemotaxis . To do so , we labeled neutrophils with cytotracker red and tracked their movement as they chemotaxed towards fMLP using the under agarose assay . To quantify the data , we color-coded the displacement tracks as a function of the imaging time; blue representing the cell’s position during initial periods of migration and red the final . We found that compared to control cells ( Fig 6Ai; S10 Movie ) , MK886-treated cells migrated shorter distances and with less direction , although speed of migration was not affected ( Fig 6Aii; S10 Movie ) . We also found that treatment with GW4869 , an nSmase2 inhibitor , and a known inhibitor of exosome production [30] ( S6A Fig ) , similarly affected chemotactic motion ( Fig 6Bi and 6Bii; S11 Movie ) . Furthermore , and consistent with the chemotaxis defect of SMPD2 KD cells using the EZ-Taxiscan system ( Fig 5E ) , we observed a similar loss in CI and distance migrated in SMPD2 shRNA KD cells ( S6B Fig ) . We then asked if these defects could be rescued by the paracrine action of an exogenous source of exosomes . For this purpose , we mixed untreated neutrophils ( labeled green ) and treated ( MK886 or GW4869 ) neutrophils ( labeled red ) in equal proportion and recorded their motility towards fMLP . We observed a dramatic improvement of both directionality and total distance migrated in MK886- ( Fig 6Aiii; S10 Movie ) and GW4869-treated neutrophils ( Fig 6Biii; S11 Movie ) . No difference was found in the motility of untreated cells labeled either with the green or red dyes ( Fig 6Ai & 6Aiv and 6Bi & 6Biv ) , excluding any dye-specific effects . Furthermore , NSshRNA cells similarly rescued the directionality defect of SMPD2 shRNA KD cells ( S6B Fig ) . While we observed a dramatic improvement in the CI of GW4869- or MK886-treated cells through the paracrine effects of exosomal LTB4 ( Fig 6A and 6B ) , we noticed that the treated cells showed defects in the time required to migrate towards fMLP ( Fig 6C ) . It required up to 35 min for cells treated with either drug to traverse 300 μm , compared with 20 min for untreated cells ( Fig 6A and 6B ) . In addition , the treated cells exhibited slower speeds as well as a loss of directional persistence in the initial phases of migration ( Fig 6C ) . These defects in migration initiation and sensitization to a chemoattractant cue highlight a key role for exosomal LTB4 in autocrine signaling . One may argue that exosomal LTB4 merely increases the robustness of the chemotactic response by regulating the cellular machinery as opposed to acting as a chemotactic beacon . We tested this by mixing cells that cannot detect fMLP with cells defective in exosome release and observed their migration towards fMLP . Due to their higher sensitivity to LTB4 , we used LTB4R1 overexpressing cells as receiving cells in these experiments ( Fig 5F ) . We treated green-labeled LTB4R1 cells with the FPR-1 antagonist cyclosporin H ( CsH ) [31] and mixed them with red-labeled cells expressing control shRNA ( NSshRNA ) , which were also treated with CsH . These cells did not migrate towards fMLP , confirming the efficacy of the antagonist treatment , and no such defects were observed when control-labeled cells were mixed ( Fig 7A ) . However , the motility defect of CsH-treated LTB4R1 cells was readily rescued when mixed with untreated NSshRNA cells , but not with MK886-treated NSshRNA cells ( Fig 7B ) . These findings reiterate that LTB4 released by NSshRNA cells is responsible for the trans-recruitment of CsH-treated cells . Importantly , no rescue in motility was observed in CsH-treated LTB4R1 cells when they were mixed with Rab27a or SMPD2 KD cells ( Fig 7C ) and in GW4869- or MK866-treated neutrophils ( S7 Fig ) . Together , these observations indicate that exosomal LTB4 relays chemotactic signals in response to primary attractants during neutrophil chemotaxis .
Our prior studies on neutrophils migrating in chemotactic gradients identified LTB4 as an important signal relay molecule that increases the recruitment range of neutrophils to sites of inflammation [6 , 11] . This observation led us to study the mode of LTB4 release from cells and its effective dissemination . In this study , we show that LTB4 is packaged in MVBs that are released as exosomes during neutrophil chemotaxis . We present this as a mechanism through which hydrophobic low-diffusible lipid molecules , like LTB4 , mediate the signal relay process . Our study proposes that neutrophils migrating in primary chemoattractant gradients release exosomes containing LTB4 and LTB4 synthesizing enzymes . These exosomes subsequently act in an autocrine manner—by sensitizing cells towards the primary chemoattractant—and in a paracrine manner , by acting as molecular beacons for following cells ( Fig 8 ) . We found that both LTB4 and 5-LO are present in MVBs and exosomes , which also contain LTA4H and FLAP . The presence of LTB4 synthesizing enzymes has previously been reported in exosomes derived from macrophages and dendritic cells [32] . However , unlike dendritic cells that constitutively produce exosomes [33] , we detected little or no release of exosomes in unstimulated neutrophils . Instead , we observed exosome release upon treatment with ionomycin as well as following the addition of fMLP , but not GM-CSF , indicating that exosome release is a result of primary stimulation and not a priming event . fMLP is also known to increase the release of plasma membrane-derived secretory vesicles and ectosomes [34] that have considerable overlap with exosomes in terms of size ( 50–200 nm ) . However , owing to their differences in lipid composition [35] , we were able to separate exosomes from ectosomes using iodixanol density gradients . Moreover , the purified exosomes were able to activate resting neutrophils , indicating proinflammatory responses compared to the reported anti-inflammatory responses of ectosomes [34] . Finally , using Rab27a and SMPD2 KD cells , we established that exosomes represent the primary pathway for LTB4 release in chemotaxing neutrophils under physiological stimulation conditions . Impaired neutrophil recruitment has been observed in Rab27a KO mice under in vivo neutrophil recruitment models [36] . Furthermore , neutrophils deficient in vesicle fusion show less migration in vitro and in an in vivo model of gout [37] . The migration defects we observed in Rab27a and SMPD2 KD cells therefore provide valuable mechanistic insights into the migration defects observed in vivo . Furthermore , the docking of 5-LO containing MVBs at the back of cells and the subsequent release of exosomes are reminiscent of the molecular beacon model first shown in chemotaxing Dictyostelium cells [17] . This is similar to recent in vivo observations made by Sung and colleagues [38] , where exosomes released by HT1080 cells bind to integrins and act as adhesion trails for cellular guidance . Under uniform stimulation conditions , exosomal LTB4 levels range from 2–10 nM and are sufficient to elicit neutrophil migration in chemotaxis assays . This , however , may not reflect the actual amount of LTB4 being released from exosomes during chemotaxis for the following reasons: ( i ) exosome release is fMLP dose-dependent and hence a function of the primary chemoattractant gradient; ( ii ) purified exosomes were prepared by uniformly stimulating neutrophils with subsaturating fMLP concentrations and losses incurred during the purification processes could lead to underestimated amounts of exosomes recovered; ( iii ) the rate of LTB4 release from exosomes and the effective LTB4 gradient across and between cells cannot be reliably measured in real time due to the lack of sensitive methodologies; and ( iv ) the presence of the LTB4 synthesizing enzymes , 5-LO , LTA4H , and FLAP in neutrophil exosomes along with the presence of the primary substrate AA [39] , suggest active exosomal LTB4 synthesis . Indeed , exosomes isolated from dendritic cells are capable of synthesizing AA metabolites [32] . Although our study elucidates the trafficking and secretion of LTB4 during neutrophil chemotaxis , the mechanisms regulating the release of exosomal LTB4 remain to be determined . These may include membrane diffusion , passive or active lysis through released neutrophil proteases , or through docking to cell surface heparan sulfate proteoglycans [40] . Alternatively , active LTB4 release may be mediated through ABC transporters [41] or other types of active processes . The availability of methods that are sensitive and , most importantly , capable of measuring LTB4 in real time and space will help elucidate how LTB4 gradients are established and propagated during neutrophil chemotaxis . EM analyses on chemotaxing neutrophils provide key insights into the mechanisms that regulate LTB4 synthesis and trafficking . In addition to its enrichment on the nuclear envelope , 5-LO is localized to the plasma membrane and to MVBs . Most interestingly , the 5-LO containing MVBs are often observed in close proximity to the nuclear envelope . The origination of MVBs from the nuclear envelope has previously been published [42 , 43] . Moreover , Record and colleagues have reported a close association of MVBs containing internalized exosomes with the nucleus in RBL-2H3 cells [39] . From these findings , we envision that the nuclear envelope represents a central hub for 5-LO distribution , leading to its segregation in two independent cellular pools: ( i ) the nucleus , where LTB4 has been shown to bind to the peroxisome proliferator-activated receptor α [44] and regulate transcriptional responses and ( ii ) MVBs , where LTB4 is destined to secretion and signal relay . We reason that the 5-LO plasma membrane labeling arises upon MVB fusion and release of exosomes into the extracellular environment . Our findings also bring into question the role of the canonical ESCRT-dependent exosomal biogenesis machinery in the synthesis and release of LTB4-containing exosomes . Indeed , the fact that nSmase2 depletion results in a near complete inhibition of exosome release suggests that neutrophil exosome biogenesis predominantly relies on a ceramide-dependent and ESCRT-independent mechanism . Our observations are consistent with a study by Trajkovic et al . showing that secretion of CD63 in a mouse oligodendroglial cell line is blocked by a sphingomyelinase inhibitor , but not by a dominant-negative ESCRT component [30] . On the other hand , ESCRTIII , one of the key members of the ESCRT complex involved in ILV formation and MVB biogenesis , has recently been shown to localize at juxtanuclear regions that mediate the reformation of the nuclear envelope during cell division [45] . It hence remains to be determined whether the ESCRT machinery is involved in regulating the assembly of 5-LO-containing MVBs from the nuclear envelope during neutrophil chemotaxis . Our findings show that exosomal LTB4 regulates neutrophil chemotaxis in an autocrine and paracrine fashion . First , we found that Rab27a and SMPD2 KD cells as well as neutrophils treated with an nSmase2 inhibitor exhibit profound defects in directionality and recruitment range towards fMLP . Similar defects were observed in neutrophils derived from Rab27a KO mice [36] and were primarily attributed to granule exocytosis , a process that is closely related to exosome biogenesis and release . Chemotaxis has not been studied in neutrophils isolated from SMPD2 KO mice , although its role may be inferred through the effects of GW4869 on neutrophil migration . GW4869 did not inhibit neutrophil speed but abrogated directional migration towards fMLP [46] , an observation that mirrors our own observation with GW4869 and SMPD2 KD cells . Second , we found that the defects in directionality and recruitment range of SMPD2 KD as well as GW4869-treated cells are rescued by the presence of control cells that provide exosomes to relay signals . One could argue that the rescue was simply due to supplementation of ceramide from the released exosomes and not related to LTB4 content [46] . However , we observed a very similar phenotypical recovery by inhibiting LTB4 synthesis alone . More importantly , unlike control cells , the KD cells were unable to send a paracrine signal to neutrophils with blocked FPR1 , clearly establishing that exosomes represent a critical bearer of the signal relay message . Exosomes are well-established mediators of intercellular communication [47] and have been known to mediate cell migration in various systems [38] . This work establishes that exosomal communication is extremely efficient and critical in fast moving cells and occurs at a faster rate compared to the slow constitutive release of exosomes reported in the literature [29] . Moreover , our findings add valuable insight into the mechanisms that underlie chronic inflammatory conditions , such as asthma and rheumatoid arthritis [48] , as well as lung cancer progression [49] , where LTB4 plays a central role . We envision that the secretion of other signals that foster communication between cells in harsh extracellular environments are similarly processed through exosome packaging .
OptiPrep , Histopaque 1077 , fMLP , IL8 , and LY294002 were obtained from Sigma-Aldrich ( St . Louis , MO ) . LTB4 , MK886 and LY223982 were purchased from Cayman Chemical ( Ann Arbor , MI ) . Anti-5-LO , Anti-p-Akt ( clone C31E5E for S473 ) , anti-AKT , anti-myosin light chain 2 , anti-p-myosin light chain 2 ( Ser19 ) , anti-total ERK1/2 , and anti-p-Erk1/2 ( clone D13 . 14 . 4E ) rabbit antibodies were all purchased from Cell Signaling Technology ( Beverly , MA ) . Anti-MMP9 , anti-MPO , anti-CD63 , anti-LAMP1 , and anti-GM130 were obtained from Abcam ( Cambridge . MA ) . CD11b APC antibody was purchase from BD Biosciences ( San Jose , CA ) . Heparinized whole blood was obtained by venipuncture from healthy donors . Neutrophils were isolated using dextran sedimentation ( 3% dextran/0 . 9% NaCl ) coupled to differential centrifugation over Histopaque 1077 [50] . Residual erythrocytes were removed using hypotonic lysis with 0 . 2% and 1 . 6% saline solutions . HEK293T cells ( ATCC ) and Phoenix cells ( Orbigen , San Diego , CA ) were maintained as previously described [51] . For virus packaging , 80% confluent cells were used for transient transfection using Lipofectamine reagent according to manufacturer’s protocol ( Life Technologies ) . PLB-985 cells were maintained in an undifferentiated state and differentiated as described [51] . The status of differentiation was monitored by Mac-1 staining . PLB985 cells expressing mCherry-5LO and CD63-GFP as well as coexpressing both CD63-GFP and mCherry-5LO were created using a retroviral approach . Rab27a and SMPD2 KD were achieved using the pGIPZ lentiviral system ( GE , Dharmacon ) . 5 x 108 cells were primed with GM-CSF ( 5 ng/ml; R&D Systems ) for 30 min in RPMI without phenol red containing 5% FBS and 10 mM HEPES , centrifuged and resuspended in ice-cold PBS for 30 min to reduce basal exocytosis levels . Cells were harvested and resuspended in mHBSS with or without 2 nM fMLP for 30 min at 37°C and centrifuged at 500 xg for 5 min . To assess the exosome-producing ability of various stimulants , unprimed cells were resuspended in basal buffer ( 25 mM HEPES , pH 7 . 4 , 140 mM NaCl , 4 . 7 mM KCl , 1 . 4 mM MgCl2 , and 10 mM glucose ) and stimulated for 30 min with either ionomycin ( 1 μM ) supplemented with CaCl2 ( 2 mM ) , fMLP ( 1 μM ) , GM-CSF ( 5ng/ml ) , or DMSO ( 0 . 1% ) . The supernatant was further centrifuged at 2 , 000 xg for 20 min . The resulting supernatants were filtered twice through low protein-binding 0 . 45 μm filters and subsequently centrifuged at 100 , 000 xg for 60 min at 4°C . The recovered pellet , containing a heterogeneous mixture of crude vesicles , was resuspended in 500 μl PBS . Further purification of the exosomes was performed as previously reported [52] . Briefly , the resuspended vesicles were loaded onto a discontinuous gradient of iodixanol ( 40% , 20% , 10% and 5% , w/v solutions in 0 . 25 M sucrose/10 mM Tris , pH 7 . 5 . ) , centrifuged at 100 , 000 xg for 18 h at 4°C . Fractions were collected , diluted in PBS , and centrifuged at 100 , 000 xg . The pellet was washed in PBS , and the recovered fractions were used as described . Purified exosomes were incubated with 3 . 9 μm diameter aldehyde/sulfate latex beads ( Life Technologies ) for 15 min at RT followed by the addition of 10 μg BSA for an additional 15 min . The incubation volume as increased to 250 μl in PBS and incubated overnight on a turnover rocker . The incubation was stopped the following day by adding glycine to a final concentration of 100 mM . The exosome-coated beads were washed twice in FACS wash ( 3% FCS and 0 . 1% NaN3 in PBS ) and resuspended in 400 μl FACS wash . The beads were incubated for 30 min with each conjugated primary antibody , washed and analyzed on a FACSCalibur flow cytometer ( BD Biosciences ) . Bead-based flow cytometry was also performed with CD63 antibody-coated dynabeads according to manufacturer’s protocol ( Invitrogen ) . Freshly isolated neutrophils ( 50 x 106 ) were incubated with DFP for 15 min at 37°C , pelleted and resuspended in 1X Disruption Buffer ( DB; 100 mM KCl , 3 mM NaCl , 3 . 5 mM MgCl2 , 1 . 5 mM EGTA , and 10mM 1 , 4-piperazinediethanesulfonic acid ( PIPES ) , pH 7 . 2 , including 1 mM sodium salt of ATP , and 0 . 5 M PMSF ) with or without 2 nM fMLP . The reaction was stopped after 2 min by adding an equal volume of ice-cold DB . Cavitation was carried out for 5 min at 500 psi on ice . Cavitates were collected and EGTA was added to the final concentration of 1 . 5 mM . Lysed cells were centrifuged at 400 xg for 20 min to remove cell debris and nuclei and 15 , 000 xg for 15 min to remove mitochondria . The supernatant was then top loaded onto a 5%–30% continuous OptiPrep gradient prepared by appropriately diluting a 60% optiprep stock with 3X disruption buffer . The gradient was centrifuged at 34 , 000 xg for 18 . 5 h , and 1 ml fractions were collected . Fractionates < 1 . 04 g/l containing primarily cytosolic fraction were discarded and the rest analyzed by immunoblotting after TCA ( Trichloroacetic acid ) precipitation . To determine the density of each fraction , a control OptiPrep gradient containing a blank containing disruption buffer was run in parallel . Fractions were collected as described , serially diluted 1:10 , 000 with water , and the iodixanol concentration determined by absorbance at 244 nm using a molar extinction coefficient of 320 L g−1cm−1 . Chemotaxis was assessed using the under-agarose assay or the EZ-Taxiscan as previously described [51 , 53] . Cells ( 5 x 105 cells/well ) were plated on 96-well plates coated with fibronectin , stimulated for 30 min at 37°C , shaken at 2 , 000 RPM ( radius = 1 cm ) for 10 s on an orbital shaker , and unbound cells were removed . The remaining cells were fixed ( 4% PFA ) and stained with crystal violet 5 mg/mL in 2% ethanol . Crystal violet was extracted by the addition of 2% SDS and absorbance was measured at 570 nm . Neutrophils were allowed to migrate under-agarose towards an fMLP gradient for 1 hr after which the agarose was removed and the cells were fixed in 0 . 1 M cacodylate buffer pH 7 . 4 , containing 5% sucrose , 4% paraformaldehyde , and 0 . 1% glutaradehyde for 2 h at 22°C . The cells were blocked and permeablized in PBS + 5% BSA , 0 . 1% NaN3 , and 0 . 1% saponin , followed by the addition of 5-LO antibody or a control rabbit antibody , and subsequently with rabbit secondary antibody conjugated to nanogold ( 4 nm; Nanoprobes ) . The cells are fixed with 2 . 5% glutaraldehyde followed by gold enhancement of the nanogold particles ( Nanoprobes ) , osmium tetroxide staining , and epoxy block preparation . Immunoelectron staining was performed on thin cut block sections ( EML , NCI ) using Hitachi H7600 Transmission electron microscope . MVBs with greater than 5 immunogold particle/μm were considered 5-LO positive MVBs for quantification purposes . In separate experiments , uranyl acetate negative-stain immuno-EM was performed to analyze the size and number of purified exosome samples . Cytotracker dyes were added to differentiated PLB-985 cells or freshly isolated neutrophils according to manufacturer’s protocol ( Life Technologies ) , washed and resuspended in modified HBSS ( mHBSS ) containing appropriate inhibitors/antagonists . Cells were then washed and 2 . 5 x 105 cells of differently labeled cells were mixed . These cells were resuspended in 5 μl mHBSS and loaded onto an under-agarose chamber . Time-lapse imaging was performed using a Plan-Neofluar 10x/0 . 30 objective ( Zeiss Axiovert S100 microscope ) . Cell tracks were extracted and analyzed using a method developed by McCann et al . [54] supplemented by tracking through ImageJ . CI and Δα ( persistance ) were used to analyze motility behavior . CI was calculated by taking the cosine of the angle between the final displacement vector and the line connecting the chemotactic source to the initial position of the cell . A value of 1 designates perfect chemotaxis . Δα is measured through the angular change between the displacement vector at any given time and the preceding displacement vector . Larger Δα designates lower persistence . Total LTB4 levels were measured using an ELISA kit ( R&D Systems ) . Briefly , neutrophils or PLB-985 cells were resuspended at 1 x 106 cells/ml in PBS and incubated for 30 min on ice . GM-CSF ( 5 ng/ml; R&D Systems ) was added and cells were further incubated for 1 hr at 37°C , centrifuged at 400 xg for 5 min , resuspended in a volume of 200 μl at 24 x 106 cells/ml in RPMI and incubated at 37°C until stimulated . The reactions were stopped by adding cold PBS . Cells were centrifuged and LTB4 in supernatants were assayed according to manufacturer’s instructions . Exosomal LTB4 was measured using the EIA kit according to manufacturer’s protocol ( Cayman Chemicals ) after disruption by sonication . To measure LTB4 in the vesicular fractions of lysed neutrophils , total lipids were first extracted from the different fractions using a method outlined by McColl et al . [55] . The extracted lipids were acidified to pH 4 . 0 loaded onto a C18 SPE column and eluted with ethyl acetate containing 10% methanol , evaporated under inert nitrogen stream and resuspended in buffer to obtain a concentrated pool of LTB4 . LTB4 levels were measured using the EIA kit ( Cayman chemicals ) according to manufacturer’s recommendations . Neutrophils were either resuspended in RPMI ( 1% bovine serum ) or plated on fibronectin-coated chambers ( for pMLCII ) and incubated with inhibitors and diisopropyl fluorophosphates ( DFP ) ( 2 mM ) for 10 min at 37°C . Cells were then stimulated with fMLP , collected at specific time points , lysed for 15 min in ice-cold lysis buffer ( 20 mM Tris pH 7 . 5 , 10 mN NaCl , 1 mM EDTA , 0 . 1% NP-40 , 3 mM DFP , 2 mM orthovanadate , 10 g/mL aprotinin , 10 g/mL leupeptin , 2 . 5 mM napyrophosphate , Complete Protease Inhibitor cocktail [Roche Diagnostics , Indianapolis , IN] ) , and processed for SDS-PAGE analyses . XM_005267109 ( SMPD2 shRNA1 ) : TGCTGTTGACAGTGAGCGCCGCCGTTGATGTGTGTGCTAATAGTGAAGCCACAGATGTATTAGCACACACATCAACGGCGATGCCTACTGCCTCGGA XM_005267109 ( SMPD2 shRNA2 ) : TGCTGTTGACAGTGAGCGCAGGCAACACAATGGTACCCAATAGTGAAGCCACAGATGTATTGGGTACCATTGTGTTGCCTTTGCCTACTGCCTCGGA NM_003080 ( SMPD2 shRNA3 ) : TGCTGTTGACAGTGAGCGAGGGGACAGAACTAAAGAACAATAGTGAAGCCACAGATGTATTGTTCTTTAGTTCTGTCCCCCTGCCTACTGCCTCGGA XM_005267109 ( SMPD2 shRNA4 ) : TGCTGTTGACAGTGAGCGCCCGCATTGACTACGTGCTTTATAGTGAAGCCACAGATGTATAAAGCACGTAGTCAATGCGGATGCCTACTGCCTCGGA NM_003080 ( SMPD2 shRNA5 ) : TGCTGTTGACAGTGAGCGATGGGTTTTACATCTCCTGTAATAGTGAAGCCACAGATGTATTACAGGAGATGTAAAACCCAGTGCCTACTGCCTCGGA NM_004580 ( Rab27a shRNA1 ) : TGCTGTTGACAGTGAGCGCGCTCAATGTCTTTGAGTATTATAGTGAAGCCACAGATGTATAATACTCAAAGACATTGAGCTTGCCTACTGCCTCGGA NM_183236 ( Rab27a shRNA2 ) : TGCTGTTGACAGTGAGCGAGCCCAGAGTCTTACATTTAAGTAGTGAAGCCACAGATGTACTTAAATGTAAGACTCTGGGCCTGCCTACTGCCTCGGA NM_004580 ( Rab27a shRNA3 ) : TGCTGTTGACAGTGAGCGCGGCTGCAGCTTTATTAGCTTATAGTGAAGCCACAGATGTATAAGCTAATAAAGCTGCAGCCTTGCCTACTGCCTCGGA XM_005254577 ( Rab27a shRNA4 ) : TGCTGTTGACAGTGAGCGCCCAGTGTACTTTACCAATATATAGTGAAGCCACAGATGTATATATTGGTAAAGTACACTGGTTGCCTACTGCCTCGGA NM_183234 ( Rab27a shRNA5 ) : TGCTGTTGACAGTGAGCGCAGGGAAGACCAGTGTACTTTATAGTGAAGCCACAGATGTATAAAGTACACTGGTCTTCCCTATGCCTACTGCCTCGGA NM_183236 ( Rab27a shRNA6 ) : TGCTGTTGACAGTGAGCGAATGCCTCTACGGATCAGTTAATAGTGAAGCCACAGATGTATTAACTGATCCGTAGAGGCATGTGCCTACTGCCTCGGA | Neutrophils represent the first line of attack against infections and inflammatory insults . The ability of neutrophils to reach these sites , a key feature in the resolution of infections , is mediated by their capacity to sense and migrate directionally to the core of the inflammation site . Chemicals released at the site of inflammation are known as primary attractants . The binding of these attractants to receptors on the surface of neutrophils leads to the secretion of secondary attractants that amplify the reach of primary attractants . We studied the mechanism by which secondary attractants are released from neutrophils . We found that the secretion of a key secondary attractant is mediated in the form of small vesicles called exosomes . These exosomes originate inside the cells , encapsulated in larger vesicles called multivesicular bodies . We purified exosomes from activated neutrophils and show that they contain the machinery to synthesize this secondary attractant and act specifically to elicit neutrophil motility . The inhibition of exosome release leads to a loss of secretion of the secondary attractant as well as a loss in directional motility . Together , our findings provide insight into the mechanisms cells use to protect labile attractants from harsh extracellular environments and communicate directional cues during inflammatory responses . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [] | 2016 | Exosomes Mediate LTB4 Release during Neutrophil Chemotaxis |
The anthelmintics ivermectin ( IVM ) and moxidectin ( MOX ) display differences in toxicity in several host species . Entrance into the brain is restricted by the P-glycoprotein ( P-gp ) efflux transporter , while toxicity is mediated through the brain GABA ( A ) receptors . This study compared the toxicity of IVM and MOX in vivo and their interaction with GABA ( A ) receptors in vitro . Drug toxicity was assessed in Mdr1ab ( −/− ) mice P-gp-deficient after subcutaneous administration of increasing doses ( 0 . 11–2 . 0 and 0 . 23–12 . 9 µmol/kg for IVM and MOX in P-gp-deficient mice and half lethal doses ( LD50 ) in wild-type mice ) . Survival was evaluated over 14-days . In Mdr1ab ( −/− ) mice , LD50 was 0 . 46 and 2 . 3 µmol/kg for IVM and MOX , respectively , demonstrating that MOX was less toxic than IVM . In P-gp-deficient mice , MOX had a lower brain-to-plasma concentration ratio and entered into the brain more slowly than IVM . The brain sublethal drug concentrations determined after administration of doses close to LD50 were , in Mdr1ab ( −/− ) and wild-type mice , respectively , 270 and 210 pmol/g for IVM and 830 and 740–1380 pmol/g for MOX , indicating that higher brain concentrations are required for MOX toxicity than IVM . In rat α1β2γ2 GABA channels expressed in Xenopus oocytes , IVM and MOX were both allosteric activators of the GABA-induced response . The Hill coefficient was 1 . 52±0 . 45 for IVM and 0 . 34±0 . 56 for MOX ( p<0 . 001 ) , while the maximum potentiation caused by IVM and MOX relative to GABA alone was 413 . 7±66 . 1 and 257 . 4±40 . 6% , respectively ( p<0 . 05 ) , showing that IVM causes a greater potentiation of GABA action on this receptor . Differences in the accumulation of IVM and MOX in the brain and in the interaction of IVM and MOX with GABA ( A ) receptors account for differences in neurotoxicity seen in intact and Mdr1-deficient animals . These differences in neurotoxicity of IVM and MOX are important in considering their use in humans .
Macrocyclic lactones ( MLs ) are a large family of broad spectrum antiparasitic drugs . Ivermectin ( IVM ) , an avermectin macrocyclic lactone , is used in humans through mass drug administration programs for the control of onchocerciasis , a tropical parasitic disease caused by the filarial nematode Onchocerca volvulus . Moxidectin ( MOX ) , a milbemycin ( non-avermectin ) macrocyclic lactone is currently being evaluated for possible use against O . volvulus in humans [1] , [2] . Besides this , both drugs are commonly used in veterinary medicine in livestock to treat diseases caused by gastrointestinal nematodes and external parasites and for the prevention of Dirofilaria immitis infection in dogs . In general , MLs have a high margin of safety in mammals ( Pulliam & Preston , 1989 ) . Indeed , P-glycoprotein ( P-gp , MDR1/ABCB1 ) , a plasma membrane efflux pump belonging to the ATP-binding cassette ( ABC ) transporters family , efficiently restricts their penetration in the brain at the blood–brain barrier [3] , thus preventing their binding to the γ-aminobutyric acid type A ( GABA ( A ) ) receptor [4] , [5] . However , neurotoxicity of IVM has been reported in mammals in cases of P-gp deficiency or overdose . In humans , IVM has been administered to tens of millions of individuals and is usually exceptionally safe when given at therapeutic doses [6] . Accumulation of the drug in the brain as a consequence of massive overdoses ( more than 100 times the normal doses ) is associated with prolonged coma and death [7] , [8] . In addition , severe adverse events ( SAEs ) have been described after IVM treatment ( 0 . 15 mg/kg ) in some individual humans carrying high burdens of the filarial nematode Loa loa [9] , [10] , [11] , [12] , [13] and IVM SAEs were recently associated with functionally relevant polymorphisms in human MDR1 gene [14] . In 2005 and 2008 , prior to initiation of MOX Phase II studies in humans infected with O . volvulus its safety was reviewed by WHO committees [http://whqlibdoc . who . int/publications/2008/9789241597333_eng . pdf , http://www . who . int/tdr/publications/tdr-research-publications/moxidectin/en/index . html] with conclusions that Phase II studies should proceed . Nevertheless , in contrast to the situation with IVM , MOX has , so far , only been administered to approximately 1 , 700 people in Phase I , II and III supervised clinical studies , without evidence of any serious adverse events [http://clinicaltrials . gov/ct2/show/NCT00856362 , http://www . clinicaltrials . gov/ct2/show/NCT00300768 , http://clinicaltrials . gov/ct2/show/NCT00790998] . IVM administrated at the therapeutic dose of 0 . 2 mg/kg to MDR1-deficient dogs provokes severe signs of neurotoxicosis including apparent depression , ataxia , somnolence and tremor [15] , [16] . However , in dogs sensitive to 120 µg/kg IVM administered orally , a similar molar dose rate of MOX given by the same route did not produce any toxicological signs [17] . In another study , P-gp-deficient dogs that were sensitive to 120 µg/ml of IVM ( 20× the therapeutic dose rate ( 6 µg/kg ) for IVM as a monthly heartworm preventative ) did not produce signs of toxicosis following exposure to MOX at 100 µg/kg ( more than the molar equivalent to 120 µg/ml IVM , and 33-fold the therapeutic dose rate ( 3 µg/kg ) for MOX as a monthly heartworm preventative ) given daily for 7 days [18] . In fact , MOX has been safely used on P-gp-deficient Collie dogs up to 32 . 5 mg/kg by spot-on application [19] . However , in addition to dose rate , route of administration can markedly affect toxicity and topical application is known to produce low bioavailability . In wild-type mice , the LD50 for oral administration of IVM is around 30 mg/kg [20] while the LD50 for MOX also given orally is 86 mg/kg [21] . Structural differences between MOX and IVM exist and include the absence of a disaccharide at position 13 of the macrocyclic ring in MOX , MOX being protonated ( −H ) at position 13 , and the presence of a 23-methoxyimino group and other substitutions which distinguish it from IVM ( Figure 1 ) . These molecular differences presumably account for differences in their interaction with various invertebrate ligand-gated ion channels . Indeed , in Caenorhabditis elegans marked differences have been observed in the effects of IVM and MOX on pharyngeal pumping and motility ( manifestations of the actions of these different MLs on ligand-gated chloride channels ) [22] . Furthermore , difference in their interaction with mammalian ABC transporters has been demonstrated , MOX being much less ( 10-fold ) effective than IVM ( and other avermectins ) in inhibiting transport activity by P-gp [23] . It is therefore reasonable to think that differences in drug interaction with mammalian GABA receptors could account for the differential toxicity of IVM and MOX . While the interaction of IVM with mammalian GABA receptors has been known for some time , little is known about the interaction of MOX with these receptors and its potential CNS toxicity . In this context , the objectives of this work were ( i ) to compare the in vivo toxicity of MOX and IVM , ( ii ) to evaluate their accumulation in brain , and ( iii ) to compare their activity on the mammalian GABA ( A ) receptor . Given the major role of P-gp in the prevention of the penetration of MLs into the brain , acute toxicity in vivo and accumulation in the brain of these two MLs was assessed with Mdr1ab ( −/− ) knockout mice , deficient for the two P-gp murine isoforms , Mdr1a and Mdr1b .
In vivo studies were conducted in mice under European laws on the protection of animals ( 86/609/EEC ) . Protocols are performed under procedure and principal for good clinical practice ( CVMP/VICH 59598 ) . The protocols for experimentation on rodents used in this manuscript have been approved by the local institutional animal care and ethics committee which is the “Direction Départementale des Services Vétérinaires de Haute-Garonne” . The specific approval number for this study approval is B31555-25 . Ivermectin , γ-aminobutyric acid ( GABA ) , collagenase type I , dimethyl sulfoxide ( DMSO ) and kanamycin solution ( 50 mg kanamycin/ml in 0 . 9% NaCl ) were purchased from Sigma-Aldrich Chimie ( St Quentin Fallavier , France ) . Moxidectin was a gift from Fort Dodge Animal Health . Penicillin-streptomycin solution ( 10 , 000 units/ml penicillin and 10 , 000 µg/ml of streptomycin ) was obtained from Invitrogen - Life Technologies ( Cergy Pontoise , France ) . All other chemicals were obtained from Sigma-Aldrich , unless otherwise stated . Rat GABA ( A ) α1 , β2 and γ2 subunit constructs were a kind gift from Dr Erwin Sigel . FVB Mdr1ab ( −/− ) mice , deficient for the two murine P-gps encoded by abcb1a and abcb1b genes ( GenBankTM Accession numbers NM011076 and NM011075 , respectively ) , were obtained from Taconic ( NY , USA ) . Animals were kept under controlled temperature with a 12/12 h light/dark cycle . They received ad libitum a standard diet ( Harlan Teklad TRM Rat/Mouse Diet; Harlan Teklad , Gannat , France ) and municipal water . Mice were randomly assigned to groups and weighed . Experiments were carried out on 10–14 week-old mice ( 25–30 g ) . Suitable dilutions of a stock solution of IVM or MOX in DMSO were made in propylene glycol/formaldehyde ( 60∶40 v/v ) for subcutaneous administration or in commercial formulations for oral administration in order to administer to each mouse the designated dose ( µmol/kg body weight ( bw ) ) in 100 µl . Each formulation was checked for drug concentration prior to administration . For drug plasma concentration and brain accumulation assessment , Mdr1ab ( −/− ) mice were injected subcutaneously with an equivalent molar dose rate for both MLs in Mdr1ab ( −/− ) mice ( 6 animals per group ) : 0 . 23 µmol/kg , corresponding to 0 . 20 mg/kg and 0 . 15 mg/kg for IVM and MOX , respectively . Mice were sacrificed at 2 or 24 h after treatment . To evaluate the plasma and brain concentrations of MLs as a function of the administrated dose , Mdr1ab ( −/− ) mice ( 6 per dose rate ) were injected subcutaneously at various doses of IVM or MOX that were not lethal in 24 h: 0 . 1–0 . 3 mg/kg ( 0 . 114–0 . 342 µmol/kg bw ) and 0 . 46 to 1 . 3 mg/kg bw ( 0 . 23–2 µmol/kg bw ) respectively . In parallel , wild-type mice ( 3 per dose rate ) were orally administered with doses close to their respective LD50: 20 and 25 mg/kg bw ( 22 . 8 and 28 . 6 µmol/kg bw ) for IVM [20] and 18 . 3 and 40 mg/kg bw ( 28 . 6 and 62 . 5 µmol/kg bw ) for MOX [21] . Mice were anesthetized 2 h or 24 h after administration with isoflurane and heparinized blood samples were collected from the orbital sinus vein . Immediately thereafter , mice were sacrificed by cervical dislocation and brains were rapidly removed . Blood samples were centrifuged at 1500 g for 10 min at 4°C and the plasma fraction was collected and stored at −20°C until analysis . The brains were washed in saline solution and frozen at −20°C until analysis . For acute toxicity experiments , Mdr1ab ( −/− ) mice ( 2–8 per dose rate ) were injected subcutaneously with drug solutions at dose rates ranging from 0 . 1–1 . 75 mg/kg bw ( 0 . 11–2 . 0 µmol/kg bw ) for IVM and 0 . 2–8 . 2 mg/kg bw ( 0 . 31–12 . 9 µmol/kg bw ) for MOX . Mice were observed for a period of 2 weeks and any neurological signs were recorded every 60 min for the first 12 hours and thence minimally twice per day . Mice were euthanized when severe tremors or ataxia were noted . The effective LD50 values ( dose rate that caused 50% lethality ) were determined graphically in Mdr1ab ( −/− ) . ML concentrations were determined in plasma and brain by high performance liquid chromatography ( HPLC ) with fluorescence detection according to previously described and validated methods [24] , [25] . In brief , plasma and tissues were homogenised in acetonitrile ( 1∶1 v/v or 1∶2 v/w , respectively ) . Samples were centrifuged at 2000 g and the supernatant applied to a Supelco C18 cartridge ( Supelco Inc . , Bellefonte , PA , USA ) by using automated solid phase extraction ( SPE ) . The extraction recoveries for the two molecules were 0 . 95 for plasma and 0 . 65 for brain . The eluate was evaporated and the dry extract was processed to obtain a fluorophore derivative by dissolving it in 1N-methylimidazole and trifluoroacetic anhydride solutions . Samples were injected into the HPLC system ( PU980 pump , Jasko , Tokyo , Japan; 360 automatic injector , Kontron , Paris , France; RF-551 fluorescence detector , Shimadzu , Kyoto , Japan ) . For IVM and MOX a Supelcosil LC18 column ( 250×4 . 6 mm , 5 µm , Supelco , Bellefonte , PA , USA ) was used with acetic acid ( 0 . 2% in water ) ∶methanol∶acetonitrile ( 4∶40∶56 , v/v/v ) as mobile phase . Oocytes from X . laevis , injected with foreign cDNA of the receptor of choice , are a commonly used tool for studying the activity of plasma membrane receptors [26] . The cDNAs coding for the α1 , β2 and γ2 subunits of the rat GABA ( A ) receptor channel have been described previously [27] , [28] . cDNAs were dissolved in water and stored at −80°C . Isolation of oocytes from the frogs , defolliculation , culturing of the oocytes and injection of cRNA were performed as described previously [29] . Oocytes were injected with 46 nl of RNA solution , with RNA coding for α1 , β2 and γ2 subunits at a ratio of 10∶10∶50 nM [30] . The injected oocytes were incubated in modified Barth's solution [90 mM NaCl , 3 mM KCl , 0 . 82 mM MgSO4 , 0 . 41 mM CaCl2 , 0 . 34 mM Ca ( NO3 ) 2 , 100 U/ml penicillin , 100 µg/ml streptomycin and 100 µg/ml kanamycin , 5 mM HEPES pH 7 . 6] at 18°C for approximately 36 h before the measurements to ensure the expression of a functional receptor . Electrophysiological experiments were performed by the two-electrode voltage-clamp method . Measurements were done in ND96 medium containing 96 mM NaCl , 2 mM KCl , 1 mM MgCl2 , 1 . 8 mM CaCl2 and 5 mM HEPES , pH 7 . 5 , at a holding potential of −80 mV . Currents were measured using a custom-made two-electrode voltage clamp amplifier in combination with an XY recorder ( 90% response time , 0 . 1 s ) . The intracellular electrodes were filled with 3 M KCl ( resistance of 0 . 5–1 . 5 MΩ ) . Oocytes were exposed to ND96 or ND96 containing GABA with or without drugs by switching the perfusate using a ValveLink 8 . 2 perfusion system ( AutoMate Scientific , Berkeley , CA ) . The perfusion solution ( 6 ml/min ) was applied through a glass capillary with an inner diameter of 1 . 35 mm . GABA was prepared as a 10 mM stock solution dissolved in ND96 . The modulatory compounds ( IVM and MOX ) were first dissolved in DMSO at 20 mM and then diluted in ND96 to the final concentration . The maximum concentration of DMSO used in perfusion was <0 . 01% , with application of DMSO alone at 0 . 1% not altering GABA responses . Control concentration-response curves for GABA alone were obtained by perfusing the oocytes with a known GABA concentrations in ND96 saline until the maximal response ( Imax ) was observed . To investigate the ability of MLs to potentiate the GABA-evoked current , oocytes were exposed to 2 µM GABA , which was responsible for approximately 10% of the maximal effect of the dose-response curve of GABA alone ( EC10 ) , followed by a 5 min recovery period . Subsequently , oocytes were exposed to a co-application of GABA ( 2 µM ) with increasing ML concentrations . Concentration ranges used were: IVM: 0 . 5 nM–10 µM; MOX: 1 nM–5 µM . Relative current potentiation by MLs was determined as [ ( I MLs+2 µM GABA/I 2 µM GABA alone ) −1]×100 where I 2 µM GABA is the control current evoked by 2 µM GABA , I MLs+2 µM GABA is the current evoked by each drug concentration in co-application with 2 µM GABA , and I ( MLs+2 µM GABA ) Max is the maximal current evoked by co-application of drugs and 2 µM GABA . A washout period of 5 min between each agonist application was introduced , allowing receptors to fully recover from desensitization . The perfusion system was cleaned , after application of MLs , between each experiment by washing with 10% DMSO in ND96 to avoid contamination . Three or four different batches of oocytes were used to collect data for each analysis . All experiments were conducted at least in triplicate . Differences in ML tissue concentrations were analyzed by one-way analysis of variance ( ANOVA ) with a Tukey post-test and are expressed as mean ± standard deviation ( S . D . ) . GABA ( A ) receptor responses were plotted by least squares fit of log ( agonist ) versus response , with variable slope and are displayed as means ± standard errors ( S . E . M . ) ( Prism 2 . 0 , GraphPad Software Inc . , San Diego , CA , USA ) . Half-maximal concentrations ( EC50 ) , slope factors ( Hill coefficients , nH ) and maximal potentiation values for ML-induced activation of current were obtained using the Hill equation . Statistical analysis was performed using unpaired t-tests with Welch's correction to allow for differences in variances ( GraphPad Instat , San Diego , CA , USA ) . Differences were considered statistically significant when p<0 . 05 .
Acute toxicity of IVM and MOX was evaluated in vivo in P-gp-deficient mice . To determine the median lethal dose of these compounds , each drug was administered subcutaneously at increasing doses . The survival time of mice after the drug administration was recorded over two weeks , as well as the number of surviving mice in each group . Percent survival as a function of administered dose was plotted and the LD50 of each compound was determined graphically ( Figure 2 ) . Results show that the lethal dose for IVM was 0 . 46 µmol/kg ( 0 . 40 mg/kg ) , in good agreement with what was described previously [3] . The LD50 of MOX was 2 . 3 µmol/kg ( 1 . 47 mg/kg ) , a 5 times higher molar dose than that of IVM , demonstrating that MOX has much lower in vivo toxicity compared with IVM . The behaviour of Mdr1ab ( −/− ) mice was observed following administration of both IVM and MOX and neurological signs are reported in Table 1 . Upon administration of IVM at 0 . 11 µmol/kg no changes in the behaviour of mice were observed . At 0 . 46 µmol/kg of IVM , hyperactivity was observed just after administration and severe neurotoxic signs were observed 6 h post-administration . In contrast , MOX did not show any toxicity when administered at the same dose rate and rapid breathing was transiently observed 2 h post-administration of 2 µmol/kg of MOX . At dose rates close to the respective LD50 for IVM and MOX ( 0 . 46 and 2 . 3 µmol/kg ) , neurotoxic signs ( lethargy , tremors and/or ataxia ) of the drugs were observed starting at 6 h for IVM and 4 h for MOX ( Table 1 ) and lethal toxicity occurred for these doses between 8 and 12 h . Because in vivo toxicity of IVM is known to be related to its entry into the central nervous system ( CNS ) , we investigated the accumulation of the two MLs in the brain in order to identify whether IVM and MOX have a different propensity to access the CNS . Plasma and brain concentrations of IVM and MOX were initially evaluated 2 and 24 h after a subcutaneous administration of each drug at an equimolar dose rate of 0 . 23 µmol/kg bw in Mdr1ab ( −/− ) mice . To compare the tendency of IVM and MOX to accumulate in the brain , the brain-to-plasma concentration ratios were calculated . Results are presented in Table 2 . As expected and previously described [31] , there was considerable accumulation of the two drugs in the brain tissue when P-gp was deficient; the brain concentration being even higher at 24 h that at 2 h . The brain-to-plasma concentration ratio for IVM was significantly higher compared with that for MOX , whatever the time studied ( 4 . 8±1 . 6 versus 2 . 2±0 . 7 ml/g , at 24 h , p<0 . 01 ) , demonstrating that IVM has a greater ability to accumulate in brain tissue than MOX . Table 2 also shows that the plasma concentration at 24 h was significantly lower for IVM ( 22 . 0±8 . 1 pmol/ml ) than for MOX ( 42 . 8±9 . 3 pmol/ml , p<0 . 001 ) . Interestingly , the IVM concentration in brain was more than 2-fold higher compared with MOX as early as 2 h after treatment ( 43 . 9±15 . 7 versus 18 . 8±2 . 8 pmol/g , p<0 . 05 ) , demonstrating that IVM enters the brain more rapidly than MOX . When the drug concentration in brain was studied 24 h after administration , no significant difference was observed between the two compounds ( 100 . 2±29 . 7 and 93 . 0±28 . 9 pmol/g for IVM and MOX , respectively ) , showing that the overall brain exposure during a 24-h period did not significantly differ between IVM and MOX . Brain uptake of IVM and MOX was then evaluated in Mdr1ab ( −/− ) and in wild-type mice , 24 h after subcutaneous administration of increasing IVM and MOX dose rates . The highest dose rate used for this purpose was sublethal so that the relationship between drug concentration in brain and the in vivo neurotoxicity could be determined . The absolute brain level was plotted against the administered doses and the positive linear correlation between the brain uptake and the administered dose in Mdr1a ( −/− ) ( Figure 3A ) and in wild-type mice ( Figure 3B ) allowed us to calculate the absolute brain level that will be reached following administration of the drug at the LD50 value for each ML . These values were approximately 270 and 830 pmol/g in Mdr1ab ( −/− ) mice , and 210 and 740–1380 pmol/g in wild-type mice for IVM and MOX , respectively ( Table 3 ) . This demonstrated that MOX had to accumulate 3 times higher concentrations in the brain than IVM to provoke neurotoxicity in P-gp-deficient and in wild-type mice . Table 3 shows a positive and significant correlation between brain concentration and plasma concentration for both compounds in the two strains of mice . In Mdr1ab ( −/− ) mice , this linear relationship reveals that there was no drug brain saturation concentration within the doses studied , in accordance with the P-gp deficiency . The slope of the linear relationship , which reliably quantifies blood-brain transport , was 5 . 5 and 2 . 6 ml/g for IVM and MOX , respectively , demonstrating that 24 h after treatment , the brain-to-plasma concentration ratio for IVM was 2-fold higher than that for MOX , whatever the dose studied ( Table 2 ) . In wild-type mice , the brain-to-plasma ratio was considerably lower than in P-gp-deficient mice for both drugs ( 0 . 08 and 0 . 09 ml/g , respectively ) in accordance with the presence of P-gp which limits the drug entrance into the brain . Differences between the toxicity of IVM and MOX could thus be related to a differential interaction of these two compounds with brain GABA receptors . We therefore compared the ability of IVM and MOX to activate GABA ( A ) receptors expressed in Xenopus laevis oocytes .
It is now well recognised that MOX is less toxic than IVM in some invertebrate species , such as dung beetles [34] and the Anopheles mosquito [35] , and in some mammals , such as wild-type mice [20] and in collie dogs sensitive to IVM . Although this information is important in the context of optimizing the use of MLs in humans and animal parasite control , the mechanisms for such differences remain unknown . The differential toxicity of the avermectins and MOX suggest several hypotheses: ( i ) ML compounds are transported differently across the blood–brain barrier ( BBB ) , ( ii ) they accumulate to a different extent in the CNS tissue leading to a different drug concentration arriving at the target and/or ( iii ) they have different effects on vertebrate CNS receptors . We therefore compared the in vivo neurotoxicity of IVM and MOX in mice , their accumulation in the brain tissue and their ability to potentiate the mammalian GABA ( A ) receptor , expressed in Xenopus oocytes . Given that IVM and MOX are differentially transported by P-gp [31] , this study was performed with P-gp-deficient Mdr1ab ( −/− ) mice , in order to remove any P-gp contribution to the entry of the drugs into the brain . In this study , first neurotoxicity signs were observed at lower doses for IVM compared with MOX . Mdr1ab ( −/− ) mice were found to be approximately 5-fold more sensitive to subcutaneous administered IVM than to MOX with an LD50 of 0 . 46 µmol/kg bw for IVM and of 2 . 3 µmol/kg bw for MOX ( p<0 . 01 ) . This result is consistent with previous work reporting an IVM LD50 of 0 . 6–0 . 8 µmol/kg bw in Mdr1a-deficient mice which had been mutated on only the abcb1a gene and had a compensatory increase in expression of the abcb1b gene [3] . Further , our results in the Mdr1ab−/− mice are consistent with observations that MOX did not induce any signs of toxicosis in IVM-sensitive dogs , which are P-gp-deficient [17] , [19] . The first hypothesis to explain this different drug tolerance is that IVM and MOX are differentially transported across the BBB in vivo and levels of accumulation of these MLs differ in the brain [36] . In the present study , the brain-to-plasma concentration ratio following subcutaneous administration of 0 . 23 µmol/kg MLs in P-gp-deficient mice was more than 2-fold higher for IVM than for MOX ( 4 . 8±1 . 6 ml/g and 2 . 2±0 . 7 ml/g at 24 h for IVM and MOX , respectively , p<0 . 01 ) . Moreover , when increasing doses were administered , the slope of the brain tissue concentration versus plasma concentration curve was higher for IVM than for MOX ( Table 3 ) . These results are in accordance with previous data [31] , demonstrating that in the absence of P-gp IVM has a higher penetration rate into the brain tissue than MOX . MOX has a higher lipophilicity than IVM ( logP MOX = 6; logP IVM = 4 . 8 ) , and one would expect a higher affinity of MOX for this tissue . This led to the hypothesis that other drug efflux transporters at the BBB besides P-glycoprotein [37] , could limit the entry of MOX but not that of IVM into brain . Knowing that the defective P-gp in the mutant Mdr1a ( −/− ) mice was associated with increased Abcg2 mRNA , encoding the efflux transporter Bcrp , at the BBB [38] , this hypothesis is in agreement with a recent study where MOX was identified as a BCRP substrate [39] . In most animals , MOX has a longer plasma half-life than IVM [39] , and the pharmacokinetics of MOX have recently been studied in humans [40] , [41] , [42] , [43] . The difference in half-life between IVM and MOX may alter the kinetics of accumulation of MOX relative to IVM in brain and the time of maximal concentration of MOX and IVM in this tissue . Interestingly , we showed in this study that the absolute level of drug accumulation in the brain in Mdr1ab ( −/− ) mice , at the LD50 dose rate was more than 3-fold higher for MOX than for IVM ( 830 versus 270 pmol/g , respectively ) . These concentrations were very similar to those measured in the brain of wild-type mice after administration of their corresponding LD50 . These data demonstrate that neurotoxicity of a ML compound is not strictly related to its ability to enter and accumulate in the brain . Another hypothesis to explain differences between the toxicity of IVM and MOX could thus be related to a differential interaction of these two compounds with GABA receptors in the brain or any tissues where GABA receptors are localized such as the enteric nervous system and sympathetic ganglia . We have shown here on the rat α1β2γ2 GABAergic Cl− channel that both IVM and MOX were able to activate and potentiate the currents elicited by the reference agonist GABA , in a concentration-dependent manner . This potentiating effect was observed when receptor occupancy was low , i . e . when co-applied with GABA at its EC10 ( allosteric effects are commonly assessed at agonist concentrations between the EC10 and EC20 for the agonist ) , demonstrating that IVM and MOX act as allosteric modulators of the mammalian GABA ( A ) receptor . This is supported by the observation that when GABA was co-applied with MLs in sub-saturating concentrations , the amplitude and time course of the elicited current were higher than when GABA was applied alone . However , the rate for dissociation of IVM and MOX from the receptor was much slower than that for GABA ( data not shown ) , indicating allosteric binding site ( s ) for the MLs on the receptor could exist . Indeed , in accordance with our results , IVM was known to interact with GABAergic receptors and was previously shown to potentiate the GABA-elicited currents of chick neuronal GABA-gated chloride channels [29] , and also of a nematode putative GABA ( A ) receptor [44] , [45] . Previous binding studies have led to the conclusion that IVM binds in a two step process , on the GABAergic receptor resulting in activation of the receptor after binding to a high affinity site and blocking it on further binding to a low-affinity site [5] . As far as MOX was concerned , direct evidence for its interaction with mammalian GABA receptor channels has been missing so far . It has only been reported that MOX , like IVM , blocked binding of 4-n-[3H]propyl-4′-ethynylbicycloorthobenzoate to GABA receptors in Drosophila melanogaster [46] . Here , we clearly demonstrate for the first time that MOX was able to potentiate the GABA-activated currents mediated by rat α1β2γ2 GABA ( A ) receptors and the ability to potentiate GABA action is different between IVM and MOX . MOX had a non significant lower EC50 compared with IVM ( p = 0 . 054 ) . Interestingly , it has been demonstrated that both MOX and IVM bind the Cooperia oncophora glutamate-gated chloride channel GluClα3 , expressed in Xenopus oocytes , and the EC50 of the MOX for opening the receptor in the presence of the natural ligand was lower compared with IVM [47] . Of considerable interest was our finding of differences in the shape of the GABA-potentiation curves , revealing differences in the actions of the two compounds . The Hill coefficient , which reflects the steepness of the dose-response plot , was 1 . 52 for IVM and only 0 . 34 for MOX . A Hill coefficient greater than one for IVM suggests positive cooperativity , i . e . , once one molecule is bound to the receptor , the affinity of the receptor for the molecule increases [48] . A Hill coefficient lower than one , found for MOX , suggests a negative cooperativity . Furthermore , IVM caused an almost 2-fold maximum potentiation of the GABA ( A ) receptor compared with MOX . These data clearly indicate that at a sublethal concentration of IVM in brain ( 270 pmol/g corresponding to 0 . 27 µM which is close to the concentration for maximal effect and is approximately 8× higher than the IVM EC50 for the potentiation of the GABA channel ) , IVM would have a greater potentiation of GABA action on this receptor than would MOX at asimilar brain concentration . Therefore , as IVM concentrations increase , it may potentiate the effects of GABA binding and opening the channel to a much greater extent than will MOX at similar elevated concentrations ( Figure 4B ) , with consequences for depolarization of neurons expressing GABA ( A ) receptors and neurotoxicity . We propose that this is the cause of the higher toxicity of IVM when compared with MOX when ML concentrations increase in the brain . The differences seen in receptor activation between IVM and MOX might be related to a difference in the structure of the MLs . Indeed , differences between IVM and MOX in the case of their interactions with mammalian ABC transporters , especially with P-gp , have already been demonstrated [23] . Moreover , a model for the IVM binding site and atomic interactions with amino acids in a C . elegans glutamate-gated chloride channel have recently been proposed [49] . Considering the structural differences between IVM and MOX , i . e . , absence of the disaccharide moiety on the C-13 of the macrocycle , a methoxime moiety at C-23 and an olefinic side chain at C-25 , it has also been postulated that the interaction of MOX with the glutamate-gated chloride channel will be different from that of IVM [50] . It is therefore reasonable to expect that MOX may also interact differently from IVM on GABA-gated chloride channels . In addition to interaction with GABA ( A ) receptors in the brain , it has been shown that IVM potentiates purinergic ( ATP ) P3X ( cationic ) receptors [51] and acetylcholine receptors [52] in the brain . However , potentiating these receptors requires relatively high concentrations of IVM , 3 and 30 µM , respectively , which is considerably higher than the concentrations that markedly potentiated the GABA receptor . Nevertheless , IVM and MOX could exert some of their neurotoxicity via receptors in the brain other than GABA ( A ) receptors and this needs further investigation . Differences in the accumulation of IVM and MOX in the brain , in the role of the P-gp transporter in the BBB and in the interaction of IVM and MOX with GABA ( A ) receptors in the brain may account for differences in neurotoxicity seen in intact and P-gp-deficient animals . These differences in neurotoxicity of IVM and MOX may be important in considering their use in humans . In summary , we have demonstrated in vivo that in the case of P-gp deficiency ( i ) IVM has a higher penetration rate into the brain whatever the dose administered and enters the brain more quickly than MOX and ( ii ) the brain uptake threshold value leading to neurotoxicity is lower for IVM than for MOX . In addition , we have shown for the first time , that in vitro , MOX can interact with the mammalian GABA ( A ) receptor as an allosteric modulator by enhancing the actions of GABA . Our data indicate that MOX at high brain concentrations is less efficient in potentiating GABA-mediated opening of the GABA ( A ) receptor than is IVM . Altogether these data show that MOX has a wider margin of safety than IVM , even when the BBB function is impaired . These observations contribute to understanding ML-induced toxicity and open new perspectives for using MOX in humans . | Ivermectin ( IVM ) is used for onchocerciasis mass drug administration and is important for control of lymphatic filariasis , strongyloidiases and Scarcoptes mange in humans . It is widely used for parasite control in livestock . Moxidectin ( MOX ) is being evaluated against Onchocerca volvulus in humans and is also widely used in veterinary medicine . Both anthelmintics are macrocyclic lactones ( MLs ) that act on ligand-gated chloride channels and share similar spectra of activity . Nevertheless , there are marked differences in their pharmacokinetics , pharmacodynamics and toxicity . Usually , both MLs are remarkably safe drugs . However , there are reports of severe adverse events to IVM , in some humans with high Loa loa burdens , and IVM can be neurotoxic in animals with defects in P-glycoproteins ( P-gp ) in the blood-brain barrier . We have compared the in vivo neurotoxicity of IVM and MOX in P-gp-deficient mice and their accumulation in brain . We also investigated their effects on mammalian GABA receptors . We show that MOX has a wider margin of safety than IVM , even when the blood-brain barrier function is impaired , and that the neurotoxicity in vivo is related to different effects of the drugs on GABA-gated channels . These observations contribute to understanding ML toxicity and open new perspectives for possible MOX use in humans . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [
"biochemistry",
"veterinary",
"science",
"veterinary",
"diseases",
"veterinary",
"parasitology",
"veterinary",
"toxicology",
"pharmacology",
"biology",
"veterinary",
"pharmacology",
"neuroscience",
"toxicokinetics",
"toxicology",
"neurotoxicology"
] | 2012 | Relative Neurotoxicity of Ivermectin and Moxidectin in Mdr1ab (−/−) Mice and Effects on Mammalian GABA(A) Channel Activity |
The midbrain superior colliculus ( SC ) generates a rapid saccadic eye movement to a sensory stimulus by recruiting a population of cells in its topographically organized motor map . Supra-threshold electrical microstimulation in the SC reveals that the site of stimulation produces a normometric saccade vector with little effect of the stimulation parameters . Moreover , electrically evoked saccades ( E-saccades ) have kinematic properties that strongly resemble natural , visual-evoked saccades ( V-saccades ) . These findings support models in which the saccade vector is determined by a center-of-gravity computation of activated neurons , while its trajectory and kinematics arise from downstream feedback circuits in the brainstem . Recent single-unit recordings , however , have indicated that the SC population also specifies instantaneous kinematics . These results support an alternative model , in which the desired saccade trajectory , including its kinematics , follows from instantaneous summation of movement effects of all SC spike trains . But how to reconcile this model with microstimulation results ? Although it is thought that microstimulation activates a large population of SC neurons , the mechanism through which it arises is unknown . We developed a spiking neural network model of the SC , in which microstimulation directly activates a relatively small set of neurons around the electrode tip , which subsequently sets up a large population response through lateral synaptic interactions . We show that through this mechanism the population drives an E-saccade with near-normal kinematics that are largely independent of the stimulation parameters . Only at very low stimulus intensities the network recruits a population with low firing rates , resulting in abnormally slow saccades .
Precisely how individual cells contribute to the saccade is still debated in the literature . Two competing models have been proposed for decoding the SC population: weighted averaging of the cell vector contributions ( [16–18]; Eq 2a ) vs . linear summation ( [3 , 9 , 19]; Eq 2b ) , respectively , which can be formally described as follows: S A V G = ∑ n = 1 N F n M n ∑ n = 1 N F n ( 2a ) S S U M ( t ) = ∑ n = 1 N ∑ k = 1 K n < t δ ( t − τ n , k ) · mn ( 2b ) N is the number of active neurons in the population , Kn < t the number of spikes in the burst of neuron n up to time t , Fn its mean ( or peak ) firing rate , and Mn = ( xn , yn ) is the saccade vector in the motor map encoded at SC site ( un , vn ) ( Eq 1b ) . mn = ζMn is the small , fixed vectorial contribution of cell n in the direction of Mn , for each of its spikes , with ζ a fixed , small scaling constant that depends on the adopted cell density in the map and the population size , and δ ( t − τk , n ) is the k’th spike of neuron n , fired at time τk , n . The vector-averaging scheme of Eq 2a only specifies the amplitude and direction of the saccade vector , and thus puts the motor map of the SC outside the kinematic control loop of its trajectory . It assumes that the nonlinear saccade kinematics are generated by the operation of horizontal and vertical dynamic feedback circuits in the brainstem [16 , 20 , 21] , or cerebellum [22 , 23] . Note also that vector averaging is a nonlinear operation because of the division by the total population activity . In contrast , the linear dynamic ensemble-coding model of Eq 2b encodes the full kinematics of the desired saccade trajectory at the level of the SC motor map through the temporal distribution of spikes by all cells in the population [9 , 19 , 24] . As a result , the instantaneous firing rates of all neurons in the population , usually estimated by their instantaneous spike-density functions , fn ( t ) , together encode the desired vectorial saccadic velocity profile: v S a c c ( t ) = ∑ n = 1 N f n ( t ) · mn with f n ( t ) = ∑ k = 1 S n 1 σ 2 π · e - ( t - t k , n ) 2 2 σ 2 ( 3 ) where Sn is the number of spikes of cell n , with the spikes occurring at times tk , n . The Gaussian acts as a linear kernel that smooths the discrete spike into a continuous function ( e . g . , [25] ) . Although the models of Eqs 2a and 2b cannot both be right , each is supported by different lines of evidence . For example , electrical microstimulation produces fixed-vector ( E- ) saccades with normal main-sequence kinematics that are insensitive to a large range of stimulation parameters [10 , 15 , 26 , 27] . If one supposes that electrical stimulation directly activates a large population of SC cells , and that the firing rates follow the ( typically rectangular ) stimulation profile , a vector-averaging scheme with downstream dynamic feedback circuitry readily explains why E-saccades are normal main-sequence , since the center of gravity of the population specifies the desired saccade vector only , regardless the firing rates . In addition , reversible inactivation of a small part of the SC motor map produces particular deficits in the metrics of visually-evoked ( V- ) saccades that may not be readily explained by the linear summation model of Eq 2b [16] . As the amplitude and direction of a V-saccade to the center of the lesioned site remain unaffected , saccades to locations around that site are directed away from the lesion . For example , V-saccades for sites rostral to the lesion undershoot the target , while V-saccades for sites caudal to the lesion will overshoot the target . The simple vector-summation model of Eq 2b yields saccades that would always undershoot targets , as the lesioned population produces fewer output spikes than under normal control conditions . However , [9 , 19] observed that their estimate of the total number of spikes from the SC population , was remarkably constant , regardless of saccade amplitude , direction , or speed . Yet , they also observed that many cells in the normal SC fire some post-saccadic spikes . They therefore assumed that saccades are actively terminated by a downstream mechanism , whenever the criterion of a fixed number of spikes , NTOT , is reached: ∑ n = 1 N ∑ k = 1 K n δ ( t - τ n , k ≤ N T O T ( 4 ) They demonstrated , by simulating the summation model of Eq 2b with actual recordings from ∼150 cells , that by including the criterion of Eq 4 ( which constitutes a cut-off nonlinearity in the model ) , the pattern of saccadic over- and undershoots to a focal SC lesion can be fully explained . In addition , the extended summation model of Eqs 2b and 4 also accounts for weighted averaging of double-target stimulation in the motor map [10 , 28 , 29] . Moreover , although the vector-averaging model ( Eq 2a ) correctly predicts the pattern of saccadic dysmetrias , it fails to explain the substantial slowing of the lesioned saccades [16] . As this latter observation is also accounted for by Eqs 2b and 4 [9] , it further supports the hypothesis that the SC population encodes both the saccade-vector , and its kinematics [24] . Interestingly , electrical microstimulation experiments have also shown that at low current strengths , just around the threshold , the evoked saccade vectors become smaller and slower than main sequence [15 , 30] . These results do not follow from vector averaging ( Eq 2a , which would always generate the same saccade , but might be predicted by dynamic summation ( Eqs 2b and 4 ) , if low-amplitude electrical stimulation were to recruit a smaller number of neurons at lower firing rates . However , if supra-threshold microstimulation would produce a large square-pulse population profile around the electrode tip ( mimicking the profile of the imposed current pulses , as is typically assumed ) , the summation model would generate severely distorted saccade-velocity profiles , which are not observed in experiments . Yet , little is known about the actual activity profiles in the motor map evoked by electrical microstimulation , as simultaneous multi-electrode recordings in the SC during microstimulation are not available and would be obscured by the large stimulation artefacts [31] . Under microstimulation , two factors contribute to neuronal activation: ( 1 ) direct ( feedforward ) current stimulation of cell bodies and axons by the stimulation pulses of the electrode , and ( 2 ) synaptic activation through lateral ( feedback ) interactions among neurons in the motor map . How each of these factors contributes to the population activity in the SC is unknown . It is conceivable , however , that current strength falls off rapidly with distance from the electrode tip ( at least by ∼1/r2 ) , and that hence a relatively small number of SC neurons would be directly stimulated by the electric field of the electrode . Indeed , a two-photon imaging study , carried out in cortical tissue from rodents and cat are V1 , showed that microstimulation at physiological current strengths directly activates only a sparse set of neurons directly around the immediate vicinity of the stimulation site [32] . These considerations therefore suggest that the major factor in explaining the effects of microstimulation in the SC motor map may be synaptic transmission through lateral excitatory-inhibitory connections among the cells . Such a functional organization in the SC is supported by anatomical studies [33 , 34] , by electrophysiological evidence [35–37] , and by pharmacological studies [38] . We recently constructed a biologically plausible , yet simple , spiking neural network model for ocular gaze-shifts by the SC population to visual targets [39] . This minimalistic ( one-dimensional ) model with lateral interactions can account for the experimentally observed firing properties of saccade-related cells in the gaze-motor map [9 , 19] , by assuming an invariant spiking input pattern from sources upstream from the motor map ( e . g . , FEF ) . We here extended that simple network model to the full two-dimensional network map that accounts for microstimulation results over a wide range of stimulation parameters . To simplify the analysis of the network properties , and to limit the number of independent parameters that describe the electrical stimulation pulses , we used rectangular current profiles with different heights ( current intensities ) and durations . In line with the evidence from previous work , the network was tuned such that microstimulation provides an initial seed that directly activates only a small set of SC neurons , which subsequently sets up a large SC population activity through lateral synaptic interactions . Our results show that stimulating the network indeed sets up a near-normal population activity profile that generates appropriate saccadic command signals across the two-dimensional oculomotor range through the linear dynamic summation mechanism of Eq 2b .
The afferent mapping function ( Eq 1a ) translates a target point in visual space to the anatomical position of the center of the corresponding Gaussian-shaped population in the SC motor map . It follows a log-polar projection of retinal coordinates onto Cartesian collicular coordinates ( Eq 1a; [13] ) . To allow for a simple 2D matrix representation of the map in our network model , we simplified the afferent motor map to the complex logarithm: u ( R ) = B u · ln ( R ) and v ( ϕ ) = B v · ϕ with R = x 2 + y 2 and ϕ = a t a n ( y x ) ( 5 ) with Bu = 1 mm and Bv = 1 mm/rad ( isotropic map ) . Thus , the contribution , m , of a single spike at site ( u , v ) to the eye movement is computed from the efferent mapping function as: m x = ζ exp ( u ) cos ( v ) and m y = ζ exp ( u ) sin ( v ) ( 6 ) We thus constructed a spiking neural network model as a rectangular grid of 201 x 201 neurons . The network represents the gaze motor-map with 0 < u < 5 mm ( i . e . , up to amplitudes of 148 deg ) , and −π/2 < v < π/2 mm . The network generates saccadic motor commands of different directions and amplitudes into the contralateral visual hemispace through a spatial-temporal population activity profile . The location of the population in the motor map determines the direction and amplitude of the saccade target , whereas the temporal activity profile encodes the eye-movement kinematics , through Eq 2b . As described below , and in our previous study [39] , the eye-movement main-sequence kinematics result from location-dependent biophysical properties of the neurons within the map , together with their lateral interconnections . We investigated the dynamics of the network model numerically through simulations developed in C++/CUDA [40] . The motor map is represented as a rectangular grid of neurons with a Mexican hat-type pattern of lateral interactions . The neural activities were simulated by custom code utilizing dynamic parallelism to accelerate spike propagation on a GPU [41] . The code was developed and tested on a Tesla K40 with CUDA Toolkit 7 . 0 , Linux Ubuntu 16 . 04 LTS ( repository under https://bitbucket . org/bkasap/sc_microstimulation ) . Simulations ran with a time resolution of 0 . 01 ms . Brute-force search and genetic algorithms , described below , were used for parameter identification and network tuning since there exists no analytical solution for the system . The neurons in the network were described by the adaptive exponential integrate-and-fire ( AdEx ) neuron model [42] , which accommodates for a variety of bursting dynamics with a minimum set of free parameters . The AdEx model is a conductance-based integrate-and-fire model with an exponential membrane potential dependence . It reduces Hodgkin-Huxley’s model to only two state variables: the membrane potential , V , and an adaptation current , q . The temporal dynamics of the system are given by the following differential equations for neuron n: C d V n d t = - g L ( V n - E L ) + g L η exp ( V n - V T η ) - q n + I i n p , n ( t ) ( 7a ) τ q , n d q n d t = a ( V n - E L ) - q n ( 7b ) where C is the membrane capacitance , gL is the leak conductance , EL is the leak reversal potential , η is a slope factor , VT is the neural spiking threshold , qn is the adaptation time constant , a is the sub-threshold adaptation constant , and Iinp , n is the total synaptic input current . In our previous paper [39] the input-layer of Frontal Eye Field ( FEF ) neurons had identical biophysical properties , and only received a fixed external input current , Iinp , n = Iext . In the present simulations , we did not include a FEF input layer , as the electrical stimulation was applied within the SC motor map as an external current . Two parameters specify the biophysical properties of the SC neurons: the adaptation time constant , τq , n ( which is assumed to be location dependent ) , and the synaptic input current , Iinp , n = Isyn , n + IE ( where Isyn , n is a location- and activity-dependent synaptic current , and IE is the applied microstimulation current ) . Both variables change systematically with the spatial location of the cells within the network ( rostral to causal ) . The remaining parameters , C , gL , EL , η , VT and a , were tuned such that the cells showed neural bursting behavior ( see Table 1 for the list and values of all parameters used in the simulations , and Fig 1 for some example responses ) . The AdEx neuron model employs a smooth spike initiation zone between VT and Vpeak , instead of a strict spiking threshold . Once the membrane potential crosses VT , the exponential term in Eq 7a starts to dominate and the membrane potential can in principle increase without bound . We applied a practical spiking ceiling threshold at Vpeak = −30 mV for the time-driven simulations . For each spiking event at time τ , the membrane potential is reset to its resting potential , Vrst , and the adaptation current , q , is increased by b to implement the spike-triggered adaptation: V ( τ ) → V r s t and q ( τ ) → q ( τ ) + b ( 8 ) After rescaling the equations , the neuron model has four free parameters ( plus the input current ) [43] . Two of these parameters characterize the sub-threshold dynamics: the ratio of time constants , τq/τm ( with the membrane time constant τm = C/gL ) and the ratio of conductances , a/gL ( a can be interpreted as the stationary adaptation conductance ) . Furthermore , the resting potential Vrst and the spike-triggered adaptation parameter b characterize the emerging spiking patterns of the model neurons ( regular/irregular spiking , fast/slow spiking , tonic/phasic bursting , etc . ) . We applied electrical stimulation by the input current , centered around the site at [uE , vE] , according to Eq 5 . We incorporated an exponential spatial decay of the electric field from the tip of the electrode: I E ( u , v , t ) = I 0 · exp ( - λ ( u - u E ) 2 + ( v - v E ) 2 ) · P ( t ) ( 9 ) with λ ( mm−1 ) a spatial decay constant , I0 the current intensity ( in pA ) , and a rectangular stimulation pulse given by P ( t ) = 1 for 0 < t < DS , and 0 elsewhere . Thus , only a small set of neurons around the stimulation site will be directly activated with this input current ( see Results ) . Throughout this paper , we used a fixed input current profile ( I0 = 150 pA ) , λ = 10 mm−1 and DS = 100 ms ) except for the final section , where we explore the effect of changing the microstimulation parameters on the resulting saccade . These parameters were determined by the neural tuning of the AdEx neurons in their bursting regime ( see Neural tuning and bursting mechanism section in Results ) . For simplicity , we incorporated a single rectangular stimulation pulse , P ( t ) , rather than a train of narrowly spaced stimulation pulses . A train of pulses would introduce additional parameters , like pulse height , pulse duration , pulse intervals , pulse polarity , and number of pulses ( stimulus duration ) , each of which would affect the network response . We have shown before that the spiking neural network model with AdEx neurons and lateral interactions can deal with such spiking input patterns [39] . However , varying these different stimulations parameters would complicate the analysis , and is deemed a topic for future work ( see Discussion ) . Note also that the AdEx neurons act as ‘leaky integrators’ for membrane potentials below VT . Therefore , a sequence of pulses and a single rectangular pulse yield qualitatively similar membrane responses . Remark on the current scale . In SC microstimulation experiments , one typically applies extracellular currents in the micro-Ampère range ( 10–50 μA ) to evoke a saccade . In our simulations , we instead take the effective intracellularly applied current , which amounts to only a tiny fraction of the total extracellular current leaving the electrode . The total input current for an SC neuron , n , located at ( un , vn ) , is governed by the spiking activity of surrounding neurons , through conductance-based synapses , and by the externally applied electrical stimulation input ( Eq 9 ) : I i n p . n ( t ) = g n e x c ( t ) ( E e - V n ( t ) ) + g n i n h ( t ) ( E i - V n ( t ) ) + I E ( u n , v n , t ) ( 10 ) where g n e x c and g n i n h are excitatory and inhibitory synaptic conductances acting upon neuron n , Ee and Ei are excitatory and inhibitory reversal potentials respectively . These conductances increase instantaneously for each presynaptic spike by a factor determined by the synaptic strength between neurons , and they decay exponentially otherwise , according to: τ e x p d g n e x c d t = - g n e x c + τ e x c ∑ i N p o p w i , n e x c ∑ s N s p k s i δ ( t - τ i , s ) ( 11a ) τ e x p d g n i n h d t = - g n i n h + τ i n h ∑ i N p o p w i , n i n h ∑ s N s p k s i δ ( t - τ i , s ) ( 11b ) with τexc and τinh , the excitatory and inhibitory time constants; w i , n e x c and w i , n i n h are the intracollicular excitatory and inhibitory lateral connection strengths between neuron i and n , respectively ( Eqs 12a and 12b ) and τi , s is the spike timing of the presynaptic SC neurons that project to neuron n . With conductance-based synaptic connections , spike propagation occurs in a biologically realistic way , since the postsynaptic projection of a presynaptic spike depends on the instantaneous membrane potential of the postsynaptic neuron . In this way , the state of a neuron determines its susceptibility to presynaptic spikes . We incorporated a Mexican hat-type lateral connection scheme in the model , where the net synaptic effect is given by the difference between two Gaussians [44] . Accordingly , neurons were connected with strong short-range excitatory and weak long-range inhibitory synapses , which implements a dynamic soft winner-take-all ( WTA ) mechanism: not only one neuron remains active , but the “winner” affects the temporal activity patterns of the other active neurons . The central neuron governs the population activity , since it is the most active one in the recruited population . As a result , all recruited neurons exhibit similarly-shaped bursting profiles as the central neuron , leading to synchronization of the spike trains within the population [39] . Two Gaussians describe the excitatory and inhibitory connection strengths between collicular neurons as function of their spatial separation: w i , n e x c = s n · w ¯ e x c exp ( - | | u i - u n | | 2 2 σ e x c 2 ) ( 12a ) w i , n i n h = s n · w ¯ i n h exp ( - | | u i - u n | | 2 2 σ i n h 2 ) ( 12b ) where w ¯ e x c > w ¯ i n h and σ ¯ i n h > σ ¯ e x c , and sn is a location-dependent synaptic weight-scaling parameter , which accounts for the location-dependent change in sensitivity of the neurons due to the variation in adaptation time constants . Electrophysiological experiments have indicated that the neural responses are well characterized by four principles: ( i ) a fixed number of spikes for each neuron associated with its preferred saccade vector Nu , v ≅ 20 spikes , ( ii ) a systematic dependence of the neuron’s cumulative spike count on the saccade vector ( dynamic movement field ) , Nu , v ( R , ϕ , t ) , ( iii ) scaled and synchronized burst profiles of the neurons in the population , resulting in a high cross-correlation , Cpop ( fn ( t ) , fm ( t ) ) ≈ δnm , between the firing rates of recruited neurons , and ( iv ) a systematic decrease of the peak firing rate of central neurons in the population , Fpeak , along the rostral-caudal axis , together with an increase of burst duration , Tburst , and burst skewness , Sburst . [19] argued that these properties follow from a systematic tuning of the gaze-motor map , and that they are responsible for the observed saccade kinematics . Here we applied these principles to determine a similarity measure between our simulated responses , and the experimentally recorded gaze motor-map features . In our network model , these features emerge from the interplay between intrinsic biophysical properties of the SC neurons , and the lateral interactions between them . Eye movements were generated by the population activity following the linear ensemble-coding model of Eqs 2b and 3 . We applied the two-dimensional efferent motor map of Eq 5 . For any network configuration throughout this paper , the unique scaling factor of the efferent motor map ( ζ ) was calibrated for a horizontal saccade at ( x , y ) = ( 21 , 0 ) deg . The resulting eye-displacement vector , S → ( t ) , was calculated from the spike trains by interpolation with a first-order spline to obtain equidistant time samples . The interpolated data were further smoothed with a Savitzky-Golay filter , to obtain smooth velocity profiles .
Fig 1 shows the membrane potential traces for three model neurons , differing in their adaptation time constants , τq , which were stimulated under different microstimulation paradigms . The electrical stimulus strength increased from a low amplitude ( I0 = 50 pA; light blue traces ) to a high intensity ( I0 = 250 pA , dark-blue traces ) , for stimulation durations between 25 and 225 ms . Note that for these different microstimulation regimes , the burst onsets and burst shapes ( i . e . , the instantaneous firing rates ) could differ , even when the number of elicited spikes would be the same . These responses illustrate how the biophysical properties of the neurons affected their bursting behavior . First , the neuron could respond after the stimulation had terminated . Such a feature , as well as the bursting behavior , is only captured by more complex spiking neuron models . Even when the input current amplitude cannot drive a neuron rapidly to its first spike to initialize the burst ( light traces ) , it suffices if the neuron’s membrane potential crosses a certain threshold ( VT in the AdEx neuron ) . The neuron can then elicit a spike after the stimulation is over ( visible for stimulation durations < 75 ms ) . Second , the stimulation amplitude determines the response onset: as the amplitude increases , the first spike occurs earlier . Such a behavior is to be expected , since the neuron model acts as an integrator [30]; higher input currents thus drive a neuron faster to its spiking threshold . Third , the different neurons respond differently to long stimulation trains ( > 175 ms ) . While the neuron with a longer adaptation time constant ( τq = 84 . 6 ms; Fig 1A ) responds with repetitive bursts of 4 to 5 spikes , separated by a silent period , the faster recovering neuron ( τq = 52 . 4 ms; Fig 1C ) elicits more and more spikes after the initial burst , especially for the higher current amplitudes ( dark traces ) . Interestingly , the neurons with the intermediate ( Fig 1B ) and short ( Fig 1C ) adaptation time constants switch between different bursting behaviors as the current amplitude increases along with longer stimulation durations . Regular short bursts with silent periods in between result from the slow decay of the adaptation current , which acts on the membrane potential as an inhibitory current . Hence , the adaptation time constant determines how fast a neuron will recover after each spike in a burst . Therefore , the strongly adapting neuron with a long will require more input current to elicit another spike ( Fig 1A and 1B for stimulation duration >175 ms ) , and thus after the fourth spike in the burst , the adaptation current is already high enough to break the bursting cycle . The fast recovering neuron ( Fig 1C , short τq ) continues its burst with more spikes ( dark traces at longer durations ( B , C ) . A phase plot of the instantaneous adaptation current vs . the membrane potential provides a graphical analysis of the effects of changing the neural parameters , the current input , and the initial state , on the evolution of the dynamical system . Fig 2 shows a number of phase-trajectories for the Adex model , for the parameters used in the simulations of the SC motor map . Nullclines illustrate the boundaries of the vector fields in the AdEx neuron’s phase plane . The V-nullcline ( Vnull; i . e . , dV/dt = 0 for Eq 7a ) and the q-nullcline ( qnull; i . e . , dq/dt = 0 for Eq 7b ) are shown as gray lines . Fixed points of the system lie at the intersections of these nullclines . A stable fixed point of the system is found at [-53 mV , 0 nA] . In all subfigures that is the starting point of the trajectories , and the state variables of the neurons will converge to this stable fixed point in the absence of input . The q-nullcline follows a linear trajectory , whereas the V-nullcline represents a convex function because of the superposition of two V-dependent parts . For V < VT , the exponential term can be omitted and the linear V dependence will have a slope of gL . For V > VT , the exponential term will dominate with a sharp increase as V increases . When a neuron receives input , the V-nullcline shifts upward by as much as the current density , and the response of the neuron follows a trajectory on the phase plane toward the spiking threshold . The blue trajectories show the evolution of the state variables for three neurons with different τq values , and stimulated at different current strengths . The horizontal arrows show the membrane potential in the spike initiation zone , V > VT . Spikes occur when the membrane potential overcomes the spiking threshold , V > Vthr . After a spike , the membrane potential is reset , and the adaptation current is increased by b ( Eq 7 ) . The spiking threshold , Vthr , and the reset potential , Vrst , are indicated by the vertical dashed lines . With each spike , the adaptive current increases more and once it reaches values above the V-nullcline , the adaptive current is high enough to suppress the neuron from continued bursting , and hyperpolarizes . In Fig 2A , the phase trajectory crosses values over Vnull = 150 pA after 5 spikes . Due to the hyperpolarization , the membrane potential starts to drop . The phase plot shows that the microstimulation is finished when the membrane potential decreases to -58 mV , and the smooth trajectory is seen disrupted . In Fig 2B , there is a second burst cycle since the microstimulation duration is much longer . After the first burst cycle crosses Vnull + 200 pA with 6 spikes ( arrows are placed closer to Vthr ) , neuron follows the trajectory to the spike initiation zone for a second burst cycle with 5 spikes . The end of the microstimulation coincides with the second burst cycle and afterwards the membrane potential decreases fast due to the high adaptive current acting on the neuron . In Fig 2C , the neuron gets stuck in its first cycle and continues spiking repetitively . This pattern is due to the fast decay of the adaptive current , which drops by more than b after each spike . Therefore , the neuron would continue spiking repetitively , as long as the current is applied . The neurons in the network were tuned to respond with a fixed number of spikes in a burst cycle ( as in Fig 2A ) . This initial burst sets up a large population activity through the lateral connections . Vnull fluctuates for each neuron with the network dynamics , depending on the input from other neurons in the population . Microstimulation parameters were chosen such that the central neuron of the population would respond with a burst cycle of 4-5 spikes ( typically , DS = 100 ms , and I0 = 150 pA ) , independent of the biophysical properties of the neuron . To that end , the adaptation time constant , τq , n , and the synaptic weight-scaling parameter , sn , for each neuron were determined by applying a fifth order polynomial fit to produce a fixed number of spikes ( N = 20 ) for self-exciting neurons: s n = ( 8 . 808 · 10 - 9 · τ q , n 5 - 3 . 280 · 10 - 6 · τ q , n 4 + 4 . 855 · 10 - 4 · τ q , n 3 - 3 . 607 · 10 - 2 · τ q , n 2 + 1 . 383 · τ q , n - 8 . 396 ) · 10 - 3 ( 15 ) The self-excitation mimics the population activity , since the central cell’s burst profile is representative for the entire population activity , due to burst synchronization across the active neurons . The adaptive time constant , τq . n , varied from 100-30 ms in a linear way with the anatomical rostral-caudal location of the neurons , according to: τ q , n = 100 - 14 * u n with u n ∈ [ 0 , 5 ] mm ( 16 ) The current density drops rapidly with distance from the microelectrode tip , as given by the current spread function ( Eq 9 , with λ = 10 mm−1 , DS = 100 ms , and I0 = 150 pA ) . Fig 3A illustrates this decay of current density on the motor map surface . The pulsed input current is presented onto the collicular surface at a site corresponding to the visual image point ( u ( R ) , v ( ϕ ) in Eq 5; Fig 3B and 3C ) . Microstimulation directly activated only a small set of neurons within a 250 μm radius . Fig 3B and 3C shows the number of spikes elicited by the activated neurons in the absence of intra-collicular lateral interactions . Each activated neuron elicited only 4-6 spikes within a given input duration range , regardless the electrode’s location . These spikes arose from the initial bursting regime of the neurons until the adaptation current built up with repetitive spikes that canceled the microstimulation input ( see Fig 2 ) . The input amplitude affected the response delay of the neurons between stimulation onset and their first spike . Thus , in the model these small neuronal subsets generated only a brief pulse signal that is supposed to set up the entire population activity through lateral connections . We next tested the collicular network response to the same microstimulation parameters as in Fig 3 , while including the lateral interactions . Fig 4A–4C shows the recruited neural population at the rostral stimulation site . Clearly , the number of recruited neurons had increased substantially as a result of the network dynamics . The diameter of the circular population extended to about 1 mm in the motor map . In addition , the cumulative activity elicited by the central cells had now increased from about 5 to 20 spikes . Fig 4B shows the neuronal bursts ( top spike patterns ) from a number of selected cells in the population , together with the associated spike-density functions . The peak firing rate of the central cells was close to 700 spikes/s and dropped in a regular fashion with distance from the population center . Note also that the cells near the fringes of the population were recruited slightly later than the central cells , but that their peak firing rates were reached nearly simultaneously . Moreover , the bursts all appeared to have the same shape . Fig 4C shows the saccade that was elicited by this neural population , together with its velocity profile . The saccade had an amplitude of 5 deg , reaching a peak velocity of about 200 deg/s . Fig 4D–4F shows the results for stimulation at the more caudal location in the motor map , yielding an oblique saccade with an amplitude of 31 deg . The size of the resulting population activity is very similar to that of the rostral population , and also the number of spikes elicited by the cells is the same . The peak firing rates of the neurons , however , were markedly lower , reaching a maximum of about 450 spikes/s . As a result , the burst durations increased accordingly , from about 50 ms at the rostral site , to more than 70 ms at the caudal site . Note that the saccade reached a much higher peak velocity ( about 900 deg/s ) than the smaller saccade in Fig 4C , but its duration was prolonged . Note also that the horizontal and vertical velocity profiles were scaled versions , indicating a straight saccade trajectory . In Fig 5 we quantified the collicular bursts in response to microstimulation at different sites along the rostral-caudal axis in the motor map . Fig 5A shows how the evoked collicular bursts of the central cells in the population systematically reduce their peak firing rates , and increase their duration , as the microelectrode moves from rostral ( R = 2 deg ) to caudal sites ( R = 31 deg ) . In Fig 5B we show three major relationships for the bursts of the central cells in the population , for saccade amplitudes between 2 and 65 deg: the peak firing rate ( green ) drops from about 750 spikes/s to 300 spikes/s , burst duration ( purple ) increases from about 40 ms to 125 ms , whereas the number of spikes in the burst ( light green ) remains constant at N = 20 spikes . These burst properties , which are due to a precise tuning of the biophysical cell parameters , underlie the kinematic main-sequence properties of saccadic eye movements [19 , 39 , 45] . Fig 6A shows the amplitudes and directions of 45 elicited saccades across the 2D oculomotor range ( stimulation parameters: I0 = 120 pA , DS = 100 ms ) . We avoided stimulating near the vertical meridian , as our model included only the left SC motor map ( e . g . , [15] ) , and stimulation at very caudal sites ( R > 40 deg ) , where edge effects of the finite motor map would lead to truncation of the elicited population at the caudal end . Crosses indicate the coordinates of the corresponding motor map locations where stimulation took place; blue dots give the coordinates of the evoked saccade vectors . There is a close correspondence between the motor map coordinates and the elicited saccade vectors . Only for the most caudal sites the saccade vectors tended to show a slight undershoot . We have not attempted to compensate for these minor effects , e . g . by including heuristic changes to the efferent mapping function . The panels of Fig 6B and 6C show the evoked saccades for the nine stimulation sites along the horizontal meridian . Note that the saccade duration increased with the saccade amplitude , and that the peak eye velocity showed a less than linear increase with saccade size . Fig 7 presents three examples of saccade position and velocity traces for stimulation at sites encoding three different directions , but with a fixed amplitude of R = 21 deg . The elicited track-velocity profiles are direction-independent . Panels Fig 7B and 7C also indicate the behavior of the horizontal and vertical saccade components . As these are precisely synchronized with the saccade vector , the ensuing saccade trajectories are straight ( not shown ) . The main-sequence behavior of the model’s E-saccades is quantified in Fig 8 . Fig 8A shows the nonlinear amplitude vs . peak eye-velocity relationship , described by the following saturating exponential function: v p e a k = 1172 · ( 1 - exp ( - 0 . 04 · R ) ) deg/s ( 17 ) From Fig 8B , the straight-line amplitude-duration relation was approximated to D s a c c = 28 . 7 + 1 . 1 · R ms ( 18 ) These main-sequence relations were combined into a single , characteristic linear relationship that captures all saccades , normal and slow ( Fig 8C ) by: v p e a k · D s a c c = 1 . 72 · R deg ( 19 ) All three relations correspond well to the normal main-sequence properties , as have been reported for monkey and human saccades ( e . g . , [2] ) . Importantly , the main-sequence behavior of E-saccades was largely insensitive to the applied current strength as soon as it exceeded the stimulation threshold . This feature of the model is illustrated in Fig 9 , which shows E-saccade peak eye-velocity as function of current strength for a fixed stimulation duration of DS = 100 ms ( Fig 9A ) . The stimulation was applied at three different sites on the horizontal meridian ( corresponding to R = 15 , 21 and 31 deg ) . Below I0 = 80 pA no movement was elicited , but around the threshold , between 90-120 pA , stimulation evoked slow eye movements , which eventually yielded the final amplitude ( Fig 9B ) . Immediately above the threshold at 130-140 pA , the evoked movement amplitudes and velocities reached their final , site-specific size ( Fig 9A and 9B ) , which did not change with current strength over the full range between 140-220 pA . The associated peak eye velocity followed a similar current-dependent behavior for changes in stimulus duration ( at a fixed current strength of 150 pA; Fig 9C ) . Thus , the quantity that determines evoked saccade initiation is the total amount of current ( current amplitude times duration; e . g . , [30] ) .
The simple linear ensemble-coding model of Eq 2b [9 , 45 , 46] seems inconsistent with the results of microstimulation , when it is assumed that ( i ) the rectangular stimulation input profile directly dictates the firing patterns of the neural population in the motor map , and ( ii ) that the neurons are independent , without synaptic interactions . We here argued that these assumptions are neither supported by experimental observation , nor do they incorporate the possibility that a major factor determining the recruitment of SC neurons is caused by synaptic transmission within the motor map , rather than by direct activation through the electrode’s electric field . We implemented circular-symmetric , Mexican-hat like interactions in a spiking neural network model of the SC motor map and assumed that the current profile from the electrode rapidly decreased with distance from the electrode tip ( Fig 3A ) . As a consequence , only neurons in the direct vicinity of the electrode were activated by the external electric field ( Fig 3B and 3C; [31 , 32] ) . Once neurons were recruited by the stimulation pulse , however , local excitatory synaptic transmission among nearby cells rapidly spread the activation to create a neural activity pattern which , within 10-15 ms , was dictated by the bursting dynamics of the most active central cells in the population ( Fig 4 ) . As a result , all cells yielded their peak firing rates at the same time , and the burst shapes of the cells within the population were highly correlated . Similar response features have been reported for natural , sensory-evoked saccadic eye movements [19] , and it was argued this high level of neuronal synchronization ensures an optimally strong input to the brainstem saccadic burst generator to accelerate the eye with the maximally possible innervation . Note that the evoked population activity does not grow without limit , but ceases automatically , both in its spatial extent , and in its bursting behavior , while the inhibitory currents acting on the neurons accumulate during the stimulation pulse . These currents are due to the synaptic far-range lateral inhibition , and to each neuron’s own adaptive current . Thus , once the network is perturbed by an excitatory input current , the SC will set up a bursting population activity , without the need of an external comparator , or external feedback by a resettable integrator . Indeed , the adaptive current functionally acts as a putative ‘spike counter’ at the single neuron level . With this spiking neural network model , we thus offer an alternative framework for the oculomotor system , in which the SC motor map not only provides a spatial signal for the saccade vector , but also the instantaneous eye-movement kinematics , through the temporal organization of its burst profiles . The site-dependent tuning of the biophysical parameters of the AdEx neurons , in particular their adaptive time constants and lateral-interaction weightings specified by Eqs 15 and 16 , caused the peak firing rates of the cells to drop systematically along the rostral-to-caudal axis , while keeping the total number of spikes constant ( Fig 5 ) . As a result , the saccade kinematics followed the nonlinear main-sequence properties that are observed for normal ( visually-evoked ) saccadic eye movements ( Figs 6–8 ) . In addition , the long-range weak inhibition ensured that the size of the population remained fixed to about 1 . 0 mm in diameter , and resulted to be largely independent of the applied current strength and the current-pulse duration ( Fig 9 ) . The lateral excitatory-inhibitory synaptic interactions ensured three important aspects of collicular firing patterns that underlie the saccade trajectories and their kinematics: ( i ) they set up a large , but limited , population of cells in which the total activity ( quantified by the number of spikes elicited by the recruited cells ) can be described by a circular-symmetric Gaussian with a width ( standard deviation ) of approximately 0 . 5 mm ( Fig 4A and 4D ) , ( ii ) the temporal firing patterns of the central cells ( their peak firing rate , burst shape , and burst duration ) solely depend on the location in the motor map ( Eq 14 ) , but the number of evoked spikes remains invariant across the map , and for a wide range of electrical stimulation parameters ( Fig 5 ) , and ( iii ) already within the first couple of spikes , the recruited neurons all became synchronized throughout the population , in which the most active cells ( those in the center ) determined the spike-density profiles of all the others ( Fig 4B and 4E ) . Here we described the consequences of this model on the ensuing kinematics and metrics of E-saccades as function of the electrical stimulation parameters . We showed that the network could be tuned such that stimulation at an intensity of 150 pA and a total input current duration of DS = 100 ms , sets up a large population of activated neurons , in which the firing rates resembled the activity patterns as measured under natural visual stimulation conditions . As a result , the kinematics of the evoked saccades faithfully followed the nonlinear main-sequence relations of normal , visually evoked saccades ( Fig 8 ) . Importantly , above threshold the saccade properties were unaffected by the electrical stimulation parameters ( Fig 9 ) . Only close to the stimulation threshold , the evoked activity remained much lower than for supra-threshold stimulation currents , leading to excessively slow eye movements , that started at a longer latency with respect to stimulation onset . Similar results have been demonstrated in microstimulation experiments ( e . g . [15 , 30] . The saccade peak eye velocity of the model saccades followed a psychometric curve as function of the amount of applied current ( Fig 9 ) . We found that the kinematics of the evoked eye movements at near-threshold microstimulation were much slower than main sequence ( Fig 9 ) . Although this property is readily predicted by the linear summation model ( Eq 2b ) , it does not follow from center-of-gravity computational schemes ( like Eq 2a ) , in which the activity patterns themselves are immaterial for the evoked saccade kinematics . Conceptually , the lateral interactions serve to normalize the population activity . Therefore , the total number of spikes emanating from the SC population remains invariant across the motor map , and to a large range of ( sensory or electrical ) stimulation parameters at any given site . The nonlinear saturation criterion of Eq 4 is thus automatically implemented through the intrinsic organization of the SC network dynamics , and do not seem to require an additional downstream ‘spike-counting’ mechanism in order to terminate the saccade response , e . g . during synchronous double stimulation at different collicular sites ( see , e . g . [28] ) . Although other network architectures , relying e . g . on presynaptic inhibition across the dendritic tree , have been proposed to accomplish normalization of the population activity and vector averaging [28 , 45 , 47–49] , substantial anatomical evidence in the oculomotor system to support such nonlinear mechanisms is lacking . We here showed , however , that simple linear summation of the effective synaptic inputs at the cell’s membrane , which is a well-recognized physiological mechanism of basic neuronal functioning , can implement the normalization when it is combined with excitatory-inhibitory communication among the neurons within the same , topographically organized structure . Such a simple mechanism could suffice to ensure ( nearly ) invariant gaze-motor commands across a wide range of competing neuronal inputs . Our model predicts near-normal activity profiles within the SC during microstimulation ( Figs 4–6 ) , and hence near-normal recruitment of the downstream brainstem circuits . Although simultaneous recordings in the SC during microstimulation are lacking , [50] described recordings from neural populations in the downstream brainstem burst generators ( EBNs ) and omnipause neurons ( OPNs ) during SC microstimulation . Their results indicated normal discharge patterns for OPNs and EBNs , and indistinguishable movement kinematics for stimulation-evoked and volitional saccades [51] . These results are nicely in line with the predictions or our model ( Figs 8 and 9 ) , at least for suprathreshold stimulation levels [26] . The two-dimensional extension of our model is a substantial improvement over our earlier one-dimensional spiking neural network model [39] . It can account for a much wider variety of neurophysiological phenomena . Yet , we have not attempted to mimic every experimental result of microstimulation . A few aspects in our model have not been incorporated yet , or some of its results seem to deviate slightly from experimental observations , which we briefly summarize here . First , although the network output is invariant across a wide variety of stimulation parameters , and evoked saccade kinematics drop markedly around the threshold ( Fig 9 ) , the present model did not produce small-amplitude , slow movements near the stimulation threshold . This behavior has sometimes been observed for near-threshold stimulation intensities [15 , 30] . In our model , the saccade amplitude behaved as an all-or-nothing phenomenon ( Fig 9B ) , which is caused by the strong intrinsic mechanisms that keep the number of spikes of the central cells fixed . Although we have not tested different parameter sets at length , we conjecture that a major factor that is lacking in the current model is the presence of intrinsic noise in the parameters and neuronal dynamics that would allow some variability of the evoked responses for small inputs . When near the threshold the elicited number of spikes starts to fluctuate , and becomes less than the cell’s maximum , the evoked saccades will become smaller ( and slower ) too . Such near-threshold responses would also explain the truncated saccades generated when stimulation train durations are shortened [26] . Second , although the main-sequence relations of the model’s E-saccades ( Eqs 17 and 19 ) faithfully capture the major kinematic properties of normal eye movements , the shape of the evoked saccade velocity profiles were not as skewed as seen for visually-evoked saccades . As a result , the peak velocity is not reached at a fixed acceleration period , but at a moment that slightly increased with the evoked saccade amplitude ( Fig 6C ) . We have not attempted to remediate this slight discrepancy , which in part depends on the applied spike-density kernels ( here: Gaussian , with width σ = 8 ms , Eq 3 ) , and in part on the biophysical tuning parameters of the AdEx neurons . However , it should also be noted that a detailed quantification of E-saccade velocity profiles , beyond the regular main-sequence parametrizations [15 , 30] , is not available in the published literature . It is therefore not known to what extent E-saccade velocity profiles and V-saccade velocity profiles are really the same or might slightly differ in particular details . Third , as explained in Methods , the electrical stimulation inputs were described by simple rectangular pulses , rather than by a train of short-duration stimulation spikes , in which case also the pulse intervals , pulse durations , pulse heights , and the stimulation frequency would all play a role in the evoked E-saccades [26 , 30] . We deemed exploring the potential results corresponding to these different current patterns as falling beyond the scope of this study , which merely concentrated on the proof-of-principle that large changes in the input for the proposed architecture of a spiking neural network led to largely invariant results . Note , however , that in our previous paper [39] the presumed input from FEF cells to the SC motor map did indeed provide individual spike trains to affect the SC-cells . We there demonstrated that the optimal network parameters could be found with the same genetic algorithm for such spiky input patterns , as applied here ( Eq 13 ) . The small differences in neuronal tuning parameters for the 1D model with FEF input , compared to the 2D model tuned to electrical pulse input , are mostly due to these fundamentally different input dynamics . Fourth , [14] recently reported an asymmetric , anisotropic representation in the afferent mapping for the upper vs . lower visual hemi-fields , that would explain kinematic differences between upward vs . downward saccades . The underlying mechanism for this anisotropy is not yet clear . For example , it could result from ( i ) differences in lateral interaction strengths for up vs . down , thus creating different population profiles in the SC; ( ii ) differences in cell density along the medial-lateral SC coordinate , or ( iii ) systematic differences in the efferent projection strengths from medial-lateral SC neurons to the up- and down burst generators . In principle , our model could accommodate an anisotropic organization for upward vs . downward saccades by incorporating parametric changes at any of these levels . Here , we focused on a simple scheme , in which the SC was taken fully isotropic ( Eqs 5 and 6 ) , and the horizontal/vertical burst-generating circuits in the brainstem , including the horizontal/vertical ocular plants , were taken identical [9] . This ensured perfectly straight saccade trajectories in all directions , with homogeneous main-sequence properties , due to a full cross-coupling between the horizontal and vertical movement components ( ‘component stretching’; see Fig 7 ) . Any change in this organization ( e . g . more realistic eye-position related differences in the oculomotor plants , or different gains and delays in the up- vs . down vs . horizontal burst generators ) will cause saccade trajectories to become curved , and direction and eye-position dependent , and may be made to resemble more closely the idiosyncratic differences observed in measured oblique saccades ( e . g . [5] ) . Although an interesting topic , working out these many different factors , however , falls beyond the scope of this paper . Fifth , double-stimulation experiments at different sites within the SC motor map have shown that the resulting saccade vector appears a weighted average between the saccades evoked at the individual sites [10 , 27] . In the present paper , we have not implemented double stimulation , although an earlier study had indicated that Mexican-hat connectivity profiles in the motor map effectively embed the necessary competition between sites to result in effective weighted averaging [28] . In a follow-up study , we recently explored the spatial-temporal dynamics of our model to double stimulation at different sites , and at different stimulus strengths [52] . Indeed , double stimulation results in weighted-averaged saccade responses , even when the SC activity is decoded by a dynamic linear-ensemble coding scheme , and without the need to implement an explicit cut-off on the total spike count , like in Eq 4 . Thus , our SC scheme with excitatory-inhibitory interactions results to automatically normalize the total activity within the SC motor map ( see also above ) . Hence , double stimulation results do not support the vector averaging scheme per se , as they can be explained by linear summation , in combination with intracollicular interactions , as well . Finally , close inspection of the burst profiles in Fig 1 ( showing stimulation results for single , isolated neurons ) suggests that prolonged stimulation at sufficient current intensities could in principle generate multiple bursts of activity in the SC cells . For example , the top-left trace ( I0 = 250 pA , DS = 225 ms ) shows a burst of 6 spikes , followed by a second burst of 5 spikes about 150 ms later . In principle , each of these bursts could be part of its own saccade , provided that the total network dynamics ( including the lateral interactions ) would preserve these properties . Indeed , the literature has shown that prolonged stimulation can lead to a series of eye movements of decreasing amplitude in the same direction ( a so-called ‘staircase’ of saccades; [10 , 50 , 51] ) . Here we haven’t tested our network for its potential to generate staircases , as we limited the stimulation durations to 250 ms . We suspect that the inhibitory currents and neural recovery may have to be balanced better to allow the prolonged input current to overcome the dynamic inhibition . Yet , although our network was not a priori designed for these staircases , their occurrence would be an interesting emerging property of the model . | The midbrain Superior Colliculus ( SC ) is crucial for generating rapid saccadic eye movements . It contains a topographically organized map of visuomotor space , in which a large population of recruited cells determines the metrics and kinematics of saccades . The dynamic spike-counting model explains how this population encodes the ensuing eye movement through linear dynamic summation of the spike-effects of each recruited neuron . Electrical microstimulation in the motor map produces saccades with a vector that corresponds with the location of the electrode in the map , and with very similar kinematics as normal visually-evoked saccades . Although the summation model accounts for the kinematics of visually-evoked saccades , it could , so far , not be reconciled with the effects of microstimulation . Here we modeled the SC motor map with a spiking neural network , in which cells are connected through tuned local excitatory and global inhibitory synapses . The network was tuned such that stimulation directly recruits only a small subset of neurons , from which activity rapidly spreads across the motor map to set up a ( near- ) normal population . Simulations with this computational model show that this scheme explains the metrics and kinematics of electrically evoked saccades . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [] | 2019 | Microstimulation in a spiking neural network model of the midbrain superior colliculus |
The classical method for detection of Lassa virus-specific antibodies is the immunofluorescence assay ( IFA ) using virus-infected cells as antigen . However , IFA requires laboratories of biosafety level 4 for assay production and an experienced investigator to interpret the fluorescence signals . Therefore , we aimed to establish and evaluate enzyme-linked immunosorbent assays ( ELISA ) using recombinant Lassa virus nucleoprotein ( NP ) as antigen . The IgM ELISA is based on capturing IgM antibodies using anti-IgM , and the IgG ELISA is based on capturing IgG antibody–antigen complexes using rheumatoid factor or Fc gamma receptor CD32a . Analytical and clinical evaluation was performed with 880 sera from Lassa fever endemic ( Nigeria ) and non-endemic ( Ghana and Germany ) areas . Using the IFA as reference method , we observed 91 . 5–94 . 3% analytical accuracy of the ELISAs in detecting Lassa virus-specific antibodies . Evaluation of the ELISAs for diagnosis of Lassa fever on admission to hospital in an endemic area revealed a clinical sensitivity for the stand-alone IgM ELISA of 31% ( 95% CI 25–37 ) and for combined IgM/IgG detection of 26% ( 95% CI 21–32 ) compared to RT-PCR . The specificity of IgM and IgG ELISA was estimated at 96% ( 95% CI 93–98 ) and 100% ( 95% CI 99–100 ) , respectively , in non-Lassa fever patients from non-endemic areas . In patients who seroconverted during follow-up , Lassa virus-specific IgM and IgG developed simultaneously rather than sequentially . Consistent with this finding , isolated IgM reactivity , i . e . IgM in the absence of IgG , had no diagnostic value . The ELISAs are not equivalent to RT-PCR for early diagnosis of Lassa fever; however , they are of value in diagnosing patients at later stage . The IgG ELISA may be useful for epidemiological studies and clinical trials due its high specificity , and the higher throughput rate and easier operation compared to IFA .
Lassa fever is a viral hemorrhagic fever endemic in West Africa [1–14] . The causative agent is Lassa virus , an RNA virus of the family Arenaviridae . Its main natural host is the rodent Mastomys natalensis [15 , 16] . Contamination of households and food by rodent excreta is a likely mode of virus transmission to humans . The virus may be further transmitted from human to human and cause epidemics mainly in the nosocomial setting [2 , 3 , 5 , 17] . However , the Lassa virus-specific seroprevalence in endemic areas is high , indicating that most Lassa virus infections in the communities are probably mild [4] . Clinically , Lassa fever mainly presents with flu-like and gastrointestinal symptoms and is difficult to distinguish from other febrile illnesses seen in West African hospitals [7 , 18 , 19] . Therefore , diagnosis requires laboratory confirmation . Reverse transcription polymerase chain reaction ( RT-PCR ) and antigen detection are valuable tools for early diagnosis of Lassa fever [12 , 20–26] . IgM and IgG antibodies are detectable in a fraction of patients during the first days of illness . While patients with fatal Lassa fever do not necessarily develop antibodies [27 , 28] , they are commonly detectable in survivors [20–22 , 29] . The classical method for detection of Lassa virus-specific antibodies is immunofluorescence assay ( IFA ) using virus-infected cells as antigen [27 , 30] . However , this method requires laboratories of biosafety level 4 for virus propagation and an experienced investigator to interpret the fluorescence signals . The aim of this study was the establishment and evaluation of IgM and IgG enzyme-linked immunosorbent assays ( ELISA ) based on recombinant Lassa virus nucleoprotein ( NP ) . We have chosen NP , as previous work provided evidence that ELISA methods detecting the NP-specific antibody response compare favorably with the conventional methods for the serological diagnosis of Lassa fever [31 , 32] . Our IgM assay is based on capturing IgM antibodies with human anti-μ chain antibodies on the solid phase ( IgM μ-capture ELISA ) . The test principle of our IgG assay is based on capturing IgG antibody–antigen complexes via rheumatoid factor ( RF ) , a human anti-IgG autoantibody , or via human Fc gamma receptor molecule CD32a on the solid phase ( IgG RF ELISA and IgG CD32 ELISA ) . Both RF and CD32 preferentially bind antibodies in complex with antigen , which facilitates sensitive detection of specific IgG in the capture assay [33–36] .
Ethical permission was granted by the Research and Ethics Committee of Irrua Specialist Teaching Hospital ( ISTH ) , Edo State , Nigeria ( ISTHREC09/14/12/2009 ) , the Ärztekammer Hamburg , Germany ( PV3187 ) , and the Institutional Review Board of the Noguchi Memorial Institute for Medical Research , Ghana ( NMIMR-IRB 003/07-08 ) . Written informed consent was obtained from all study subjects . If study participants were minors , the parents/guardians provided consent on behalf of all child participants . In addition , leftover specimens , i . e . remnants of specimens collected for routine clinical care that would otherwise have been discarded , were included in the assay evaluation . These specimens were anonymized and used without informed consent . A total of 576 sera from the diagnostic service of the Institute of Lassa Fever Research and Control at ISTH , which is located in a Lassa fever endemic zone in Edo State , Nigeria , were chosen for the study; 394 sera were left over from the first specimen of patients with suspected Lassa fever that was sent to the laboratory for Lassa virus diagnostics between 2008 and 2011 ( also called diagnostic specimen ) [10] . Of those , 270 sera tested positive by Lassa virus RT-PCR [25] establishing the diagnosis of Lassa fever; 101 sera tested negative by Lassa virus RT-PCR; and 23 had no RT-PCR result . From 47 RT-PCR confirmed Lassa fever patients , 182 ( 1–9 per patient ) follow-up sera were available . Three follow-up specimens were excluded from further analysis , as the data obtained with these specimens were implausible within the context of the data obtained with other follow-up specimens of the respective patients ( S1 Data ) , suggesting mix-up of samples during sampling or processing . The diagnostic specimen was missing for one follow-up patient . From Lassa fever non-endemic areas , 199 samples collected between 2008 and 2011 from patients with suspected viral hemorrhagic fever or viral hepatitis in Ghana , all of whom tested negative by Lassa virus RT-PCR [37] , and 105 diagnostic leftover samples from German patients with various unknown diseases were included in the study . The travel history of the patients from non-endemic areas was not known . The specimens for this study were randomly chosen from the available specimen pools . However , to provide a more meaningful collection for the purpose of this study , patients who tested positive in RT-PCR were intentionally overrepresented . Thus , the prevalence of Lassa fever patients among the study population does not reflect the true prevalence of Lassa fever patients among all patients tested at ISTH . All samples were analyzed retrospectively and therefore , the information generated in this study was not known to those who performed the RT-PCR assays . The investigator who performed IFA and ELISA knew the origin of the samples during evaluation , though not the RT-PCR result . IFA and ELISA were performed in a blinded fashion , i . e . the investigator did not know the corresponding assay result . Samples from each setting ( Nigeria , Ghana , Germany ) were tested consecutively according to the identification number . Diagnosis ( Lassa fever or non-Lassa fever ) , IFA result , and ELISA result were linked after testing . Presence and absence of disease were defined as follows: Presence of Lassa fever was defined by a positive result in the Lassa virus RT-PCR on the diagnostic specimen . RT-PCR was chosen as reference standard , as it is the method of choice for early diagnosis of Lassa fever [20 , 21] . Details on the RT-PCR assay used in our study have been published previously [10 , 25] . Absence of Lassa fever was primarily defined by origin of the patient from a non-endemic area . As sporadic Lassa fever cases in Ghana cannot definitively be excluded , classification as "non-Lassa fever" in Ghana included a negative RT-PCR result . Due to the definitive absence of Lassa fever and the virus reservoir in Germany , patients from this country were defined as "non-Lassa fever" without further laboratory testing . RT-PCR-negative patients from Nigeria were not classified as "non-Lassa fever" , as a negative RT-PCR result does not rule out Lassa fever in the endemic area , for example if the patients present at late stage or have a mild course of disease . Because of this ambiguity and because the high prevalence of pre-existing Lassa virus IgG antibodies in these patients would hamper estimation of assay specificity , they were not included in the analysis of clinical accuracy . In summary , origin of patient from a non-endemic country in combination with negative RT-PCR was chosen as reference to define absence of disease , as this deemed more accurate in ruling out Lassa virus infection than a negative RT-PCR result in a patient from endemic area . All Nigerian samples were used to estimate analytical performance using IFA as reference method . IFA was chosen as analytical reference , as it is the classical test for detection of Lassa virus-specific antibodies in patient sera and the diagnostic standard test in our laboratory [27 , 30] . We aimed at a sample size of approximately 300 patients per group , as this number facilitates detection of small proportions at reasonable precision , for example a 2%-false positive rate with specified limits of the 95% confidence interval at 0 . 5% and 3 . 5% ( http://sampsize . sourceforge . net/iface/index . html#prev ) . Information on compliance of this study with standards for reporting diagnostic accuracy studies is given in S1 Checklist and S1 Diagram . The data generated in this study are listed in S1 Data . Lassa virus strain AV [38] was propagated in Vero cells in the biosafety level 4 facility , Hamburg , Germany . Infected cells were spread onto immunofluorescence slides , air-dried , and acetone-fixed . Serum samples were diluted 1:20 and 1:80 and incubated for 2 h with the cells . After washing with phosphate buffered saline ( PBS ) , bound antibodies were detected by anti-human IgG or IgM labeled with fluorescein isothiocyanate ( Dianova ) . Signals were evaluated by fluorescence microscopy and classified as "clearly negative" , "probable positive" , and "clearly positive" by the investigator . For consistency , a single investigator evaluated all slides . Spodoptera frugiperda ( Sf9 ) cells ( Invitrogen ) in several T75 cell culture flasks were inoculated with a high-titered stock of recombinant baculovirus expressing NP of Lassa virus strain AV ( GenBank no . AAG41803 ) fused to C-terminal FLAG-6xHis sequence [38 , 39] . Six days post infection , cells were lysed in 50 mM NaH2PO4 ( pH 8 ) –500 mM NaCl–10 mM imidazole–0 . 5% NP-40–1x Complete Protease Inhibitor Cocktail ( Roche ) –25 U/ml Benzonase ( Novagen ) and treated by sonication . Cell debris was removed by centrifugation and the supernatant was incubated with Ni-NTA agarose ( Invitrogen ) overnight at 4°C . The agarose was washed 5 times with 50 mM NaH2PO4 ( pH 8 ) –500 mM NaCl–30 mM imidazole and NP was eluted in 50 mM NaH2PO4 ( pH 8 ) –500 mM NaCl–500 mM imidazole . Purity of the protein was verified by polyacrylamide gel electrophoresis . NP was dialyzed overnight against PBS , quantified by Bradford assay , and about 0 . 4 mg were biotinylated using EZ-Link Sulfo-NHS-Biotin reagent ( Thermo Fisher ) at 20-fold molar excess on ice for 2 h according to the manufacturer’s instructions . The protein was desalted using Zeba Spin Desalting Columns ( 7K MWCO , 2 ml; Thermo Fisher ) and stored 1:1 in glycerol at –20°C . For use as antigen in ELISA , biotinylated NP ( 100–300 μg/ml ) was diluted 1:200–1:800 in 1% Triton X-100–1% bovine serum albumin–PBS . Optimal dilutions for each lot of antigen were determined empirically . ELISA plates ( Nunc Thermo Scientific 469949 , Immuno Clear Standard Modules , F8 , MaxiSorp ) were coated with recombinant human Fc gamma RIIA/CD32a ( R&D Systems 1330-CD-050/CF ) at a concentration of 8 μg/ml in PBS–0 . 01% sodium azide for 3–5 days at 4°C and washed three times with 0 . 05% Tween 20–PBS before use ( CD32-coated IgG ELISA plates ) . ELISA plates coated with rheumatoid factor ( 4 μg/well ) , an autoantibody prepared from blood of patients with rheumatoid arthritis that is directed against the Fc portion of IgG , were custom-made by Medac Company , Hamburg , Germany ( RF-coated IgG ELISA plates ) . ELISA plates coated with anti-human IgM antibodies were also obtained from Medac ( μ-capture IgM ELISA plates ) . For detection of NP-specific IgG , serum diluted 1:20 in 1% Triton X-100–PBS was mixed 1:1 with biotinylated NP antigen , and 50 μl thereof were incubated in a well of a CD32- or RF-coated ELISA plate overnight at 4°C . For detection of NP-specific IgM , 50 μl serum diluted 1:20 in 1% Triton X-100–PBS was incubated in a well of a μ-capture ELISA plate for 2 h at 4°C; then the wells were washed three times with 0 . 05% Tween 20–PBS and incubated with 50 μl biotinylated NP antigen overnight at 4°C . Following incubation with serum and antigen , ELISA plates were washed three times with 0 . 05% Tween 20–PBS and incubated with 50 μl streptavidin–horseradish peroxidase conjugate ( 1:1 , 000–1:2 , 000 in PBS; Sigma-Aldrich ) per well for 1 h at 4°C . Wells were washed three times with 0 . 05% Tween 20–PBS and incubated with 50 μl 3 , 3' , 5 , 5'-tetramethylbenzidine substrate for 10 minutes . The reaction was stopped by adding 50 μl 0 . 5 M H2SO4 and the optical density ( OD ) was measured in a 96-well plate reader at 450/630 nm . A variety of arbitrary formulas exist for calculation of ELISA cut-off , most of which are based on negative controls [40] . Our cut-off determination considered the following aspects: ( i ) The cut-off is calculated using negative controls , which can be easier generated and replenished compared to positive control sera with a specific titer . ( ii ) The formula does not include the standard deviation of negative controls , as reliable estimation of this parameter requires testing of a larger number of negative controls on each ELISA plate . ( iii ) To account for differences in the end-point of the colorimetric reaction on individual ELISA plates , the cut-off is proportional to the mean OD of negative controls ( scale factor a ) . ( iv ) A constant ( b ) serves as baseline cut-off . According to these criteria , we used the formula: Cut-off = a × Mean OD of negative standards + b , which is a modification of the formulas F1 and F2 in Lardeux et al . [40] . Three representative negative sera were selected as standards for cut-off determination and included in all assays performed in this study . After all data had been collected , we found empirically that a = 3 and b = 0 . 06 facilitate a good correlation between ELISA and IFA results independent of the colorimetric end-point of a plate , in particular for sera classified as "clearly negative" or "clearly positive" in IFA . Except for outliers in the IgM ELISA , sera from non-endemic countries were negative according to this formula . Cut-off values ranged from 0 . 094 to 0 . 457 reflecting the plate-specific colorimetric end-points . The cut-off and sample OD values for individual ELISA plates are shown in S1 Fig . Each assay also included two positive control sera . The reactivity of a serum sample was expressed as sample to cut-off ratio ( S/CO ) or log10 S/CO . Samples with S/C > 1 or log10 S/CO > 0 , respectively , were considered positive . Analytical performance characteristics of the ELISAs were calculated as follows: Sensitivity=NpositivebothbyIFAandELISA/NpositivebyIFA Specificity=NnegativebothbyIFAandELISA/NnegativebyIFA PositivePredictiveValue ( PPV ( =NpositivebothbyIFAandELISA/NpositivebyELISA NegativePredictiveValue ( NPV ( =NnegativebothbyIFAandELISA/NnegativebyELISA Accuracy= ( NpositivebothbyIFAandELISA+NnegativebothbyIFAandELISA ( /Nalltested The term positive may also represent a specific serological constellation arising from the combined use of IgM and IgG assays , i . e . IgM+/IgG+ , IgM+/IgG– , or IgM–/IgG+ . Clinical performance characteristics of the ELISA for diagnosis of acute Lassa fever , such as sensitivity , specificity , positive likelihood ratio , and negative likelihood ratio including 95% confidence intervals ( 95% CI ) were estimated using the statistical calculator at https://www . medcalc . org/calc/diagnostic_test . php . Calculation of clinical PPV and NPV considered the prevalence of Lassa fever patients among all patients with suspected Lassa fever tested ( LF Prevalence ) : PPV=Sensitivity×LFPrevalence/[Sensitivity×LFPrevalence+ ( 1–Specificity ( × ( 1–LFPrevalence ) ] NPV=Specificity× ( 1−LFPrevalence ( /[ ( 1–Sensitivity ) ×LFPrevalence+Specificity× ( 1–LFPrevalence ) ] Sensitivity=NLassafeverpatientspositivebyELISA/NallLassafeverpatients Specificity=Nnon-LassafeverpatientsnegativebyELISA/Nallnon-Lassafeverpatients The specificity for combined use of IgM and IgG ELISA was calculated as: Specificity of combined test = 1 –False Positive Rate of IgM ELISA × False Positive Rate of IgG ELISA . Pre-existing Lassa virus-specific IgG among patients with suspected Lassa fever ( IgG Prevalence ) was considered in the false positive rate of the IgG ELISA , as IgG from past infection occurs exclusively in non-Lassa fever patients ( there is no evidence for secondary Lassa fever ) : False Positive Rate of IgG ELISA = IgG Prevalence / ( 1 –LF Prevalence ) . To simplify calculation , we assumed a linear 1:3 relationship between prevalence of Lassa fever patients and IgG seroprevalence: IgG Prevalence = 3 × LF Prevalence .
Both IgG and IgM ELISA were designed as indirect capture assays as described previously [32–36] . The IgM ELISA plates were coated with anti-human IgM for capturing IgM antibodies and the IgG ELISA plates were coated with RF or human Fc gamma receptor CD32a for capturing specific IgG antibody–antigen complexes . Lassa virus NP was expressed in insect cells and used as antigen . NP was labeled with biotin for detection of antibody–antigen complexes via streptavidin–peroxidase conjugate . The concentrations of the ELISA components were adjusted so that negative control sera yielded OD values between 0 . 02 and 0 . 08 and positive control sera yielded OD values between 1 . 0 and 2 . 5 . Cut-off values typically ranged from 0 . 15–0 . 3 . All experiments were performed with the μ-capture IgM ELISA , the RF-based IgG ELISA , and the CD32-based IgG ELISA . The sample to cut-off ( S/CO ) values obtained for the 880 sera included in this study are summarized in Fig 1 . A comparison of the S/CO values of the RF-based vs . the CD32-based IgG ELISA revealed a high degree of congruence between both read-outs ( Fig 2 ) , indicating that technically both assays are comparable . The analytical performance of the ELISAs to detect Lassa virus-specific antibodies was first evaluated in comparison to IFA as serological reference method . The test panel included 576 sera from a hospital in a Lassa fever endemic area in Nigeria . All sera were tested by IgM and IgG IFA and classified as "clearly negative" , "probable positive" , and "clearly positive" . The S/CO values obtained with IgM , RF IgG , and CD32 IgG ELISA correlated well with the IFA categories "clearly negative" and "clearly positive" resulting in ≥95% correspondence of ELISA and IFA results ( Fig 1 ) . As expected , the S/CO range of sera classified as "probable positive" in IFA was broader , though for the IgG ELISAs the S/CO ratios in this category tended to be >1 ( Fig 1 ) . For calculation of analytical performance characteristics of the ELISAs , the "probable" and "clearly positive" IFA categories were merged into one "positive" category . The analytical specificity of the ELISAs ranged from 95–98% for all assay combinations ( Table 1 ) . The analytical sensitivity of the individual IgM and IgG ELISA was 84% and 90% , respectively , indicating good agreement with IFA in detecting positive samples . When IgM and IgG assays were used in combination , the analytical sensitivity depended on the serological constellation ( Table 1 ) . There was poor agreement of the ELISAs with IFA in detecting the IgM+/IgG–status ( analytical sensitivity 12 . 5% ) , while the IgM+/IgG+ status was detected with a sensitivity of 87% . In summary , there was good analytical performance of the ELISAs in detecting Lassa virus-specific IgM and IgG antibodies or antibody combinations , except for the IgM+/IgG–status . The clinical evaluation of the ELISAs was performed using sera from 270 Lassa fever patients . About one third of these patients was IgM-positive in the first—diagnostic—specimen , i . e . the specimens that had been used to establish the diagnosis by Lassa virus RT-PCR ( Table 2 ) . The majority ( 83% ) of IgM positives was also positive for IgG and only a minority ( 17% ) had isolated IgM . Follow-up sera were available for 47 Lassa fever patients . The percentage of IgM positive as well as IgM/IgG positive sera among these follow-up sera was about 60% , while isolated IgM was less prevalent than in the first specimen ( Table 2 ) . The low prevalence of patients with isolated IgM compared to patients showing IgM as well as IgG suggested that both antibody classes develop concurrently in the majority of Lassa fever patients . This conclusion was further substantiated in 21 patients , who seroconverted during follow-up . In 12/21 ( 57% ) IgM and IgG emerged at the same time ( Fig 3 , patients B , D , E , I , K , L , O , P , R , S , T , U ) . In 7/12 ( 33% ) , IgM seroconversion was delayed or absent relative to IgG seroconversion ( Fig 3 , patients A , C , H , J , M , N , Q ) . Only in 2/21 ( 9 . 5% ) ( Fig 3 , patients F and G ) , IgM emerged shortly before IgG . There was no significant difference between ELISA and IFA regarding the time point when seroconversion was detected . However , IgG IFA results around seroconversion often scored "probable positive" , while the IgG ELISAs already showed clear positive results with S/CO ratios >>1 ( Fig 3 ) . Nearly 20% of Lassa virus RT-PCR-negative patients from the endemic area in Nigeria—pre-classified as inconclusive Lassa fever status—showed isolated IgG reflecting the seroprevalence in the area ( Table 2 ) . Five percent were found positive in IgM ELISA and most of those were also positive in IgG ELISA ( Table 2 ) . These findings were confirmed by IFA , suggesting that the IgM-positive patients might have been in convalescence phase or had Lassa fever recently . Sera from non-Lassa fever patients provided information on the specificity of the ELISA . All non-Lassa fever patients from Ghana ( Lassa virus RT-PCR negative ) and Germany were negative in IgG ELISA ( Table 2 ) . The patients from Germany were also negative in IgM ELISA , while 13 sera from Ghana showed a positive signal in IgM ELISA ( Table 2 ) . However , the latter sera tested negative for IgM in IFA and showed a positive signal in ELISA even in absence of NP antigen , indicating that the signals result from unspecific binding . Therefore , these samples were classified as IgM false positives . Based on the data presented in Table 2 , we estimated the performance characteristics of the ELISAs for diagnosis of acute Lassa fever in the first specimen taken from a patient ( Table 3 ) . The characteristics were calculated for the IgM ELISA as stand-alone test and in combination with the IgG ELISA . Due to the high seroprevalence in endemic areas as described here and elsewhere [4 , 7] , the IgG ELISA was considered for diagnostics of acute patients only in combination with the IgM ELISA . The sensitivity of the IgM ELISA was 31% irrespective of the IgG result; 5 . 2% for isolated IgM detection ( IgM+/IgG– ) ; and 26% for detection of both IgM and IgG ( IgM+/IgG+ ) . The specificity of the IgM ELISA was estimated to be 96% based on the frequency of false positive reactions among non-Lassa fever patients from non-endemic areas . The specificity of the IgG ELISAs was estimated to be 100% , as there was no evidence for false positive reactions among patients from non-endemic areas . The likelihood ratios indicate good diagnostic value for the stand-alone IgM ELISA and combined IgM and IgG detection ( IgM+/IgG+ ) . However , isolated IgM detection ( IgM+/IgG– ) has no diagnostic value , as the likelihood ratios are close to 1 and their 95% CI overlap with 1 ( Table 3 ) . PPV and NPV were calculated for a prevalence of Lassa fever patients among all patients tested ranging from 0% to 20% ( Fig 4 ) . The calculation for combined IgM and IgG detection also considered the impact of pre-existing IgG in the local population ( IgG seroprevalence ) on the diagnosis of acute Lassa fever . The PPV of a stand-alone IgM finding approximates zero at low prevalence and improves with increasing prevalence of Lassa fever cases , while combined detection of IgM and IgG shows a PPV of about 70% over the entire prevalence range . The NPV of the IgM assay alone and in combination with the IgG assay is 100% at low prevalence and decreases to 85% at high prevalence . With increasing prevalence of pre-existing IgG , specificity and positive likelihood ratio for combined IgM/IgG detection decrease from 100% to 97% and from ∞ to 10 , respectively ( Fig 4 , right ) .
This study describes the establishment and evaluation of novel capture ELISA for detection of Lassa virus NP-specific IgM and IgG . The ELISAs detected Lassa virus-specific antibodies with high analytical accuracy . Evaluation of the ELISAs for diagnosis of acute Lassa fever revealed sensitivity for the stand-alone IgM ELISA of 31% and for combined IgM/IgG detection of 26% . Isolated IgM reactivity , i . e . IgM in the absence of IgG , has no diagnostic value . The clinical specificity of IgM and IgG ELISA was estimated at 96% and 100% , respectively . In agreement with previous serological studies [20–22 , 27–29] , the ability of the ELISAs to diagnose Lassa fever at early stage is limited . The sensitivity of our stand-alone IgM ELISA is comparable to that of a previously published IgM ELISA , which detected 26% of virus culture-confirmed Lassa fever cases in the first specimen and 72% in all blood draws [22] . However , the sensitivity of combined IgM/IgG detection was lower than in our study . Three main factors affected the performance of our assays . First , as most Lassa fever patients have not ( yet ) developed antibodies on admission to the hospital , the clinical sensitivity of IgM detection is low . Second , due to high IgG seroprevalence in endemic regions , the detection of IgG is not suitable for diagnosis of acute infection but improves performance in combination with the IgM assay , as discussed below . And third , we found evidence for false positive reactions in the IgM ELISA . Specificity issues are common to IgM detection assays and may stem from various conditions , including polyclonal B cell activation , vigorous immune response e . g . during acute malaria , or naturally occurring biotin IgM antibodies [41] . An observation that actually improves clinical performance of the ELISA was the rare occurrence of isolated IgM during early stage of disease . The majority of IgM positive patients already developed IgG . This is consistent with our follow-up data showing that Lassa virus-specific IgM and IgG develop simultaneously rather than sequentially . In some patients , IgG emerged even slightly earlier than IgM . The immunological mechanisms behind the rapid class switch from IgM to IgG or suppression of the IgM response are not known [42 , 43] . On the other hand , this observation might be due to a higher analytical sensitivity of the IgG ELISA compared to the IgM ELISA . However , due to this specific feature , specificity , positive likelihood ratio , and PPV of the diagnostic method can be improved by combining IgM and IgG detection . The PPV of the stand-alone IgM assay , i . e . without considering IgG , strongly depends on the prevalence of Lassa fever among the hospital admissions . At low prevalence a stand-alone IgM finding is likely to be false positive . The additional information on the IgG status facilitates classification of IgM positive findings into two groups: ( 1 ) isolated IgM reactivity that is of no diagnostic value , and ( 2 ) IgM/IgG reactivity with a positive likelihood ratio >10 and a PPV around 70% independent of Lassa fever prevalence . This is in line with the improvement of the analytical performance ( i . e . concurrence of the ELISA with IFA results ) for detection of both IgM and IgG versus stand-alone or isolated IgM detection . In summary , due to the low sensitivity ( 26% ) of IgM/IgG detection on admission , the ELISAs may play only a supplementary role in diagnostics of Lassa fever in the early stage . Even IgM/IgG-positive findings have to be interpreted with caution due to the moderate PPV ( 70% ) of such finding . Importantly , our data confirm that the method of choice for early detection of Lassa fever is RT-PCR or sensitive antigen detection [12 , 20–26] . A firm diagnosis of Lassa fever may be established by ELISA in patients in whom IgM and/or IgG seroconversion can be demonstrated . However , the window period before seroconversion delays diagnosis , and thus proper patient management , and fatal cases do not necessarily develop antibodies [27 , 28] . The latter suggests that development of antibodies is a marker for a favorable prognosis , which might be of value in clinical management . The ELISAs may also have diagnostic value in Lassa fever patients presenting during convalescence after virus had been cleared , in patients with symptomatic virus persistence in CSF or other body fluids in the absence of viremia [44] , or in patients with mild or asymptomatic infections [4] , in whom the viremia might be below the detection limit of the PCR . Unfortunately , due to the lack of appropriate specimens , we cannot provide estimates for diagnostic accuracy of the ELISAs to detect these clinical conditions . Another potential limitation of our study is that we have evaluated the assays in one endemic and two non-endemic areas only . The diagnostic accuracy of the assays might be somewhat different in other settings . Possible factors affecting performance include the average time between onset of Lassa fever and presentation at the hospital , the prevalence of Lassa fever , the ratio between clinical and subclinical Lassa virus infections , the prevalence of Lassa virus-specific antibodies in the population , and the frequency of cross-reactive antibodies or unspecific reactions . In order to generalize our data and provide estimates of performance in various settings , we have taken some of these factors into account for estimation of specificity , likelihood ratio , PPV and NPV . Nevertheless , before applying the ELISA in Lassa fever diagnostics it is essential that the performance characteristics be estimated—or even validated—according to the local conditions . Another factor influencing performance might be virus variability . The NP antigen used in the assays is derived from strain AV belonging to Lassa lineage IV circulating in Mali , Côte d’Ivoire , Guinea , Sierra Leone , and Liberia , while we evaluated the assay in an area where a heterologous lineage ( II ) is prevailing [6 , 9 , 10 , 14 , 38] . Therefore , we assume that the diagnostic accuracy in regions where other heterologous lineages circulate will essentially correspond to what we estimated here with lineage II-specific sera . In addition to diagnostics in acute patients , the assays may be of value in epidemiological studies and clinical trials , e . g . to distinguish between natural immunity and immunity induced by glycoprotein-based vaccines [45] . The high specificity of the IgG ELISA , the clear distinction of negatives and positives in that assay via the S/CO ratio , and the higher throughput rate and easier operation compared to IFA will be of advantage in any kind of large-scale seroepidemiological study . The established assays do not require expensive equipment; ELISA readers are available in many diagnostic laboratories in West Africa . The use of recombinant antigen will facilitate future assay production according to standards of good manufacturing practice and potentially commercialization . We are making efforts to provide the assays described here in industry-standard quality in future . | Lassa fever is endemic in several West African countries . However , only few hospitals and laboratories in the region have the capacity to conduct molecular or serological Lassa fever diagnostics . One reason is that the classical serological technique for Lassa fever—the immunofluorescence assay ( IFA ) —requires biosafety level 4 laboratories , which are not available in the Lassa fever endemic countries . In addition , IFA does not feature an objective read-out . Therefore , we established enzyme-linked immunosorbent assays ( ELISA ) for detection of Lassa virus-specific IgM and IgG in 96-well format , which do not require expensive equipment and can be implemented in diagnostic laboratories in West Africa . The ELISAs are based on recombinant antigen facilitating future production according to industry standards . In our evaluation , the ELISAs have shown a performance comparable to IFA . They allow in particular the diagnosis of Lassa fever patients at later stages of the acute illness . In addition , reliable serological assays , such as those described here , are urgently needed to conduct large-scale epidemiological investigations to better understand the epidemiology of Lassa fever across West Africa as well as for clinical trials evaluating novel medical countermeasures including vaccines and drugs . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"reverse",
"transcriptase-polymerase",
"chain",
"reaction",
"medicine",
"and",
"health",
"sciences",
"enzyme-linked",
"immunoassays",
"immune",
"physiology",
"pathology",
"and",
"laboratory",
"medicine",
"pathogens",
"immunology",
"tropical",
"diseases",
"microbiology",
"bi... | 2018 | Development and evaluation of antibody-capture immunoassays for detection of Lassa virus nucleoprotein-specific immunoglobulin M and G |
Rabies is a devastating yet preventable disease that causes around 59 , 000 human deaths annually . Almost all human rabies cases are caused by bites from rabies-infected dogs . A large proportion of these cases occur in Sub Saharan Africa ( SSA ) . Annual vaccination of at least 70% of the dog population is recommended by the World Health Organisation in order to eliminate rabies . However , achieving such high vaccination coverage has proven challenging , especially in low resource settings . Despite being logistically and economically more feasible than door-to-door approaches , static point ( SP ) vaccination campaigns often suffer from low attendance and therefore result in low vaccination coverage . Here , we investigated the barriers to attendance at SP offering free rabies vaccinations for dogs in Blantyre , Malawi . We analysed data for 22 , 924 dogs from a city-wide vaccination campaign in combination with GIS and household questionnaire data using multivariable logistic regression and distance estimation techniques . We found that distance plays a crucial role in SP attendance ( i . e . for every km closer the odds of attending a SP point are 3 . 3 times higher ) and that very few people are willing to travel more than 1 . 5 km to bring their dog for vaccination . Additionally , we found that dogs from areas with higher proportions of people living in poverty are more likely to be presented for vaccination ( ORs 1 . 58-2 . 22 ) . Furthermore , puppies ( OR 0 . 26 ) , pregnant or lactating female dogs ( OR 0 . 60 ) are less likely to be presented for vaccination . Owners also reported that they did not attend an SP because they were not aware of the campaign ( 27% ) or they could not handle their dog ( 19% ) . Our findings will inform the design of future rabies vaccination programmes in SSA which may lead to improved vaccination coverage achieved by SP alone .
Rabies has been estimated to cause around 59 , 000 human deaths per year [1] . Globally , rabies has been estimated to cause 3 . 7 million disability-adjusted life years and 8 . 6 billion US dollars economic losses annually [1] . Almost all human rabies cases are acquired from contact with rabies infected dogs [2] . Case fatality for patients who develop clinical signs related to rabies infection approaches 100% and successful treatment has rarely been reported [2] . Rabies disproportionately affects Sub Saharan African countries [1 , 2] . Despite significant regional and international healthcare intervention initiatives , no African country has been reported rabies free to date [3] . Since 99% of all human rabies deaths are caused by bites from rabies infected dogs [2] , mass dog vaccination campaigns are the single most effective strategy to eliminate rabies amongst humans and dogs [1 , 4 , 5] . To effectively eliminate rabies from canine and human populations , a critical requirement of mass dog vaccination programmes is to ensure that a sufficiently high proportion of dogs are vaccinated [6] . Empirical data has shown that annual vaccination coverage of 70% is sufficient to eliminate rabies from dog and human populations [6 , 7] . This has been further validated by mathematical modelling [8] . For example , mathematical models have demonstrated that a cut-off of 70% would prevent a major disease outbreak at least 96 . 5% of the time based on rabies field data from USA , Mexico , Malaysia and Indonesia [8] . Collectively , these findings have resulted in the recommendation by the World Health Organisation ( WHO ) that rabies vaccination programmes should vaccinate at least 70% of all dogs annually [6 , 7 , 9 , 10] . However , vaccinating large numbers of dogs at over 70% coverage has proved challenging despite the development of a range of mass rabies vaccination strategies [11] . Vaccination approaches which have been used include door-to-door campaigns ( D2D ) ; static point ( SP ) campaigns , using both fixed and temporary posts; and a combination of the two . Door-to-door programmes , which typically achieve a high vaccination coverage , are labour intensive , expensive and challenging to roll out on a large scale . Consequently , most rabies vaccination programmes in Sub Saharan Africa ( SSA ) have used SP vaccination approaches where the vaccination teams remain at a static location within a community and the local inhabitants present dogs to the vaccination teams . Although widely used in Africa as they are logistically and economically more feasible than door-to-door approaches , SP approaches have often failed to reach a high coverage [9 , 11–14] . Consequently , many organisations have to manage a trade off when rolling out dog vaccination programmes of either utilising a door-to-door approach , which typically results in high coverage but lower numbers of dogs vaccinated , or a SP approach , which often achieves a lower coverage but facilitates the vaccination of more dogs . This trade-off between coverage and number of dogs vaccinated would be eliminated if a higher proportion of dogs could be vaccinated through SPs . The reasons why attendance at SPs is low in many countries has surprisingly received little attention despite it being a major reason why rabies elimination programmes have been so challenging to effectively roll out in SSA . If the barriers to attendance at SPs could be understood and then overcome , mass vaccination programmes could become more feasible thereby allowing high vaccination coverage to be achieved without the need for expensive and logistically challenging door-to-door programmes . Only a small number of studies have explored why attendance to SPs is often suboptimal . In previous small scale studies in Chad , Mali , Peru and urban Tanzania the most common reasons reported by dog owners for not attending a static vaccination point included; lack of information about the campaign [14–18] , difficulty in handling dogs [13–15 , 17 , 18] , lack of time [14 , 17 , 18] , lack of information about rabies [15] , mistrust [15] , distance/location of SP [13 , 15 , 18] , the dog being too young [16–18] or lactating [17 , 18] and the lack of money to pay [14 , 16] . In campaigns using a combination of SPs and D2D vaccination strategies , it is also possible that owners do not attend SPs as they expect to get their dogs vaccinated during the door-to-door campaign [16] . Furthermore , while giving out dog collars or wristbands can increase participation [18] , charging the owners for vaccinations can result in lower vaccination rates [16] . Despite these studies , there is still an incomplete understanding of the barriers which limit attendance at static points . The need to understand and overcome barriers to SP attendance is particularly important in Blantyre , Malawi where rabies is an important cause of mortality , especially in children [19] . In order to address the high incidence of rabies in this population , we have embarked on an annual mass dog vaccination campaign throughout the city . We have previously reported that although 97% of the dog population is owned , only 53% out of the 79% overall vaccination coverage we achieved in our 2015 vaccination programme was attributed to dogs vaccinated at a SP , with the remaining 26% achieved by vaccinating dogs at door-to-door [20] . In order to make the vaccination campaign financially sustainable in the longer term , we need to reduce the reliance on D2D vaccination and encourage higher attendance at SP vaccination stations . Consequently , the aim of this study was to investigate the barriers to attendance at SP vaccination clinics using a multi-faceted approach including modeling the relationship between distance to travel and attendance at SP together with dog owner questionnaires . Our study is the first large scale , city-wide study investigating the reasons for failure to attend static vaccination points in SSA .
Prior to vaccination of owned dogs , verbal informed consent was obtained from the person presenting the dog for vaccination . In the cases where an owner could not be identified , dogs were vaccinated in accordance with Government Public Health protocol , as the work was part of a non-research public health campaign . This study was conducted in Blantyre city , the second largest city in Malawi with an estimated human population of 881 , 074 in 2015 [21] . The city’s dog population in 2015 was estimated to be 45 , 526 based on mark re-sight methods [20] . The city covers an area of 220 km2 , which is divided into 25 administrative wards [22] . The campaign took place throughout the whole of Blantyre city . The vaccination campaign has been described in detail by Gibson et al . [20] . Briefly , the city was divided in 204 working zones and their sizes were subjectively dictated according to an area that could be covered by a vaccination team in one day . Each zone was assigned a land type based on appearance in Google Satellite Maps™: a ) housing category ( HS ) 1 ( small houses—high density ) , b ) HS 2 ( small houses—medium density ) , c ) HS 3 ( small houses-low density ) , d ) HS 4 ( medium houses—ordered ) , e ) HS 5 ( large houses-medium/low density ) , f ) industrial/commercial , g ) agriculture/open space . For the purposes of the regression analysis described below these were regrouped in high ( a ) , medium ( b , d ) and low ( c , e , f , g ) housing density areas . Mass dog vaccination across the city was carried out between the 30th of April and the 25th of May 2016 using two approaches; static point ( SP ) and door-to-door ( D2D ) . Using 8 vaccination teams working simultaneously , SP vaccinations were conducted at weekends followed by D2D vaccinations in the same area on the following Monday , Tuesday and Wednesday . All data analysis was carried out within the R statistical software environment [24] . Specific packages used are mentioned below .
22 , 924 dogs recorded during the door-to-door campaign were used for this analysis . This excluded 700 dogs that had been vaccinated by someone other than MR earlier that year . 91% included in the analysis were adults and 10% were neutered . 40% of dogs were female out of which 13% were either pregnant or lactating . The vast majority of dogs were owned , out of which 24% were recorded as always roaming , 34% as roaming daily but restricted at some point during the day , less than 1% as roaming weekly and 33% as never roaming . 10 , 476 dogs ( 0 . 46 ) were reported to be vaccinated at SP locations and for 6 , 271 of those we could retrieve their unique SP location ID . 4 , 756 ( 0 . 76 ) of those went to their nearest SP location , while 994 ( 0 . 16 ) went to their second nearest SP . Fig 1 shows the location of the SPs used in this campaign , as well as household to SP paths drawn for people who presented their unique SP ID number during the door-to-door campaign . Lastly , there was no statistical evidence of a relationship between distance to the nearest SP and missingness of vaccination card evidence . Based on the Google Maps route the distance attending dog owners were willing to travel to a SP vaccination clinic was on average 1 . 22 km with 75% of attending dog owners walking up to 1 . 5 km to the SP . Similarly , the mean straight line distance was estimated to be 0 . 812 km with an upper quartile of 1 . 016 km . S1 Fig demonstrates the difference between the two methods for calculating the distance and demonstrates why distance estimates can vary using the two methods . Fig 2 shows distance traveled to each of the 47 SP clinics , using the two distance estimation methods . It shows that there is great variation in the range of distances SPs manage to attract individuals from . This implies that there might be underlying reasons why some SPs attract individuals from much greater distances than other . In order to ensure that these differences are adjusted for , the addition of SP-level random effects was considered in the regression model . Data used to build a multivariable logistic regression model predicting attendance to a SP included dog related data , household related data extracted from poverty , land use and land type GIS data and straight line distance from nearest SP . A summary of data used as predictor variables is presented in S1 and S2 Tables . Furthermore , Fig 3 shows how the proportion of attendance to SP decreases as distance from nearest SP increases . Univariable analysis results are shown in Table 1 . Land use data were not considered for the model as almost all dogs were located within residential areas . Similarly , ownership status was excluded as very few of the dogs seen were strays ( 1% ) . All other variables were considered for the final model . Fig 4 shows the final multivariable logistic regression model predicting attendance to SP . Numerical results of the regression model can be found in S3 Table . While increasing distance from SP , being a puppy or pregnant/lactating decreased the odds of a dog being taken to a SP for vaccination , high proportions of poor people among a region , as well as living in a high and medium housing density area were positive predictors of attendance to SP . The model also showed that the effect of distance was increased with increasing levels of poverty i . e . there was an increased drop of attendance with distance in poorer people . Regarding dog characteristics being healthy or neutered increased the odds of a dog being taken to SP for vaccination . Lastly , compared to dogs who always roamed , dogs who were reported as never roaming were less likely to be taken to a SP , while dogs who were allowed to roam daily , but restrained for part of the day had increased odds for being taken to a SP for vaccination . The predictive ability of the model was assessed by using the model to predict whether a dog was taken to a SP or not using the test dataset . The AUC was calculated as 0 . 77 , indicating that the model was reasonably good at predicting the outcome . During the door-to-door survey , people who did not attend a SP clinic were asked why . Reasons quoted for not having attended a SP clinic are shown in Fig 5 . The most common ones included the owners being unaware , unavailable or unable to handle their dogs , distance and the puppies being too young . This result complements the results of our model by emphasising the importance of distance and the fact that the age of the dog will influence their decision on whether to bring it for vaccination . It also provides further information on other possible reasons why dogs might not be presented at a SP clinic , which our model was unable to take into consideration . These include owner related factors such as lack of awareness , availability and difficulty in handling dogs . Investigating further into the relationship between people being unaware of the SP vaccination campaign and distance to their nearest SP , it was found that people who said that they did not attend the SP because they were not aware of the campaign were located further to a SP than those who quoted different reasons for not attending ( mean unaware = 1 . 10 km , mean other reason = 0 . 92 km , p-value = < 0 . 01 ) .
This paper presents the results of the first large scale study investigating the reasons why attendance at SPs offering free rabies vaccinations for dogs is suboptimal in SSA . We were able to interrogate data from a city-wide vaccination campaign in Blantyre , Malawi using a combination of GIS and household questionnaire type data . We found that distance from household played an important part in SP attendance . Specifically , our regression model showed that for every km closer the odds that the dog was taken to a SP for vaccination were 3 . 3 times higher . Distance was also one of the main reasons for not attending SPs most commonly quoted by the owners ( 17% ) . This finding has been replicated in other studies , which also found distance decay in the use of health services in developing countries [33–37] . Our findings were consistent with smaller scale studies in rural Tanzania , where vaccination coverage decreased as distance from sub-village [18] or household [13] to SP increased . In order to better inform the planning of future vaccination campaigns we also estimated the distance people were actually willing to walk to a SP , using data from 6 , 271 dogs for which we could retrieve vaccination cards with SP IDs . Our approach was unique in mapping both straight line distance and actual path based distance . We estimated that people were willing to travel on average 1 . 22 km to a SP vaccination clinic with 75% of the people walking up to 1 . 5 km to the SP . Similarly , the mean straight line distance was estimated to be 0 . 812 km with an upper quartile of 1 . 016 km . This information is crucial and should be used in planning efficient vaccination campaigns in urban sub Saharan settings in order to improve vaccination coverage using SPs only . In addition , our study highlighted the different uses of straight line distance versus path distance . While straight line distance is very useful when designing mass campaigns as it is much easier to estimate , path distance is more accurate and would be a more valuable tool in estimating for example the cost of travel of each SP attendant . Our study clearly demonstrates that the path distance is on average 50 per cent greater than the straight line distance in this setting . We also found that socio-economic status influenced attendance to SP vaccinations . Our model shows that dogs from areas with higher proportions of people living in poverty are more likely to be presented for vaccination . Interestingly , the model also shows that the effect of distance described above is increased at increasing levels of poverty . In other words , there is an increased drop of attendance with distance in areas with higher proportions of people living in poverty . This is the first study to report this relationship , which highlights the importance of understanding more about which groups of people might be more inclined to bring their dog to a SP for vaccination . The only other study that has looked at this relationship has found no difference in vaccination coverage between households with high and low socio-economic status in rural Tanzania [13] . The conflicting results might arise due to the fact that our study was carried out in an urban setting . According to our experience in Blantyre , dogs are often brought to SPs by younger members of the family . Middle and high income parents might be less inclined to send their children alone to a vaccination point . Similarly , affluent people may consider their time more costly and be less willing to spend it waiting in queues in order to get their dogs vaccinated . The signalment of the dogs was also important in influencing likelihood of attending a SP . Our model shows that young dogs , pregnant or lactating females were less likely to be brought to SP vaccination stations . Young age was also reported by 9% of the owners themselves as a reason for not bringing a dog to a SP both in our study and other studies in SSA [16–18] . Puppies less than three months old are often excluded from vaccination campaigns , either due to the misconception that they cannot mount an immune response or because it would require administration of the vaccine off-label [38] . Nevertheless , previous experimental and field studies have shown that puppies can mount a protective immune response as young as 4 weeks old [38–40] . Puppies constitute up to 30% of the dog population in SSA [11] and can therefore play a crucial role in maintaining vaccination coverage beyond the 70% threshold . In fact , WHO guidelines on mass vaccination campaigns advise vaccinating all dogs including those under three months of age [10] . This important issue needs to be addressed through improved advertising and education in order to increase vaccination coverage by ensuring that puppies as well as adult dogs are presented to static vaccination points . Another interesting finding was the relationship between the reported dog confinement level and SP attendance . We found that compared to dogs who always roamed , dogs reported as never roaming were less likely to be taken to a SP . This might be because people believe that if dogs are not allowed to roam , they are not at risk of rabies . While this might be true for dogs that are kept in a protected area , many dogs will just be kept on a leash or in a garden where other dogs have access to and are therefore at risk of contracting rabies . This is another important issue to be raised during rabies education sessions and vaccination campaign advertisement . In comparison , dogs who were allowed to roam daily , but restrained for part of the day had increased odds of being taken to a SP for vaccination when compared to dogs who roamed all the time . This might reflect the fact that people who interact more with their dogs , are also keen to provide health care or simply be proxy for whether people were able to handle their dogs in order to bring them to the SP . We also found that a lack of awareness of the vaccination programme was important despite high local profile within the local media , communities and schools . SP vaccination stations were advertised using posters and local radio during the weeks preceding the campaign and announced using a loud speaker in the communities around each station in the days before the actual vaccinating teams arrived at the SPs [20] . Despite these efforts , the most common reason for not attending a SP quoted by the owners ( 27% ) was that they did not know about it . In fact , people further away from SP were less likely to be aware of the vaccination campaign . Promotion of a vaccination campaign is massively important and indeed being unaware was one of the most commonly quoted reasons for failure to attend a vaccination SP in other developing countries including Chad [16 , 17] , Mali [14] , Tanzania [18] and Peru [15] . Timely and accurate provision of information about upcoming SP vaccination stations is likely to increase participation at SPs , and might therefore be cost-effective for future campaigns to invest a greater proportion of resources on campaign advertisement and promotion making sure they cover the area of interest homogeneously . Another important reason for not bringing a dog to a SP identified by 19% of the owners in this study was difficulty in dog handling . This supports findings of previous studies in developing countries [13–15 , 17 , 18] . In settings where most dogs are owned for guarding or hunting [11] , dogs may be less accustomed to being walked on a leash , making it very difficult for owners to bring them to SP vaccination stations . In order to achieve greater coverage at SP this problem cannot be ignored . Promotional campaigns and rabies education work need to include information on how to safely handle and walk dogs . Such information might need to be provided throughout the year in order for the dogs to be more likely to be able to be handled at vaccination time . Examples of programmes focusing on improving dog handling have been used in several countries in Latin America [15] , but have not been previously described in SSA possibly due to economic constraints . With rapidly rising mobile phone ownership in SSA , regional mass SMS delivery through the most popular networks has the potential to greatly increase dissemination of information about time and location of up and coming SPs and therefore possibly increase turn-out . The present study has several strengths and limitations . Data used for our regression model were sourced from an intensive vaccination campaign which aimed to cover the whole city and is therefore likely to be very representative of the dog population in Blantyre city . This provided us with data about each dog’s signalment as well as GPS locations used to estimate distance to nearest SP and extract GIS data corresponding to each location . This resulted in a detailed dataset and enabled us to extensively explore factors affecting attendance at SP locations . Our model validation showed that our model was reasonably good at predicting the outcome , but there was some unexplained variation . This might have arisen due to the fact that GIS data sources for Malawi are limited and not very detailed or due to information we did not collect such as households who did not respond , number of dogs per households and whether they had equipment to restrain dogs . Similarly , this might be due to information we were unable to include in this kind of model and indeed our household questionnaire showed that two of the most common reasons for not attending SP were being unaware of the campaign and having difficulty in handling which were not included in the initial model . Lastly , we have used google maps to calculate the path distance people were willing to travel to SP stations . This is an innovative way of estimating path distance , which has not been used in rabies relevant studies before , providing a more realistic estimate compared to straight line distance . Nevertheless , it is important to remember that the accuracy of this estimate greatly depends on the accuracy of google maps data in each region and might not be applicable in all areas .
Overall , this is first large scale study investigating the barriers to obtaining adequate rabies vaccination coverage through SPs in an urban setting in SSA . Our results suggest that future vaccination campaigns should increase efforts on improving positioning of SPs so that they become more accessible . We have also shown that there is a clear need to provide timely and accurate information about upcoming campaigns , emphasing the importance of puppies being vaccinated and identifying ways to improve dog handling . Estimates from our model could be used to estimate the impact on vaccination coverage of adapting several measures such as increasing vaccination points or increasing the proportion of puppies vaccinated , however caution should be exercised due to potential factors not accounted for by the model . In conclusion , this study has provided valuable insight into the barriers to attendance at SPs in urban settings and this should be taken into consideration when designing future mass vaccination programmes using SP vaccination stations in order to allow high vaccination coverage to be achieved without the need for expensive and logistically challenging door-to-door programmes . | Rabies is a devastating yet preventable disease that causes around 59 , 000 human deaths annually of which a large proportion occurs in Sub Saharan Africa ( SSA ) . In order to eliminate rabies , annual vaccination of at least 70% of the dog population is recommended . In SSA most rabies vaccination programmes use static point ( SP ) vaccination approaches . Despite being logistically and economically more feasible than door-to-door approaches , SP vaccination campaigns often result in low vaccination coverage . Here we investigated the reasons why attendance at SPs offering free rabies vaccinations for dogs is suboptimal in SSA . We analysed data from a citywide vaccination campaign in Blantyre city , Malawi in combination with household related data . Our results found that the distance from home to SP influences attendance at SPs . We also found a clear need for provision of timely and accurate information about upcoming campaigns , including information on the importance of puppies being vaccinated as well as ways to improve dog handling . Understanding the barriers to attendance at SPs and taking them into consideration , would make mass vaccination programmes more feasible thereby allowing high vaccination coverage to be achieved without the need for expensive and logistically challenging door-to-door programmes . | [
"Abstract",
"Introduction",
"Materials",
"and",
"methods",
"Results",
"Discussion",
"Conclusion"
] | [
"medicine",
"and",
"health",
"sciences",
"maternal",
"health",
"obstetrics",
"and",
"gynecology",
"immunology",
"tropical",
"diseases",
"geographical",
"locations",
"vertebrates",
"social",
"sciences",
"malawi",
"dogs",
"animals",
"mammals",
"preventive",
"medicine",
"r... | 2018 | Barriers of attendance to dog rabies static point vaccination clinics in Blantyre, Malawi |
Urea-induced protein denaturation is widely used to study protein folding and stability; however , the molecular mechanism and driving forces of this process are not yet fully understood . In particular , it is unclear whether either hydrophobic or polar interactions between urea molecules and residues at the protein surface drive denaturation . To address this question , here , many molecular dynamics simulations totalling ca . 7 µs of the CI2 protein in aqueous solution served to perform a computational thought experiment , in which we varied the polarity of urea . For apolar driving forces , hypopolar urea should show increased denaturation power; for polar driving forces , hyperpolar urea should be the stronger denaturant . Indeed , protein unfolding was observed in all simulations with decreased urea polarity . Hyperpolar urea , in contrast , turned out to stabilize the native state . Moreover , the differential interaction preferences between urea and the 20 amino acids turned out to be enhanced for hypopolar urea and suppressed ( or even inverted ) for hyperpolar urea . These results strongly suggest that apolar urea–protein interactions , and not polar interactions , are the dominant driving force for denaturation . Further , the observed interactions provide a detailed picture of the underlying molecular driving forces . Our simulations finally allowed characterization of CI2 unfolding pathways . Unfolding proceeds sequentially with alternating loss of secondary or tertiary structure . After the transition state , unfolding pathways show large structural heterogeneity .
Protein denaturation by osmolytes such as urea or guanidinium is widely used to study protein folding and stability . The underlying mechanism , however , is not yet fully understood on the molecular level . Despite the large number of theoretical and experimental studies carried out in the past decades to shed light on the molecular details of this process , no clear picture has emerged yet . On the one hand , the microsecond to millisecond timescales at which individual folding and unfolding events occur , as well as the need for synchronization in ensemble measurements , and the structural heterogeneity of unfolding pathways renders it difficult to gain atomistic insight from experiments . On the other hand , for computer simulations , folding/unfolding processes are typically too slow or too rare to be accessible . Two basic model classes have guided the study of the driving forces of urea-induced protein denaturation , and still set the framework for ongoing discussions . According to the first model , urea induces changes in the water structure , which in turn weaken the hydrophobic effect and thus cause protein denaturation [1]–[3] . In this model of indirect interactions , two alternative views have been put forward in which urea is regarded either to break [1] , [2] , or to enhance [3] water structure . The second model , in contrast , attributes the denaturing effect of urea to direct interactions between urea and the protein [4]–[6] . Also this model comprises different aspects: either the interaction of urea with polar residues or the peptide backbone , mainly via hydrogen bonding [5]—or hydrophobic interaction with apolar residues [4] . All of these possibilities , and various combinations thereof , have been suggested as the primary driving force of denaturation , and are still controversially discussed . Whereas some studies have provided support for the primacy of indirect effects [7]–[12] , this concept has been challenged by many authors [6] , [13] , [14] , and many recent studies provide increasing evidence for direct interactions as the primary driving force for denaturation [14]–[25] . Within this framework , however , it is controversially discussed whether either polar [11] , [12] , [17] , [22] , [23] , [25] or apolar [4] , [15] , [16] , [18]–[21] , [24] , [26] interactions between urea and the protein dominate . Here we address this question by studying the relevance of direct polar and apolar contacts with all-atom molecular dynamics ( MD ) simulations . We have chosen the chymotrypsin inhibitor 2 ( CI2 ) protein as an example , the folding kinetics and thermodynamics of which have been extensively studied experimentally [27] . We consider the CI2 in water as well as in aqueous urea solution , and perform a thought experiment ( “Gedankenexperiment” ) , in which urea polarity is varied by scaling its partial atomic charges . The rational of this computer experiment is as follows . If polar contacts such as hydrogen bonds between urea and the protein constituted the determinant interaction for denaturation , one would expect hyperpolar urea to be an even stronger denaturant than real urea . If , in contrast , apolar contacts played the major role for denaturation , one would expect hypopolar urea to be the stronger denaturant . Therefore , by monitoring the respective denaturation strengths in the simulations , we will be able to decide which of the two interaction types drives urea-induced unfolding .
All simulations were performed using the Gromacs [28]–[30] program suite , versions 3 . 2 . 1 and 3 . 3 , with the OPLS-all-atom force-field [31] , [32] . The TIP4P water model [33] was used , and the urea force field was adopted from Smith et al . [34] , which is a refined version of the original OPLS parametrization by Duffy et al . [18] . A cutoff of 1 . 0 nm was used for short-range Coulomb as well as Lennard-Jones interactions . Particle Mesh Ewald summation ( PME ) [35] , [36] was used to calculate the long-range electrostatic interactions with a grid-spacing of 0 . 12 nm and an interpolation order of 4 . All simulations were performed in the NpT-ensemble using Berendsen-type temperature-coupling [37] with a coupling coefficient of τT = 0 . 1 ps and Berendsen-type pressure-coupling [37] at 1 bar with a coupling coefficient of τp = 1 ps . To allow comparison with the simulations reported in [11] , the simulation temperature was set to 333 K ( except for one simulation at 300 K ) , and the same CI2 double mutant ( E33A , E34A ) was used . An integration timestep of 2 fs was used together with the LINCS constraint solver [38] for all covalent bonds . The structure of the CI2 protein was taken from the Protein Data Bank [39] , PDB-code 1YPC [40] . Unresolved side chain atoms for residue MET40 ( residue number 59 in the pdb file ) were added using the program WHAT IF [41] . The box-size was chosen such that a minimum distance of 1 . 5 nm between protein atoms and the box was kept . For the solvation of the protein , pre-equilibrated structures of water and 8 M urea were used ( taken from [42] ) . Sodium and chloride ions were added to yield a 150 mM ion concentration and mimic physiological conditions . Prior to each simulation , a 200 step steepest descent energy minimization and a 500 ps equilibration run with position restraints on the protein heavy atoms were carried out . To avoid over-interpretation of possibly anecdotal events , multiple simulation runs were carried out for each parameter set ( Table 1 ) . Two simulations of CI2 in water , three simulations with regular urea charges , two simulations with 25% urea charges , five simulations with 50% urea charges , four simulations with 75% urea charges , two simulations with 150% urea charges and two simulations with 200% urea charges were performed , each at 333 K . In addition , one simulation in water at 300 K was performed to define native contacts and native secondary structure ( see below ) . The total simulation time of all simulations was ca . 7 µs . We note that a computational thought experiment not dissimilar to the one performed here was conducted by Sorin et al . [43] , who investigated the relationship between solvent and protein structure in a “hydrophobic titration” experiment employing different TIP3P variants . Solvent accessible hydrophobic surface areas ( SAS ) were calculated using the double cubic lattice method [44] with a 0 . 14 nm probe radius . Native contacts and native secondary structure were defined using the simulation at 300 K in water ( W300 K ) , rather than the crystal structure . This approach has the advantage that fluctuations of the native state were captured which allowed a more direct comparison with the unfolding simulations . Residues were defined to be in contact if the distance between the closest atom pair was not larger than 0 . 4 nm . Contacts were defined as native if they were present during more than 50% of the time in simulation W300K . Contacts between neighboring residues were not considered for the calculation of the native contact fraction . Secondary structure was classified using DSSP [45] . The native secondary structure was defined as the most frequently occurring structure type for each residue seen in simulation W300K , which was similar to that of the crystal structure . Helix , β-sheet , and turn-elements were considered to calculate the fraction of native secondary structure content . To quantify the frequency of interactions between urea and the amino acids , we used the contact coefficient CUW [46] for a particular amino acid X , ( 1 ) where NX–U and NX–W are the numbers of atomic contacts of amino acid X with urea and water molecules , respectively . Atoms were defined to be in contact if close than 0 . 35 nm . CUW is normalized using the total numbers of urea atoms ( MU ) and water atoms ( MW ) . Accordingly , a residue with a contact coefficient of CUW = 1 . 0 has no interaction preference for either urea or water . Values above 1 . 0 indicate preferential interaction with urea , values below 1 . 0 indicate preferential interaction with water .
As a reference , we first analyzed the dynamics and stability of the folded CI2 protein as well as its protein-solvent interactions both in water and in 8 M aqueous urea solution . Figure 1 shows the Cα root-mean-square-deviation ( RMSD , panel A ) and the solvent accessible hydrophobic surface area ( SAS , panel B ) for the simulations in water ( W1 , 2 , blue ) and in 8 M urea solution ( , green ) . As can be seen , the Cα-RMSD of the protein in both solvents shows similar fluctuations with an average value of 0 . 3 nm , and no significant differences between both solvents are seen . In particular , no unfolding is observed , which is expected from the measured millisecond time scale for CI2 denaturation [47] . In contrast , and perhaps unexpectedly , the average SAS in aqueous urea is 2–3 nm2 larger than in water . As can be seen in Figure 1B , this difference is significantly larger than the SAS fluctuations of single trajectories . Closer inspection reveals that this difference results mainly from few specific residues whose side chains are more solvent-exposed in aqueous urea than in water . In particular , MET1 , LEU32 , ILE44 and PHE50 contribute dominantly to this difference ( 0 . 22 nm2 , 0 . 17 nm2 , 0 . 30 nm2 and 0 . 29 nm2 , respectively ) . With only a few exceptions ( e . g . , ARG43 ) , however , also the side chains of almost all other residues are slightly more exposed in aqueous urea solution than in water . Because these amino acids are among those which were found to have particularly strong contact preferences for urea ( see [46] ) , we expect that the increased exposure of these side chains is caused by favorable interactions with urea molecules . To check whether this trend holds not only for tripeptides [46] , but also for the whole protein , we quantified these interactions using the contact coefficient CUW . Figure 2C shows the CUW values for each amino acid type in the CI2 , averaged over time and over the three simulations in aqueous urea solution ( ) . Indeed , the obtained contact coefficients are largely similar to those calculated for the individual amino acids in tripeptides [46] . In particular , apolar and aromatic amino acids , as well as the backbone , have pronounced contact preferences for urea , whereas charged amino acids have preferences for water contact . This finding confirms that polarity/apolarity is clearly a determining factor for the specific interactions of urea with the CI2 protein residues , and provides further motivation for our approach to investigate protein stability in solutions of urea with modified polarity . We note that the remaining differences between the contact coefficients of tripeptides versus those observed here for CI2—quantified by a correlation coefficient of r2 = 0 . 69—suggests that effects from sequence and structure of the folded CI2 protein account for ca . 30% of the contact preferences . To investigate the denaturation strengths of hyper- or hypopolar urea , the partial charges of urea were scaled to values of 25% , 50% , 75% , 150% , and 200% . For each of these modified degrees of polarity , the CI2 protein was simulated in aqueous solution . Urea with partial charge scaling of x% will be denoted as “ureax%” . Since it is a priori not clear that upscaling or downscaling urea partial charges does in fact enhance polar or apolar , respectively , interactions with the protein , we investigated the contact coefficients of each amino acid type in the CI2 for hypo- and hyperpolar urea . As can be seen in Figure 2A and 2B , hypopolar urea indeed shows less interactions with charged and polar amino acids , and enhanced interactions with less polar residues . Hyperpolar urea150% , in contrast , exhibits fewer interactions with those amino acids preferentially interacting with “regular” urea100% ( Figure 2D ) . Interactions with charged residues are even preferred by urea150% over interactions with less polar residues . In summary , lowering the polarity of urea enhances its interaction preferences: less preferred interactions become even less frequent , and preferred interactions become even more frequent . An exception is ARG , which does not show enhanced interactions for urea150% . We attribute this effect to the fact that ARG contains large polar as well as apolar parts . Having shown that upscaling or downscaling urea partial charges has the desired effect on the interaction strengths between urea and the different amino acids , we can now turn our attention to the influence on protein stability . Accordingly , we monitored the SAS for the different urea partial charge scalings ( Figure 3 ) . As can be seen , for hypopolar urea , the protein unfolds in all nine simulations ( urea75% and urea50% , magenta and orange lines , respectively ) . In contrast , for hyperpolar urea150% , the SAS remains close to the native value and the protein remains stable in all simulations ( black lines ) . In fact , the SAS is even smaller for hyperpolar urea than for regular urea , which suggests that hyperpolar urea compacts the folded state . Furthermore , this result suggests that urea150% would actually be a weaker denaturant than urea100% . In summary , enhanced apolar interactions between urea and the protein destabilize the native state and induce unfolding of the CI2 . Strengthening apolar interactions yields a stronger denaturant , while strengthening polar interactions yields a weaker denaturant . We have also performed simulations with “extreme” urea25% and urea200% . However , these simulations exhibit artifacts which render them irrelevant for the present purpose and are therefore not shown in Figure 3 . For partial charges scaled down to 25% , on the one hand , urea shows a strong tendency to self-aggregate to a hydrophobic layer in the periodic simulation box , which does not any more interact with the protein . Urea200% , on the other hand , induces a glass transition in the solvent , with drastically reduced urea diffusion coefficients ( from ≈2 . 2·10−5 cm2/s to <0 . 001·10−5 cm2/s ) . As a result of the vanishing mobility , the urea molecules do not interact with the protein either . Similar underestimations of the diffusion coefficients in common force-fields has previously been observed for high ion concentrations [48] . We note that these two side-effects , urea aggregation and reduced diffusion coefficients , were also observed for the simulations with urea50% and urea150% , respectively , albeit to a ( much ) lesser extent . Care has to be taken , therefore , that these side-effects do not affect our main conclusions . In particular , one might argue that protein unfolding in urea50% is not necessarily a direct consequence of reduced urea polarity . Rather , it might be caused by inhomogeneities of urea concentration . However , since the observed unfolding events are quite similar to those observed for urea75% , where no significant aggregation is seen , we do not expect locally enhanced urea concentration to play a significant role . For the simulations with hyperpolar urea150% , one might object that not enhanced protein stability , but reduced urea diffusion coefficient for urea150% ( from ≈2 . 2·10−5 cm2/s to ≈0 . 1·10−5 cm2/s ) is the reason that no unfolding is observed . To address this concern , two effects of this reduced urea mobility need to be considered . First , the reduced mobility of urea molecules implies much slower thermodynamic equilibration . And therefore , the thermodynamic equilibrium distribution at the protein surface might not be reached within the available simulation time . However , the diffusion time for a urea molecule to cross the whole box length is well within the simulations time ( ≈75 ns for urea150% ) , such that this effect can be excluded . Second , the reduced mobility of urea molecules might slow down conformational changes of the protein due to higher solvent viscosity . Note , however , that conformational changes are seen on the simulation timescale , which lead to the observed compaction . Furthermore , as can be seen from the fast 10 ns SAS jumps in urea75% and urea50% , even a 20-fold enhanced viscosity is unlikely to prevent motions on a 500 ns timescale . This observation , together with the fact that other proteins , e . g . the Cold Shock protein , are observed to undergo large conformational changes in hyperpolar urea150% ( data not shown ) strongly suggests that the increased solvent viscosity does not compromise our interpretation . The extent of both side-effects , self-diffusion slowdown and urea aggregation , is shown in the Supporting Information ( Text S1 ) . In our simulations , the CI2 protein unfolds reproducibly in urea75% ( all four simulations ) and urea50% ( all five simulations ) which allows us to analyze unfolding pathways in more detail . To this aim , Figure 4 shows the unfolding pathways observed for the four simulations in urea75% . Here , the unfolding pathway is characterized by the fraction of native secondary structure versus fraction of native tertiary structure , measured by the fraction of native contacts . ( The respective data for urea50% is provided as Figure S1 in the Supporting Information ) . In each of the nine cases , starting from the folded state ( top right ) , the protein undergoes conformational changes eventually leading to denaturation and unfolding in all nine trajectories . We first describe one unfolding trajectory ( ) in detail , and subsequently discuss common features and differences of all nine unfolding trajectories . In simulation , reversible fluctuations of the secondary-structure ( β-strand 3 , ILE57-ARG62 ) trigger the first unfolding step . After 29 ns , a part of the coil region between β-strand 1 and β-strand 2 reorients . In particular , the sidechains of THR36 , ILE37 and VAL38 rotate by about 180° , which apparently triggers , at 30 ns , a subsequent flip of the turn region formed by residues 22–25 . This irreversible and fast unfolding step implies significant loss of native contacts and is followed by a longer phase of 80 ns during which the α-helix ( res . 13–22 ) unfolds , with the ALA-rich region ( ALA14 , ALA15 , ALA16 ) unfolding last at 110 ns . Subsequently , the turn ( and former α- ) region between residues 18–25 detaches from the protein core , while the ALA-region of the helix undergoes several partial refolding and unfolding events . Between 140 ns and 150 ns , further global unfolding rearrangements of the tertiary structure occur . At 150 ns , unfolding is completed with the disruption of β-strands 2 ( res . 46–52 ) and 3 ( res . 56–62 ) . Whereas the sequence and all the details of the described unfolding events are not necessarily similar in all unfolding trajectories , several common features emerge . In all simulations , unfolding proceeds stepwise , with alternating phases of loss of secondary and tertiary structure . In none of the simulations , both structure levels are seen to break down simultaneously; also not seen is the complete loss of one structure level before the other . Often , meta-stable parts of the trajectory , each sampled for a longer time ( typically 100 ns and longer ) and characterized by reversible fluctuations , are connected by fast transitions ( about 5 ns ) , during which irreversible loss of native contacts occurs . Such alternating stepwise unfolding pattern is consistent with the nucleation-condensation mechanism of folding for the CI2 protein which has been derived from φ-value analysis [49] . In summary , a sequence of alternating unfolding steps is observed , which supports unfolding models that assume a strong coupling between tertiary and secondary loss of structure . We would like to emphasize that the sequence of meta-stable states seen in our simulations is consistent with the fact that CI2 is a two-state folder [27] , because the observed transient states are both too short-lived and too heterogeneous to be resolved in current ensemble- or equilibrium-unfolding experiments . Next , we investigated whether regions of the CI2 exist where unfolding is particularly likely to start . To this end , the RMSD per residue was calculated for the initial phase of unfolding ( defined by a significant increase in the SAS from the native value ) for each of the simulations U75% and U50% ( Figure 5 ) . For comparison , the top row shows the root-mean-square-fluctuations per residue in the native state ( simulation W300K ) . Many initial unfolding steps are seen to occur in regions that exhibit large fluctuations already in the native state in water at 300 K . Examples are the C-terminal end of the α-helix ( res . Q22 ) and the adjacent turn-region ( res . D23–E26 , simulations ) , as well as the coil- and turn-regions between β-strands 2 and 3 ( simulations ) . In contrast , regions that show only small fluctuation in the native state , e . g . res . 5–18 in simulations , tend to unfold later . In summary , no unique unfolding “hot-spot” is found , but rather several regions where unfolding likely begins . This observation led us to investigate whether common transient structures or putative intermediates exist in the unfolding pathways . To this end , for every unfolding trajectory i , the RMSD was calculated with respect to every structure Xj ( t ) ( with a time resolution of Δt = 100 ps ) of each of the other unfolding trajectories j ( data not shown ) . In this analysis , conformations which occur in trajectory i as well as in trajectory j , would be revealed by a minimum in the respective RMSD . Unexpectedly , no pronounced minima were found , which indicates that no pair of trajectories shares common global structures , and that unfolding proceeds structurally different in all nine cases . Rather , during unfolding as well as after complete denaturation , the protein explores quite different regions of phase space . This finding further implies that the transition state ensemble consists of conformations which are structurally more heterogeneous than the thermal fluctuations of the native state . These results are consistent with the previous observations of a broad transition state ensemble [50] and the fact that the CI2 protein is a two-state folder without pronounced intermediates [27] . We therefore attempted to analyze the transition state ( TS ) ensemble in more detail . Although the TS ensemble can not be rigorously defined from the nine trajectories at hand , a reasonable estimate can be given . To that end , we calculated the ( non-equilibrium ) density ρ of states for the SAS as reaction coordinate for each of the nine unfolding trajectories , which served to provide a rough free energy estimate , −kT log ρ . In all nine trajectories , the native state showed up as a minimum at low SAS values , with an adjacent clear maximum , which served to locate the TS ( data not shown ) . In most simulations , this maximum was consistently seen at an increased SAS of 3–5 nm2 . This agreement suggests that our approach provides a reliable estimate for the TS . In all nine simulations , the overall structure of the TS is found to be similar to that of the native state ( about 70% native contacts ) , but more expanded , in agreement with previous experimental [51]–[53] and simulation results [11] , [54] . The α-helix is still intact , albeit with its central region bent away from the molecule's center in most simulations , whereas the β-sheet is already partially disrupted in most cases . In agreement with a previous simulation study [11] , we find the TS ensemble to be heterogeneous with respect to the loops , turns , and terminal regions . After the TS , unfolding proceeds in six out of the nine trajectories with disruption of the β-structure before unfolding of the α-helix; conversely , in the remaining three simulations , the α-helix unfolds before the β-strands . In all cases , the time span between α- and β-disruption was rather short; therefore , no defined sequence of the two processes was established . We finally focus on the residual structure in the denatured state . In particular , we investigate a possible polyproline II helix structure ( PPII , φ = −75° , ψ = 150° ) which has been suggested as prevalent configuration for the denatured ensemble from CD-spectroscopy results [55] . Recently , this suggestion has gained considerable attention due to accumulating evidence for residual structure of denatured proteins [56]–[59] . We note that the sampling of the denatured state is very limited in our simulations ( ≈650 ns in total for urea75% ) , such that we expect this analysis to provide rough estimates rather than accurate numbers . Figure 6 shows the Ramachandran plot for the folded CI2 protein ( averaged over all simulations with urea100% , panel A ) , and the denatured protein ( averaged over the denatured ensemble in all simulations with urea75% , panel B ) . Similar distributions of the folded or unfolded ensemble were seen in the other simulations ( data not shown ) . As expected , the native state predominantly occupies three regions in ( φ , ψ ) space; the α region around ( −70° , −27° , “α” ) , as well as the β-sheet regions around ( −83° , 128° , parallel or PPII , “pp2/pβ” ) and ( −142° , 149° , antiparallel , “apβ” ) . For the denatured protein , the same three regions are populated , although with different occupancies ( Figure 6B ) . In particular , the PPII/pβ region becomes the most populated one , particularly when the denatured protein is very extended ( SAS >40 nm2 ) , which supports the pronounced role of this secondary structure element . However , the other two regions remain populated: the population of the antiparallel-β region increases from 11% to 17% , while the population of the α-region decreases significantly from 27% to 13% . Note that the presence of these backbone angle configurations does not imply correctly formed secondary structure elements in the denatured state . The unambiguous classification of PPII from ( φ , ψ ) is complicated by the fact that other secondary structure elements share similar backbone configurations . Therefore , several different definitions to calculate PPII content have been developed and applied in the past . Integration over the pp2/p β-peak in Figure 6 yields an increase of 26% to 32% relative population . With the definition from Jha et al [60] ( −100°<φ<0° , 50°<ψ<280° ) , we found a PPII population of ca . 35% in the denatured structures as compared to ca . 30% for the folded CI2 . A third definition ( −120°<φ<60° , 120°<ψ<240° ) , from Makowska et al [59] , yields similar results with increase from 35% to 40% PPII . Hence , for all three definitions , we observed a pronounced , but not absolute prevalence of PPII configuration in the denatured ensemble . This finding corroborates recent results by Makowska et al . [59] , who argued that PPII might be one of several possible backbone conformations in the denatured state .
To elucidate whether polar or , in contrast , apolar urea-protein interactions are the key driving force for urea-induced denaturation , thought experiment simulations were performed , in which the respective denaturation strengths of hyperpolar urea ( with strengthened polar interactions ) or hypopolar urea ( with strengthened apolar interactions ) were compared . To this end , the CI2 protein was simulated in water , in regular urea , and in hypo- and hyperpolar urea , which was realized by scaling the partial charges of the urea force field . In all nine simulations with reduced urea polarity , the CI2 protein unfolded within 300 ns . In contrast , the protein remained stable in the simulations with increased urea polarity , and the folded state was found to be even slightly more compact than in water . These results provide strong evidence that interactions with less polar parts—rather than polar interactions—are the main driving force for urea-induced protein denaturation . Together with previous results [46] , a coherent picture for urea-induced protein denaturation emerges . Urea molecules accumulate around less polar side chains and exposed backbone , forming an interface between less polar protein surface and water . The resulting displacement of water molecules from the protein surface into bulk water is entropically and enthalpically favorable and reduces the hydrophobic effect , such that unfolding of the protein becomes favorable . The ability of urea to form hydrogen bonds to the protein backbone is not the main driving force for denaturation , but contributes to the overall energetics by preventing unsatisfied hydrogen bond sites at the protein backbone . This view is also in agreement with recent spectroscopic results which provide evidence against the dominant role of polar interactions and hydrogen bonds [61] . It is interesting to note a relation to the mechanism of chaperone-mediated folding . Recent investigations of the chaperone GroEL [62] provide support for the suggestion that the hydrophobic environment of the open state of GroEL facilitates unfolding , whereas the hydrophilic environment of the closed state of GroEL facilitates folding [63] , [64] . In our simulations , we also find a more hydrophobic environment ( aqueous solution of hypopolar urea ) to facilitate unfolding , and a more hydrophilic environment ( aqueous solution of hyperpolar urea ) to facilitate folding . For regular urea , the preferences of the 20 natural amino acids for contacts with either urea or water were largely similar to those found previously for tripeptides [46] . In particular , less polar residues interacted preferentially with urea , whereas polar and particularly charged residues had stronger preferences for interaction with water . As expected , the characteristics of this interaction profile were amplified for hypopolar urea , and inverted for hyperpolar urea . The observation that the CI2 protein does not unfold within several hundred microseconds in urea with regular charges is consistent with the measured millisecond unfolding time [47] . We could not reproduce the complete nanosecond-unfolding seen in previous simulations [11] , which however employed a cutoff-approximation for the long-range electrostatics . On the structure level , our simulations suggest that denaturation proceeds rather heterogeneously and not via narrow , distinct pathways . In particular , unfolding of the CI2 was observed to start in stochastically one of several regions rather than one . However , regions with large structural fluctuations already in the folded state often turned out to be primary unfolding regions . Moreover , the nine unfolding pathways in the simulations with urea75% and urea50% turned out to share no common conformations during unfolding , which is consistent with the fact that CI2 is a two-state-folder without meta-stable folding intermediates . This heterogeneity of unfolding pathways prompts us to suggest an “inverted funnel”-scenario for the unfolding energy landscape , with multiple pathways leading from the narrow mesa of the folded state down to the relatively flat and extended region of the denatured ensemble . Whereas no shared conformations were found in the different unfolding pathways on the detailed level , more general common features of the unfolding process emerge . In particular , in most of the simulations unfolding was observed to proceed with alternating and sequential loss of secondary and tertiary structure . This finding is consistent with the coupling between secondary and tertiary structure formation in the nucleation-condensation folding process of the CI2 inferred from spectroscopic and mutation studies [49] , [65] , [66] . It further suggests that the processes of structure-formation during folding and structure-loss during denaturation share common features . Finally , our simulations allowed us to analyze the residual structure in the denatured state . Overall , relatively little residual secondary structure was seen , in agreement with previous CD studies [65] . Polyproline II turned out to be the most prominent , however not dominant residual structure in the unfolded ensemble . This finding supports the recent suggestion that polyproline II is one of several possible backbone conformations in the denatured state [59] . α-helical structure was found to be drastically reduced , whereas the population of β-sheet like backbone conformations was even slightly enhanced in the denatured state . Should such increase of β-sheet like backbone conformations turn out to be a common feature of unfolded protein ensembles , it might be relevant for the structural understanding of β-amyloid formation . | To perform their physiological function , proteins have to fold into their characteristic three-dimensional structure . While the folded state is stable under physiological conditions , changes in the solvent can destabilize the folded state and even induce denaturation . One of the most commonly used denaturants is urea . Despite its widespread use to study protein folding and stability , however , the molecular mechanism and particularly the driving forces of urea-induced protein denaturation are not yet understood . Two mechanisms have been suggested , according to which denaturation is driven either by polar interactions via hydrogen bonds or by hydrophobic interactions with apolar amino acids . By systematically varying urea polarity and quantifying the interactions of the solvent molecules with all amino acids of the protein , the present simulation study reveals that it is mainly the apolar interactions that drive denaturation . Our results suggest a coherent microscopic picture for urea-induced denaturation and bear more general implications for protein stability in other environments , e . g . , in chaperone-assisted folding . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"computational",
"biology/molecular",
"dynamics",
"biophysics/theory",
"and",
"simulation",
"biophysics/protein",
"folding"
] | 2008 | Polar or Apolar—The Role of Polarity for Urea-Induced Protein Denaturation |
In recent years , comparative genome sequence analysis of African Mycobacterium ulcerans strains isolated from Buruli ulcer ( BU ) lesion specimen has revealed a very limited genetic diversity of closely related isolates and a striking association between genotype and geographical origin of the patients . Here , we compared whole genome sequences of five M . ulcerans strains isolated in 2004 or 2013 from BU lesions of four residents of the Offin river valley with 48 strains isolated between 2002 and 2005 from BU lesions of individuals residing in the Densu river valley of Ghana . While all M . ulcerans isolates from the Densu river valley belonged to the same clonal complex , members of two distinct clonal complexes were found in the Offin river valley over space and time . The Offin strains were closely related to genotypes from either the Densu region or from the Asante Akim North district of Ghana . These results point towards an occasional involvement of a mobile reservoir in the transmission of M . ulcerans , enabling the spread of bacteria across different regions .
Mycobacterium ulcerans is an emerging pathogen with elusive reservoirs and transmission pathways . It causes the devastating skin disease Buruli ulcer ( BU ) that mainly affects rural populations in West Africa [1] . M . ulcerans is a descendant of the fish and occasionally human pathogen Mycobacterium marinum [2] , from which the new species has evolved through the acquisition of a plasmid encoding the enzymatic machinery for the synthesis of the macrolide toxin mycolactone [3] . From this common ancestor at least three different lineages or ecovars have evolved through genome reduction [4] . Clinical isolates from Africa belong to the classical lineage and differ from each other only in a very limited number of single nucleotide polymorphisms ( SNPs ) [4 , 5] , indicative for a highly clonal recent expansion of the pathogen in Africa . BU is characterized by a focal distribution of cases within endemic countries . Previous studies have revealed a strong association between genotype and the geographic origin of strains [4 , 6 , 7] , speaking for the development of local clonal M . ulcerans complexes following the introduction of this pathogen into a new area . The limited genomic diversity found within these local clonal complexes is however sufficient for studies on the distribution of variants at a micro-epidemiological level [8] . Since human-to-human transmission seems to be rare , findings point towards infection from a relatively localized environmental reservoir of the pathogen . In view of the association of BU outbreaks with stagnant and slow-flowing water bodies , a reservoir in the aquatic ecosystem is considered likely [9] . While in several African endemic areas a single unique clonal complex has been identified [4 , 6–8] , a recent comparative whole genome sequencing study of isolates from residents of the Asante Akim North district of Ghana showed for the first time the concurrent presence of two distinct clonal complexes within one BU endemic area [10] . Within the framework of a comprehensive genome analysis of clinical M . ulcerans isolates from Ghana , we analyzed genomes of a limited number of strains from the Offin river valley and equally observed a co-existence of two clonal complexes .
Ethical approval for the study was obtained from the institutional review board of the Noguchi Memorial Institute for Medical Research ( Federal-wide Assurance number FWA00001824 ) . Written informed consent was provided by all study participants . In an exhaustive active BU case search conducted in 2013 and a subsequent continuous monitoring of cases over a 17-months period in 13 randomly selected communities located in the historically highly BU endemic Offin river valley , an unexpectedly low prevalence of BU was revealed with only 11 laboratory-confirmed cases identified [11] . Two M . ulcerans strains that could be isolated from two of the 11 patients , as well as three M . ulcerans strains isolated in 2004 from lesions of two BU patients residing in the valley were analyzed in this study ( Table 1 ) . In addition , we included 48 M . ulcerans isolates from patients residing in BU endemic areas located in the Densu river valley of Ghana , which were part of a previous SNP typing study [8] . All M . ulcerans strains were subjected to whole genome sequencing . Genomic DNA was extracted from M . ulcerans cultures by phenol-chlorophorm extraction and ethanol precipitation as described previously [12] . Multiplexed genomic DNA libraries were prepared and sequenced on an Illumina HiSeq 2000 on 75-bp paired-end runs [13] . Illumina reads were aligned to the complete reference genome of M . ulcerans strain Agy99 ( GenBank accession number CP000325 . 1 ) with an insert size between 50 and 400 bp using BWA version 0 . 7 . 10 . SNPs were identified using SAMtools [14] as described [15] and were filtered for a minimum mapping quality of 30 and a quality cutoff of 75% . SNPs called in repetitive regions of the M . ulcerans reference genome ( 737 , 280 bp ) were excluded from the analysis and only the SNPs mapped in the core genome ( 4 , 894 , 326 bp ) were used to construct the phylogenetic trees . Maximum-likelihood phylogenetic analysis was performed using RAxML [16] on the alignment of identified SNPs from across the Ghanaian genomes sequenced here , together with genomes sequenced in previous studies [4 , 10 , 17] . Additional strains isolated from BU patients from other regions of Ghana and from Benin and Australia were included in the analysis to provide a comprehensive genetic context for the analysis of genetic diversity among the Offin and Densu isolates .
Phylogenetic analysis demonstrated the expansion of a single clonal complex in the Densu river valley ( Fig 1 ) . This complex has diversified substantially , but still forms a separate cluster , distinct from other African local clonal complexes ( Figs 2 and 3 ) . When compared to the second branch of the classical lineage of M . ulcerans—isolates from Australia—it is evident , that all African isolates are genomically extremely closely related ( Fig 2 ) . In contrast to the observation of a single clonal complex in the Densu river valley , our analysis revealed for the Offin river valley the presence of members of two distinct clonal complexes ( Fig 3 ) . Two Offin isolates ( NM031 and NM997 ) were closely related to isolates from the Densu river valley of Ghana ( Figs 1 and 3 ) . The other three Offin isolates ( NM022B , NM022D and NM972 ) –separated from the two Densu-like Offin strains by 29 SNPs–clustered with strains ( belonging to a clonal complex designated Agogo-1; [10] ) from the Asante Akim North district in the Ashanti region of Ghana ( Fig 3 ) . Intra-genotype average diversity was low with 12 and 17 identified SNPs among Densu-like and Agogo-1-like Offin isolates , respectively . Not a single SNP difference was found between the genomes of two strains ( NM022B and NM022D ) isolated from two different lesions of the same patient . In a next step , we combined the phylogenetic analysis of the Offin isolates with information on the residence of the patients within the river valley and the year of strain isolation ( Fig 4 ) . In both 2004 and 2013 one member each of the two clonal complexes , isolated from BU patients resident in different communities was found . Due to the limited number of isolates and missing details on the travel history of the BU patients from which these strains have been isolated , no firm conclusions could be drawn concerning the apparent lack of geographical clustering . However , the data revealed a co-circulation of two distinct M . ulcerans clonal complexes in the Offin river valley over space and time . In contrast to the remarkably strong link between genotype and geographical origin of clinical M . ulcerans isolates reported in previous genotyping studies conducted in African BU endemic foci [6–8] , two distinct M . ulcerans clonal complexes were recently found to co-exist in the Asante Akim North district of Ghana among strains isolated within the short time frame of two years and an area of only 30km2 [10] . It was concluded that M . ulcerans genotypes might be spread across larger areas , suggesting the presence of a rather mobile reservoir of infection in addition to the postulated more focalized aquatic niche environment typically associated with the pathogen [9] . In this context , recent data indicated that M . ulcerans is able to persist for several months in underwater decaying organic matter [18] , possibly as a commensal in protective aquatic host environments [19–21] . While the specific factors favoring the persistence of M . ulcerans in the environment and its transmission are yet to be explored , a complex interplay between environmental factors as well as biotic and abiotic drivers is assumed [22 , 23] . Reductive genome evolution of M . ulcerans speaks for niche adaptation [17] . In the present study we revealed the co-existence of both Densu-like and Agogo-1-like M . ulcerans genotypes in communities along the Offin river at two time points separated by ten years . Our data thus show that co-existence of clonal complexes in one BU endemic area may prevail over longer time periods . A mobile mammalian host , allowing the bacteria to replicate and to be shed to the environment , that way forming a reservoir [18] from which humans may be infected by unknown mechanisms , could be the missing link explaining the spread of M . ulcerans from an established BU endemic region to a new area . However , as demonstrated here by the presence of only a single clonal complex in the Densu river valley , the exchange of genetic M . ulcerans variants between BU endemic areas appears to be an extremely rare event . While in Australia possums have been identified as a host for M . ulcerans [24] , another line of evidence points to the involvement of humans with chronic ulcerative BU lesions in the spread of the bacteria in African BU environments . Extensive whole-genome sequencing studies are required to further unravel the evolutionary history and population structure of M . ulcerans in Africa . | Infection with Mycobacterium ulcerans causes the debilitating skin disease Buruli ulcer . Until today , transmission pathways and reservoirs of this emerging pathogen are not well understood . Generally , it is assumed that infection occurs after contact with potential environmental sources of M . ulcerans through puncture wounds or lacerations or via invertebrate vectors , such as aquatic insects contaminated with the bacteria . Comparative genome analyses of M . ulcerans strains isolated from patients living in the same BU endemic areas have revealed a close relationship between the genotype detected and the geographical origin , indicating that the reservoir of the pathogen is relatively fixed in space . In the present study , we report the co-circulation of two distinct M . ulcerans clonal complexes in the same BU endemic area over space and time . Since members of these two clonal complexes were closely related to strains from either the Densu river valley or the Asante Akim North district of Ghana , we conclude that a mobile reservoir of M . ulcerans may be involved in the occasional spread of the bacteria across different regions . | [
"Abstract",
"Introduction",
"Methods",
"Results/Discussion"
] | [
"ecology",
"and",
"environmental",
"sciences",
"medicine",
"and",
"health",
"sciences",
"rivers",
"pathology",
"and",
"laboratory",
"medicine",
"geographical",
"locations",
"tropical",
"diseases",
"genomic",
"library",
"construction",
"bacterial",
"diseases",
"signs",
"... | 2016 | Spatiotemporal Co-existence of Two Mycobacterium ulcerans Clonal Complexes in the Offin River Valley of Ghana |
Seasonal influenza epidemics cause consistent , considerable , widespread loss annually in terms of economic burden , morbidity , and mortality . With access to accurate and reliable forecasts of a current or upcoming influenza epidemic’s behavior , policy makers can design and implement more effective countermeasures . This past year , the Centers for Disease Control and Prevention hosted the “Predict the Influenza Season Challenge” , with the task of predicting key epidemiological measures for the 2013–2014 U . S . influenza season with the help of digital surveillance data . We developed a framework for in-season forecasts of epidemics using a semiparametric Empirical Bayes framework , and applied it to predict the weekly percentage of outpatient doctors visits for influenza-like illness , and the season onset , duration , peak time , and peak height , with and without using Google Flu Trends data . Previous work on epidemic modeling has focused on developing mechanistic models of disease behavior and applying time series tools to explain historical data . However , tailoring these models to certain types of surveillance data can be challenging , and overly complex models with many parameters can compromise forecasting ability . Our approach instead produces possibilities for the epidemic curve of the season of interest using modified versions of data from previous seasons , allowing for reasonable variations in the timing , pace , and intensity of the seasonal epidemics , as well as noise in observations . Since the framework does not make strict domain-specific assumptions , it can easily be applied to some other diseases with seasonal epidemics . This method produces a complete posterior distribution over epidemic curves , rather than , for example , solely point predictions of forecasting targets . We report prospective influenza-like-illness forecasts made for the 2013–2014 U . S . influenza season , and compare the framework’s cross-validated prediction error on historical data to that of a variety of simpler baseline predictors .
Seasonal influenza epidemics occur each year and incur significant economic burden , morbidity , and mortality . The annual impact in the United States has been estimated at 611K lost undiscounted life-years , 3 . 1M hospitalized days , 31 . 4M outpatient visits , and $87 . 1B in economic burden [1] . Accurate and reliable forecasts offer many opportunities to improve preparedness and response to influenza epidemics . Long-term predictions could be used to help select a vaccine for the next season . Forecasts within a season can help policy makers to tailor vaccination campaigns and advisories , hospitals to prepare staff and beds , and individuals and organizations to plan for vaccination and potential sickness . Despite the notable impacts of the disease , though , many weaknesses of influenza surveillance and prediction systems in the past [2] remain today . Capabilities to observe and forecast the prevalence of influenza and similar diseases lag considerably , e . g . , behind analogues in meteorology . During the 2013–2014 flu season , the Centers for Disease Control and Prevention ( CDC ) hosted the “Predict the Influenza Season Challenge” [3] , which encouraged teams to forecast features of the current epidemic progression that would be useful to policy makers , and to take advantage of digital surveillance such as search engine and social network data . The competition established a closer relationship between forecasters and policy makers , and provided valuable assessment of the performance of true ( prospective ) within-season forecasts . Existing work on modeling influenza epidemic curves generally falls into one of three categories: Compartmental models estimate the number of people in various states related to a disease [4] . For example , the SIR model approximates dynamics between the proportions of the population susceptible to influenza , infected with the virus , and recovered from infection . Common assumptions include that any pair of individuals in a population are equally likely to interact , and that different strains of influenza behave identically . Careful construction of compartmental models incorporating additional states and exogenous variables can improve on the results of more basic alternatives , and have outperformed alternative models in other settings [5 , 6] . Agent-based models generate synthetic populations based on census data and build complex schemes of interaction and disease behavior in synthetic humans [7–11] . It is common for these systems to be applied to the special case of a single , novel strain of influenza . Parametric statistical models are tools from time series modeling that are less closely tied with mechanistic assumptions of how flu is transmitted . Simple approaches include linear autoregression , which estimates flu activity at some time with a linear function of the flu activity in the recent past . A referee identified beta regression [12 , 13] as an alternative with observations constrained within the range of possible wILI values ( 0%–100% ) . More complex methods include generalized linear models ( GLM ) , Box-Jenkins analysis [14 , 15] , seasonal autoregressive integrated moving-average models [16] , and generalized autoregressive moving-average models [17] . Past forecasting efforts [18 , 19] usually take a compartmental model [20 , 21] , agent-based model [22] , or parametric statistical model [23–26] , and condition on partial data to predict flu activity levels one to ten weeks in the future . Other methods include prediction markets [27] , which combine expert predictions using a stock market-like system , and the method of analogues ( k nearest neighbors ) [28] , which makes predictions of future flu activity levels using similar patterns from the past , without assuming a strict model . The forecasting targets and ( sometimes qualitative ) evaluation metrics selected vary widely between works [18 , 19] , making it difficult to compare results for different methods . The 2013–2014 CDC challenge provided a standardized set of forecasting targets , allowing for some qualitative and quantitative comparisons on a single season . The contest winner [29] used an SIRS compartmental model approach [20 , 30 , 31] . We take a nonmechanistic approach , generating possibilities for the current season’s epidemic curve using modified versions of past seasons’ curves , incorporating adjustments in the timing , pace , and intensity of the epidemic informed by variability in historical data , and accommodating noise in observations . Our method models the process generating the data nonparametrically , using a large family of smooth curves to produce fairly close fits to historical data , relying on a small set of transformations of these curves to construct a probability distribution for the underlying level of ILI this season . While the method of analogues is similar in this regard as a nonparametric method , our framework considers the entire season as a unit and models observational noise , which differs from the traditional perspective in nearest neighbor modeling . Our framework outputs a distribution over epidemic curves , which can be used to produce histograms , credible intervals , and point predictions of the season’s onset , peak week , peak , and duration , as well as individual wILI measurements; existing applications of the method of analogues generate separate point predictions for each wILI measurement .
The forecasting framework is composed of five major procedures: Model past seasons’ epidemic curves as smoothed versions plus noise . Construct prior for the current season’s epidemic curve by considering sets of transformations of past seasons’ curves . Estimate what the wILI values in recent past will be after their final revisions , using non-final wILI and GFT . Weight possibilities for current season’s epidemic curve using estimates of final revised wILI . Calculate forecasting targets for each possibility , and report results . The first two steps only need to be executed once , at the beginning of the current season . As additional data becomes available throughout the season , we generate forecasts using steps 3–5 . We perform predictions for each geographical unit—the U . S . as a whole or individual HHS regions—separately . Historically , surveillance has focused on influenza activity between epidemiological weeks 40 and 20 , inclusive . We define seasons as epidemic weeks 21 to 39 , the “preseason” , together with weeks 40 to 20 . During the competition , data was available for 15 historical seasonal influenza epidemics . We excluded the 2009–2010 season from the data since it included nonseasonal behavior from the 2009 pandemic in the preseason . Additionally , there was partial data available for the 2013–2014 season .
For the CDC challenge , we generated biweekly forecasts from December 5 ( epidemiological week 49 ) to March 27 ( week 9 ) , for the nation as a whole , and individually for each the 10 HHS regions . Included below is a summary of our framework’s forecasts throughout the season , based on revised wILI data and no GFT . Fig 2 shows 10 draws from the posterior representing likely wILI curves , as well as the posterior mean and 5th and 95th posterior percentiles for the wILI value for each week . S1 Fig contains these forecasts for the entire 2013–2014 season , along with histograms and point predictions for the onset , peak week , peak height , and duration . Fig 3 shows ( A ) the observed national onset , peak week , peak height , and duration for the 2013–2014 season; ( B ) retrospective forecasts using revised wILI data only ( no GFT ) ; ( C ) real-time forecasts using wILI data only; and ( D ) real-time forecasts using both wILI and GFT . The real-time forecasts ( submitted to the CDC as part of the prediction challenge ) used older versions of the forecasting framework and wILI data . The small error in the onset before it occurred , as well as some of the error in peak week and height predictions , can be attributed to not factoring in holiday effects; at least some of these effects are smoothed out by the trend filtering process , or shifted to different times and heights by the peak week and height transformations . Later errors in the peak week , peak , and duration result from latching onto transformed versions of one or two past epidemic curves with two peaks . A forecasting method’s performance can vary greatly between seasons , so a single season provides limited evaluation power . We use leave-one-out cross-validation on historical data to provide a more stable estimate of the average point prediction error from retrospective forecasts . For each historical season scv , we produced forecasts using the rest of the historical seasons to build the prior , and recorded the average error of our point predictions across these 15 seasons for each week in the flu season . One detail to note is that these error estimates were generated using the final revision of the wILI data , and do not include any effects from approximating the most recent wILI values from the tentative values available in real-time . Fig 4 shows the cross-validated error for national point predictions of our current empirical Bayes framework , as well a few other approaches , for each for the four forecasting targets , aligned by the observed onset , peak week , or epidemiological week . S3 Fig shows these plots for both national and regional predictions , as well as the estimated accuracy and reliability of forecasts aligned by the point predictions for onset , peak week , and season end . S1 Table provides an alternative summary of the national cross-validation results , estimating the bias and variance of each forecasting approach , and aligning by epidemiological week . We use locally linear kernel regression [49] to estimate the mean absolute error when aligning by point predictions , GNU Parallel [50] to generate plots on multiple processors in parallel , and the xtable package [51] while typesetting the table . We display predictions of tar jr ( y r , s cv ) for a few methods . Baseline ( Mean of Other Seasons ) : takes the average target value across the 14 other seasons , completely ignoring any data from the current season; provides an idea of whether other forecasters provide reasonable levels of error at the beginning of the season , and how much they benefit from incorporating data from the season they are forecasting . Pinned Baseline ( Mean of Other Seasons , Conditioned on Current Season to Date ) : constructs 14 possible wILI trajectories for the current season by using the available observations for previous weeks and other historical curves for future weeks; reports the mean target value across these 14 trajectories; this is another very generic baseline that allows us to see the effect of using more complex wILI models and forecasting methods . Pointwise Percentile ( P2014 ) [44]: Constructs a single possible future wILI trajectory using the pointwise qth quantile from other seasons; estimates an appropriate value of q from the observed data so far , trying to match more recent observations more closely than less recent ones . k Nearest Neighbors ( knn ) : Uses a method similar to existing systems for shorter-term prediction [28] to identify k sections of other seasons’ data that best match recent observations , and uses them to construct and weight k possible future wILI trajectories . Empirical Bayes ( Transformed Versions of Other Seasons’ Curves ) : Our current framework , using transformed versions of other seasons’ curves to form the prior . Empirical Bayes ( SIR Curves ) : Our current framework , using scaled and shifted SIR curves rather than other seasons’ curves to form the prior; this is a somewhat similar approach to the SIRS-EAKF method used by the contest winner [20] . S5 Fig shows the fitted ( not forecasted ) SIR curves for national historical data , which were used to estimate a distribution over SIR , scale , and shift parameters , and S7 Fig shows two fits to regional data . Fig 4 indicates that , for all forecasting targets and most weeks , the average point prediction error for the EB method is similar ( overlapping error bars ) or lower than the average error for the best predictor for that target and week . An important feature of this approach is that it provides a smooth distribution over possible curves and target values , rather than just a single point . From this distribution , we can calculate point predictions to minimize some expected type of error or loss , build credible intervals , and make probabilistic statements about future wILI and target values .
The framework has a tendency to “latch” onto a particular shape in the mid to late season , forming predictions for the current season using transformed versions of a single past season . This phenomenon is undesirable in many cases , and was one motivation for using transformations of past curves , rather than just the curves themselves . S2 Fig illustrates how latching would be much more frequent and problematic if we did not use transformations . We find that latching occurs less frequently as more historical data becomes available , and S4 Fig shows that forecast error decreases as well . S2 Text discusses current limitations of the framework and future work , such as ways to improve forecasts by incorporating additional types of surveillance data ( e . g . , Twitter activity , thermometer sales , lab tests , weather , and vaccination data ) , dependencies between geographical units , and more accurate models of reporting behavior ( e . g . , by modeling holiday effects and improving the noise model ) . It should also be possible to automatically select what transformations and data to use by minimizing cross-validated prediction error on historical data . The strength of our non-mechanistic forecasting technique is that it relies more on the raw data and less on models of how that data came about . Current biological and epidemiological models of influenza , while grounded in 100 years of significant theoretical and mathematical development , still neglect or grossly simplify much that is not well understood or not yet well estimated , including subtype cross-protection , spatial dynamics , and demographic , behavioral and climatic conditions . Furthermore , since true influenza incidence is not known even in hindsight , we fall back on forecasting wILI . But to do so mechanistically , a large number of other processes must be understood and estimated , including the contributions of and interactions with non-influenza respiratory illnesses , the non iid nature of ILINet , variability in medical care seeking among both ILI and non-ILI patients , and more . In our opinion , these complications make non-mechanistic methods an attractive starting point for developing forecasting technology . The flip side of this approach is that our methods provide only modest insight into the biological and epidemiological processes underlying influenza . To boot: The usefulness of time-shifting the wILI curves confirms our intuition that the same week-on-week dynamics of influenza can be at play at different times of the year in different seasons . The usefulness of the Empirical Bayes approach in general suggests that the universe of wILI curves may well differ substantially from the conventionally parameterized compartmental models , likely due to the complex interaction of subtypes , the presence on non-influenza ILI , and the spread dynamics over the large regions involved . Our analysis revealed the Holiday Effect ( mentioned in the discussion of the distributional and point predictions for the 2013–2014 season , and visible in S5 , S7 and S8 Figs ) as a systematic and significant phenomenon in current wILI surveillance data . This effect consists of both a drop in the number of non-ILI office visits , as well as ( in some seasons ) a rise in the absolute number of ILI office visits , during the major holidays . While it is not unexpected that non-acute office visits are down during the holiday period , a deeper investigation of acute-care seeking behavior may be called for . Regardless , both these phenomena should be accounted for in any modeling or forecasting approach that uses this data . The fact that our method’s accuracy continues to improve with more historical seasons ( see S4 Fig ) suggests that the universe of wILI curves is not adequately sampled with 15 seasons , and that adding more seasons as they become available will likely further improve our method’s accuracy . Since the presented framework models epidemic curves rather than the underlying epidemiological process , it can be more readily applied to similar settings than complex mechanistic models which require adjustment based on some of the factors listed above . We have already used it to predict dengue incidence in the 2014 World Cup game cities with little modification [44] , and expect that application to additional diseases with semi-regular seasonal outbreaks would require little adjustment , and could be considered as a baseline for other , more specialized , predictors; it would not , however , apply to diseases with non-seasonal behavior , emerging diseases or invasion scenarios . While we provide detailed analysis of our method’s accuracy in the supporting information , to be useful to decision makers , the accuracy and reliability of any method must be distilled down to a few numbers . How should accuracy results be aggregated across regions , seasons , and time-of-forecast ? This is a non-trivial question for forecasting time varying events . Aggregating as function of time relative to the events occurrence ( as in S3A Fig ) is useful for analysis , but is not actionable because the time of the event is not known at forecast time . Aggregating as a function of absolute time ( as in S1 Table ) , while actionable , is not very useful for events whose timing varies considerably from season to season and from region to region . Perhaps the most useful way to aggregate accuracy results is by predicted-time-to-event ( as in S3B Fig ) . Whether these forecasts are already sufficient to influence action regarding influenza in the US or else must first be further improved is a question best left to public health officials , but it is our hope that by offering our methods and cross validated results we will both enrich the growing body of forecasting technologies and stimulate others to publish the results of their methods on these same test sets , targets , and metrics . | Influenza epidemics occur annually , and incur significant losses in terms of lost productivity , sickness , and death . Policy makers employ countermeasures , such as vaccination campaigns , to combat the occurrence and spread of infectious diseases , but epidemics exhibit a wide range of behavior , which makes designing and planning these efforts difficult . Accurate and reliable numerical forecasts of how an epidemic will behave , as well as advance notice of key events , could enable policy makers to further specialize countermeasures for a particular season . While a large amount of work already exists on modeling epidemics in past seasons , work on forecasting is relatively sparse . Specially tailored models for historical data may be overly strict and fail to produce behavior similar to the current season . We designed a framework for predicting epidemics without making strong assumptions about how the disease propagates by relying on slightly modified versions of past epidemics to form possibilities for the current season . We report forecasts generated for the 2013–2014 Centers for Disease Control and Prevention ( CDC ) “Predict the Influenza Season Challenge” , and assess its accuracy retrospectively . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [] | 2015 | Flexible Modeling of Epidemics with an Empirical Bayes Framework |
The long-term treatment outcome of visceral leishmaniasis ( VL ) patients with HIV co-infection is complicated by a high rate of relapse , especially when the CD4 count is low . Although use of secondary prophylaxis is recommended , it is not routinely practiced and data on its effectiveness and safety are limited . A prospective cohort study was conducted in Northwest Ethiopia from August 2014 to August 2017 ( NCT02011958 ) . HIV-VL patients were followed for up to 12 months . Patients with CD4 cell counts below 200/μL at the end of VL treatment received pentamidine prophylaxis starting one month after parasitological cure , while those with CD4 count ≥200 cells/μL were followed without secondary prophylaxis . Compliance , safety and relapse-free survival , using Kaplan-Meier analysis methods to account for variable time at risk , were summarised . Risk factors for relapse or death were analysed . Fifty-four HIV patients were followed . The probability of relapse-free survival at one year was 50% ( 95% confidence interval [CI]: 35–63% ) : 53% ( 30–71% ) in 22 patients with CD4 ≥200 cells/μL without pentamidine prophylaxis and 46% ( 26–63% ) in 29 with CD4 <200 cells/μL who started pentamidine . Three patients with CD4 <200 cells/μL did not start pentamidine . Amongst those with CD4 ≥200 cells/μL , VL relapse was an independent risk factor for subsequent relapse or death ( adjusted rate ratio: 5 . 42 , 95% CI: 1 . 1–25 . 8 ) . Except for one case of renal failure which was considered possibly related to pentamidine , there were no drug-related safety concerns . The relapse-free survival rate for VL patients with HIV was low . Relapse-free survival of patients with CD4 count <200cells/μL given pentamidine secondary prophylaxis appeared to be comparable to patients with a CD4 count ≥200 cells/μL not given prophylaxis . Patients with relapsed VL are at higher risk for subsequent relapse and should be considered a priority for secondary prophylaxis , irrespective of their CD4 count .
When visceral leishmaniasis ( VL ) occurs in HIV patients , it presents several challenges [1] . These include changes in clinical manifestations that may result in delayed diagnosis; changes in immunological response to the infection that affect the performance of diagnostic tools; and poor treatment response , in terms of low initial cure , relapse and mortality , due mainly to the combined effects of both infections causing profound immunosuppression [2 , 3] . The anti-leishmanial medicines available cannot completely eradicate the Leishmania parasites from the body [4] . Even those patients who are declared parasitologically cured at the end of treatment are , in reality , left with some parasites in the tissues that are not undetectable by microscopy [5] . In immunocompetent individuals , these are contained by cell mediated immunity , probably providing some degree of protection . However , in HIV patients , this small number of remaining parasites continues to replicate resulting in relapse of disease [6] , which then becomes more difficult to cure . These patients tend to have a persistent infection with flare-ups of clinical disease , described as active chronic disease [7] . Despite repeated treatment courses , such patients remain poorly responsive to treatment and deteriorate in their clinical and immune status . This leads to a high rate of failure to both VL and HIV treatments , and risk of death [8] . Hence , it is logical to introduce a maintenance therapy or secondary prophylaxis for this group of patients , as is done for other opportunistic infections in HIV , to continually suppress the multiplication of the parasite . Although secondary prophylaxis has been recommended in some international guidelines [9 , 10] , this is based on small studies from L infantum transmission in Europe [11 , 12] . In anthroponotic transmission regions like Eastern Africa , patients with persistent Leishmania parasites may serve as reservoirs of infection . Their need for repeated treatments with the limited available drugs increases the risk that they will be a source of emergent drug-resistant parasites [13] . Thus , first line drugs used to treat leishmaniasis in the region are not good options for use as secondary prophylaxis for fear of enhancing resistance development . The use of secondary prophylaxis for VL has not been routinely practiced in high endemic regions such as Northwest Ethiopia . Previous studies have demonstrated that the relapse rate of VL in HIV patients is 60–70% within a year of VL treatment [14 , 15] . Patients with low CD4 count and previous VL episodes were found to be at the highest risk of relapse . A recent prospective cohort study in Northwest Ethiopia has demonstrated 71% relapse free survival at one year among VL-HIV patients using pentamidine as a secondary prophylaxis for patients with CD4 <200 cells/μL or relapsed VL [16] . This is the only report on secondary prophylaxis for VL in the region . The objective of the current study is to document the long-term treatment outcomes of VL in HIV infected patients , namely the relapse-free survival , and to assess risk factors for relapse or death for up to one year after treatment for VL . Pentamidine was used as secondary prophylaxis for patients with CD4 cell count < 200 cells/μL after VL was successfully treated while those patients with CD4 ≥200 cells/μL were followed without secondary prophylaxis .
This was a prospective cohort study of parasitologically cured VL patients with HIV co-infection . It followed another randomised clinical trial ( NCT02011958 ) , that had a non-comparative design to evaluate the efficacy and safety of two treatment regimens for VL in HIV co-infected patients ( AmBisome total dose of 40 mg/kg and Ambisome total dose of 30 mg/kg + miltefosine 100mg/day/28days ) . In the preceding trial , one or two courses of the allocated treatment were given until patients achieved parasitological cure . One course of treatment was defined as the standard dose/duration of therapy with a specified anti-leishmanial drug e . g . a total dose of 40 mg/Kg of Ambisome monotherapy or 30 mg/kg of Ambisome plus 28-day course of miltefosine regimens . Patients who still had not reached cure after two cycles of VL therapy received rescue treatment at the discretion of the treating physician . At the end of VL therapy , patients who had a negative tissue microscopy result for Leishman-Donovan ( LD ) bodies were eligible for this prospective cohort study . The study was conducted in Northwest Ethiopia at two large leishmaniasis treatment centres: the Leishmaniasis Research and Treatment Centre at University of Gondar Hospital , supported by Drugs for Neglected Diseases initiative; and the Abdurafi Health Centre Médecins Sans Frontières Leishmaniasis Treatment Centre . Both are referral centres for complicated leishmaniasis cases . Each of these centres treats over 400 VL patients every year , of whom around 15–20% are co-infected with HIV . Most are adult seasonal migrant workers who travel from highland areas to work on large lowland farms where VL is endemic . Patients were enrolled from 14th August 2014 , and follow-up ended on 12th August 2016 . There were 55 patients who achieved parasitological cure with VL treatment ( s ) during the trial preceding this study . One patient was subsequently lost to follow-up and the remaining 54 patients were included in this cohort study ( Fig 1 ) . This research aimed to study the long-term outcomes of VL patients with HIV co-infection enrolled in the above-mentioned clinical trial . The sample size of the clinical trial was determined based on expected efficacy of the initial treatment , and as such provides a fixed sample size for this cohort study as the number of patients surviving to negative parasitology post treatment for VL . There were no pre-specified sample size calculations specifically for the cohort study objectives reported here . Patients who achieved parasitological cure but remained with a CD4 count below 200 cells/μL at the end of VL treatment were approached and consent was sought for pentamidine secondary prophylaxis . Patients with contraindications for pentamidine ( renal impairment , diabetes , known hypersensitivity ) were excluded from this intervention but were followed up as per their initial consent . Patients with CD4 cell counts above 200/μL were followed without secondary prophylaxis . Pentamidine isethionate ( Pentacarinat ) secondary prophylactic treatment was started one month after the negative test of cure ( completion of VL treatment ) . Every month , a dose of 4 mg/kg body weight of the salt reconstituted in 5 mL distilled water was re-diluted in 200 mL of 5% dextrose in saline or normal saline solution and infused over one hour with the patient in supine position . Patients were kept in the ward during administration and observed with frequent blood pressure monitoring for one hour before discharge . The blood glucose level was monitored prior to each infusion , and other metabolic panel tests ( blood sugar level , renal function , liver function , serum electrolytes ) were run every 6 months . Patients were offered continued ART , co-trimoxazole prophylaxis and adherence counselling . Follow-up arrangements for HIV care were made at ART clinics . Blood samples for HIV viral load were sent to a regional reference laboratory every six months . Arrangements for ART regimen changes were made when indicated in communication with the respective ART clinics . Follow-up started on the date negative parasitology was achieved and ended 390 days ( D390 ) after the initiation of the first VL treatment in the preceding trial . This means that those patients requiring more than one course of treatment to clear parasites during their VL treatment , and therefore achieving negative parasitology later , were followed up for less time than those who responded to one course of VL treatment ( Fig 1 ) . In particular , such patients would typically have less than one year of follow up in the cohort study . Patients eligible for pentamidine treatment were seen monthly for prophylactic treatment . There were two pre-specified follow-up time points for all patients , regardless of pentamidine prophylaxis; at 6 and 12 months after initiation of treatment for the current VL episode to check on long-term outcomes . They could present for unscheduled assessments as needed during the follow-up period ( e . g . relapse , intercurrent diseases , serious adverse events ( SAE ) ) . All SAEs , drug related adverse events ( AEs ) and any event that could lead to pentamidine interruption were documented during the follow-up period but other non-serious AEs were not systematically captured . Relapse-free survival by D390 was specified in the protocol as the primary outcome of this cohort study . Time at risk was defined as the time from negative parasitology following VL treatment to the earliest of the following events: i ) death , ii ) relapse , iii ) date of the D390 visit , and iv ) date last seen ( in case of loss to follow-up ) . Relapse-free survival was defined as reaching the end of the time at risk having neither died nor relapsed . Death ( i ) or relapse confirmed by positive parasitology ( ii ) correspond to a “failure” outcome . Conversely , alive and relapse-free at D390 ( iii ) or lost to follow up ( iv ) correspond to censoring . Relapse-free survival at one year was estimated using Kaplan-Meier methods to account for variable time at risk in all patients and within two sub-groups; i ) CD4 ≥200 cells/μL not receiving pentamidine secondary prophylaxis and ii ) CD4<200 cells/μL receiving pentamidine secondary prophylaxis . Although these groups were not defined a priori , risk factors were found to differ substantially between them . Graphical presentation of Kaplan-Meier analyses shows probability of failure rather than survival to illustrate timing of relapses and deaths in this cohort . Poisson regression was used to investigate univariable risk factors associated with relapse or death , again accounting for variable time at risk . This analysis was performed within the two sub-groups mentioned above . A multivariable model was built including those factors found to be associated with relapse or death in univariable analyses , i . e . with 95% confidence intervals ( CIs ) excluding the null effect of 1 . In principle a single model could be fitted to both subgroups by including interactions between subgroup and the risk factors . In practice , however , this was not possible due to small cell frequencies . Compliance to pentamidine in those who were eligible was calculated as the percentage of patients who received all monthly treatments for which they were eligible , i . e . until relapse , death or D390 in surviving relapse-free patients . Safety analyses comprised of summarizing the proportion of patients experiencing SAEs , an SAE related to pentamidine , and an AE that required discontinuation of pentamidine treatment . AE is defined as any untoward medical occurrence ( any unfavourable and unintended sign , symptom or disease , including an abnormal laboratory finding ) in temporal association with the use of the investigational treatment i . e . after the start of pentamidine . Causality relation to pentamidine is based on the known AEs listed in the Summary of Product Characteristics available from the manufacturer . Grading of the severity of the events was based on Common Terminology Criteria for Adverse Events ( CTCAE ) , Version 4 . 0 [17] . For events not described in CTCAE , severity of AE is graded as mild ( symptoms that did not require additional treatment ) , moderate ( symptoms that require additional treatment and get controlled ) , severe ( symptoms that require multiple treatments and may not resolve despite treatment ) . The definition of SAE was according to ICH-GCP guidelines ( life threatening events or events that led to disability , hospitalization , death , or congenital anomalies ) . The research protocol was approved by the Ethiopian regulatory authority ( Food , Medicine , Health Care Administration and Control Authority , FMHACA ) , the National Research Ethics Review Committee ( NRERC ) , the Institutional Review Board of the University of Gondar in Ethiopia , the Ethics Review Board of Médecins Sans Frontières , the London School of Hygiene and Tropical Medicine Research Ethics Committee , the Antwerp University Hospital Ethics Committee , and the Institute of Tropical Medicine , Antwerp Institutional Review Board . All patients were >18years old and were included into the study after written informed consent was given . Patients received all the required treatments free of charge including treatments for adverse events and intercurrent diseases . Food and transport support were provided .
Except for one female , all the patients were male migrant workers with a median age of 33 years . Among the 54 patients , 28 ( 52% ) were relapse VL cases , and 27 ( 50% ) were malnourished with body mass index ( BMI ) < 18 . 5kg/m2 . Most of the patients , 39 ( 72% ) , were already on antiretroviral treatment ( ART ) when VL was diagnosed with 27 ( 50% ) being on ART for six months or more . About two-thirds had a high Leishmania parasite load ( grade of +5 and +6 ) at VL diagnosis . Two-thirds of the patients had previously been treated with the Ambisome+miltefosine combination . Overall , 50% required more than one treatment course ( Table 1 ) . More than one course of treatment was required for 67% of patients on Ambisome monotherapy and 40% of patients on the combination regimen . Patients who required more than one treatment course received the same treatment twice , or at least one course of rescue treatment ( Table 2 ) . Of the 54 patients who were followed up , 22 had a CD4 count ≥200 cells/μL at the time of achieving negative parasitology . Of the 32 with CD4 count <200 cells/μL , 29 were started on pentamidine prophylaxis ( Table 1 ) . The other three had a contraindication , refused to participate in the prophylaxis or withdrew before the first pentamidine infusion , and are not included in the analysis of relapse-free survival or risk factors for relapse or death ( Fig 1 ) . One of them required multiple VL treatments and relapsed around 4 months after parasitological cure . The other two patients were followed up for 9 months and 12 months , respectively , without relapse . The type of VL ( primary or relapse ) , Leishmania parasite load , BMI , ART status and duration and the VL treatment regimen used were comparable between those with a CD4 cell count <200/μL who started pentamidine , and those with a CD4 cell count ≥200/μL who did not receive pentamidine ( Table 1 ) . Of those with low CD4 at the end of VL treatment ( <200 cells/μL ) , 62% of patients had received only one course of VL treatment , whereas in those with higher CD4 ( ≥200 cells/μL ) , it was lower at 32% . The percentage of patients who received all pentamidine treatments for which they were eligible was 76% ( Table 2 ) . The Kaplan-Meier estimates of the percentage of all patients with relapse-free survival at one year was 50% ( 95%CI: 35–63% ) ; 53% ( 95%CI: 30–71% ) in the 22 patients with CD4 ≥200 cells/μL and 46% ( 95%CI: 26–63% ) in the 29 patients with CD4<200 cells/μL who started pentamidine ( Fig 2 ) . The endpoint was reached because of death in one patient in the CD4 ≥200 cells/μL group . In the group with CD4 <200 cells/μL and on pentamidine , the endpoint was reached because of death in three patients , and two further patients died after reaching the endpoint due to VL relapse . If the events of interest are restricted to relapses , with pre-relapse deaths being considered as censored , then 57% ( 95%CI: 33–75% ) of the 22 patients with CD4 ≥200 cells/μL , and 54% ( 95%CI: 33–71% ) of the 29 patients with CD4 <200 cells/μL who started pentamidine were relapse-free at one year . Pentamidine treatment was not interrupted or stopped due to any adverse drug reaction . One expected SAE was reported: renal failure , considered possibly related to pentamidine that led to death . This patient was also experiencing pyelonephritis , sepsis , VL relapse and multiple myeloma which might have contributed to the renal failure , either directly or by toxicity of the concomitant medications ( Table 3 ) . Among patients with low CD4 counts ( <200 cells/μL ) at the time of VL cure and who started pentamidine prophylaxis , no statistically significant risk factors for relapse or death were identified ( Table 4 ) . In univariable analyses of patients with higher CD4 counts ( ≥200 cells/μL ) at the time of VL cure , higher rates of relapse or death were detected in relapse cases compared to primary cases , and in patients with normal BMI compared to low BMI ( <18 . 5kg/m2 , Table 4 ) . Patients previously treated with the combination regimen ( Ambisome+miltefosine ) for the VL episode had a lower rate of relapse or death , compared to those on Ambisome monotherapy ( Table 4 ) . The relation between relapse-free survival and the number of treatments required to clear parasites was also investigated . Of note , among those with CD4 cell count above 200 cells/μl at the end of treatment , there were no relapses or deaths in the subgroup for whom one course of treatment was sufficient to clear parasites . Moreover , such patients had all been treated with the Ambisome+miltefosine combination . The number of relapses being zero prevents the calculation of a rate ratio for the ≥200 cells/μL group by the end of treatment . Conversely , all patients with CD4 ≥200 cells/μL at the time of negative parasitology who received Ambisome monotherapy had required more than one course of treatment to clear parasites . In adjusted analyses , simultaneously accounting for the VL treatment regimen , relapse status ( primary vs relapse ) and BMI , only relapsed patients remained significantly associated with subsequent relapse or death ( adjusted rate ratio ( ARR ) = 5 . 42 , 95%CI: 1 . 1–25 . 8 , Table 4 ) .
The treatment of VL in HIV patients is complicated by a high rate of relapse in the first year after VL treatment [14 , 15] . Relapse leads to further immunosuppression , progression of HIV disease , predisposition to a number of opportunistic infections , failure of ART and death . Thus , it is important to comprehensively manage VL in HIV patients to ensure an effective initial treatment that is complemented by subsequent relapse prevention . Long-term outcomes of VL in HIV patients are described in this prospective cohort study , with the use of pentamidine secondary prophylaxis for those with CD4 cell counts <200/μL . Parasitologically cured VL patients with HIV co-infection were followed up for one year . The study was conducted in Northwest Ethiopia which is one of the highest HIV-VL co-infection regions of the world . The population is largely young adult males , with relapsed VL status and malnutrition accounting for about half of the cases respectively and patients at different time period on ART . The results are likely to be generalizable to the rest of Ethiopia and East Africa . The management of patients in this cohort included offering ART to all and secondary prophylaxis to those with a CD4 count <200 cells/μL after achieving parasitological cure of VL . Three-quarters of the patients had 100% compliance for the monthly pentamidine infusions . Low CD4 count is a known risk factor for relapse of VL [15 , 16] . Despite this care package , several VL relapses occurred , regardless of CD4 count at the end of VL treatment . This indicates the possibility of other factors influencing the long-term outcomes . The one death that was possibly related to pentamidine was due to acute renal failure in a patient with multiple co-existing diseases that can affect renal status . The strength of the study was that , as a continuation of a clinical trial , it complied with GCP , had substantial resources for intensive follow up , which helped ensure few missing data and consequently reduced bias . The probability of relapse-free survival at one year was 50% ( 95% CI: 35–63% ) in all patients . One limitation of this study was that it was not adequately powered to test differences between the two sub-groups; CD4 ≥200 cells/μL without pentamidine versus CD4 <200 cells/μL and receiving pentamidine prophylaxis . Another limitation was that the allocation of pentamidine was dependent on CD4 count which complicates the interpretation of the results and does not allow for unbiased estimation of an overall effect of pentamidine prophylaxis , or an overall effect of CD4 count . Interestingly , comparable probabilities of relapse-free survival were seen in the two groups , 53% ( 95%CI 30–71% ) and 46% ( 95%CI 26–63% ) respectively . A direct comparison of the two groups ( patients with CD4<200 given prophylaxis and patients with CD4>200 without prophylaxis ) based on a hypothesis of a difference was not included in our study as we may not expect a difference between these two groups if pentamidine can reduce the risk of relapse or death due to low CD4 , bringing the risk more in line with risk in those with higher CD4 and not on pentamidine , and we propose this as a hypothesis . According to previous reports , 60–70% of VL cases with HIV co-infection and on ART relapse within one year in the absence of secondary prophylaxis , the relapse rate being higher with a lower CD4 count [14 , 15] . The few existing reports on the use of secondary prophylaxis showed a relapse-free survival rate at one year ranging from 40–80% and a relapse rate among patients without secondary prophylaxis of 50–100% [11 , 18–21] . Most of these studies were case series studies from a region with L . infantum transmission , using different drugs and different follow-up patterns . In a recently conducted interventional cohort at the same study sites in Ethiopia , primary , relapse and past VL cases were enrolled . Pentamidine was given to those who achieved negative parasitology for the VL episode if their CD4 cell count was <200 cells/μL , and to all relapse cases regardless of CD4 count . The relapse-free survival rate at one year in all patients was 71% [16] . Looking more in depth at subgroups , the relapse-free survival rate for patients with CD4 <200 cells/μL ( n = 45 ) in this previous study was 68% ( 95%CI: 52%-80% ) and in those with CD4 ≥200 cells/μL ( n = 12 ) , it was 82% ( 95%CI: 45%-95% ) , with up to 12 months of pentamidine prophylaxis . There was no apparent difference in relapse-free survival observed in the groups with CD4 <200 cells/μL between the two studies , based on the overlap between the 95% confidence intervals around relapse-free survival in each study . The potential benefit of pentamidine amongst patients in the current study can be hypothesized to result from pentamidine treatment reducing the rate of relapse amongst those more likely to otherwise do so , to a rate closer to that observed in patients with a higher CD4 count . Although the use of secondary prophylaxis reduced the risk of relapse , there was still a notably high number of patients who continued to relapse in both groups ( CD4 <200 and CD4 ≥200 cells/μL ) . There are earlier reports showing relapse at higher CD4 count and also with secondary prophylaxis use [6 , 11] . Multiple factors may play a role in the risk of VL relapse . In this study , CD4 count , duration on ART , VL relapse status ( primary vs relapse ) , the Leishmania parasite load at diagnosis , the antileishmanial treatment regimen ( Ambisome monotherapy vs Ambisome+miltefosine combination ) , the duration of VL treatment and BMI were evaluated . In the group of patients with CD4 cell count <200/μL and receiving pentamidine , no additional independent risk factors were found to be associated with an increased rate of relapse . There is a need for additional , adequately powered studies to assess the risk factors for persistent immunosuppression in these patients . VL disease while on ART and low CD4 count is a WHO AIDS-defining illness and an indicator of advanced HIV disease and possibly undiagnosed ART failure [22 , 23] . For patients with profound immunodeficiency , ART , anti-Leishmania combination therapy and secondary prophylaxis seem insufficient to prevent relapse . The high relapse rate among these groups of patients indicates the need to explore other treatment modalities . These can be more frequent and/or higher doses of the prophylaxis or other interventions that can rapidly improve immunity [24] . Serum level of pentamidine was not checked in the study and the optimal prophylaxis dose is not known . Due to the differences in the formulation of pentamidine , the dose used in this study might be lower [25] . Beyond that , the goal should be for early detection and management of VL before profound immune deficiency sets in [26] . In the group of patients who had a CD4 cell count above 200/μl at the end of VL treatment ( stable on ART ) , past history of VL treatment ( relapse ) was an independent risk factor for subsequent relapse , as has previously been observed [15 , 27] . The current study reconfirms that patients with a previous history of relapse have a higher risk of relapse regardless of their CD4 cell count , adjusted rate ratio of 5 . 42 ( 1 . 14 , 25 . 8 ) . The small sample size in this group did not allow for further breakdown of the CD4 level and subgroup analysis . However , the findings indicate that CD4 level recommendation for secondary prophylaxis has to be re-visited and higher cut-off values be recommended [28] . Although the study was not adequately powered for a pre-specified magnitude of rate ratio , it is plausible to hypothesize a worse prognosis for patients whose VL episodes were hard-to-treat . Hard-to-treat patients are likely to include those who develop VL while on ART for more than 6 months , have a persistently low CD4 count ( <200 cells/μL ) after VL treatment ( advanced HIV disease ) , those who require longer treatment to cure VL , those with treatment failure for a combination regimen and previous history of VL treatment [15] . Because CD4 recovery can take time , those who received longer VL treatment ( negative parasitology at or after D58 ) were likely to achieve CD4 cell counts above 200/μL and thus become ineligible for secondary prophylaxis in this study , while they still fall in the category of hard-to-treat patients . It could be hypothesized that those patients would have benefited from the secondary prophylaxis , hence avoiding some of the observed relapses in the group with the higher CD4 cell count . In this study 50% of those with CD4 cell count ≥200cells/μl were relapsed patients who might have benefited from secondary prophylaxis . Patients who require prolonged initial VL treatment ( >1 month ) might be those with advanced disease and high initial parasite load , or those receiving less effective initial treatment ( e . g . Ambisome monotherapy ) . A highly effective initial VL treatment regimen ( e . g . combination therapy ) seems a more favorable approach , especially for hard-to-treat VL patients . Likewise , faster parasitological cure of VL and CD4 recovery ≥200 cells/μL ( cure by D29 ) can be a sign of mild disease and relatively well-preserved immunity . Of the seven patients who responded to initial treatment ( parasite clearance ) by D29 and had CD4 count ≥200 cells/μL ( thus without pentamidine prophylaxis ) , none relapsed or died . Although based on a very small number of patients , this could suggest that secondary prophylaxis may not be a priority for patients who respond to one course of treatment and achieve a CD4 level ≥200 cells/μL . This might be related to early diagnosis of VL in ART stable patients . On the other hand , the decision for secondary prophylaxis may need to be based on the CD4 level at the time of VL diagnosis rather than after VL treatment . In general , we have observed that the long-term outcome of VL in HIV patients is affected by multiple factors–importantly the level of immunity , history of VL treatment and the use of secondary prophylaxis . While a clear trend of benefit from the secondary prophylaxis is observed among those with low immunity , it should be noted that the treatment of these patients requires a multifactorial approach . Effective ART is a crucial component . Although the management of HIV included ART provision , clinical and CD4 cell monitoring; regular viral load determination was not possible due to the limited services available in the country during the study period and the delayed provision of results . The available data suggest that not all patients were on successful ART treatment and only a few had second line ART ( S1 , S2 and S3 Tables ) . Sustainable treatment of HIV-related opportunistic infection without effective ART is impossible . Integration of HIV treatment within the treatment programs of endemic opportunistic diseases is important for effective disease control [8 , 29] . HIV viral load results are important for patient management decisions and facilities in high endemic regions need to be upgraded with such services . In conclusion , the VL relapse rate in HIV co-infected patients is high irrespective of CD4 level . Secondary prophylaxis with pentamidine was found to be safe except in patients with risk factors for renal failure and could help prolong the disease-free survival of those with a CD4 count below 200cells/μL to a rate comparable to that for patients with CD4 count above 200 cells/μL and not receiving secondary prophylaxis . There are studies supporting its effectiveness and safety from both L . donovani and L . infantum regions [12 , 16 , 30 , 31] . However , the available data to date are based on small numbers of patients and from non-randomized studies , and are therefore below the ideal level of evidence needed to recommend implementation . Taking into consideration the high mortality and morbidity of VL-HIV co-infection and the urgent need for better management , we strongly recommend the use of secondary prophylaxis as an integral part of VL management in HIV . Priority cases for secondary prophylaxis are patients whose CD4 cell count remain <200/μL after effective VL treatment and those with a history of VL treatment ( VL relapses ) . A monthly infusion of pentamidine is a suitable option in terms of feasibility and safety , except for patients with renal diseases . Future prospective research studies could investigate alternative prophylactic regimens , different dosing and frequency to improve relapse-free survival , alongside new treatment approaches for hard-to-treat patients . | Achieving parasitological cure at the end of visceral leishmaniasis ( VL ) treatment in HIV co-infected patients does not assure definitive cure , as the disease will recur within a year in many patients . In this cohort study , the probability of relapse-free survival at one-year was 50% in all patients . The use of monthly pentamidine infusion for those with lower CD4 counts ( <200 cells/μL ) at the time of VL cure appeared to result in a comparable relapse-free survival rate to those patients with higher CD4 count ( ≥200 cells/μL ) who did not receive secondary prophylaxis . On the other hand , patients with a history of previous VL treatment ( VL relapse ) remained at high risk of relapse despite achieving CD4 count ≥200 cells/μL at the end of the VL treatment . While all VL patients with HIV co-infection may benefit from secondary prophylaxis , those with CD4 <200 cells/μL and previous history of treatment should be prioritized for secondary prophylaxis . New modalities for prevention of VL relapse in HIV patients should also be explored . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"medicine",
"and",
"health",
"sciences",
"pathology",
"and",
"laboratory",
"medicine",
"pathogens",
"tropical",
"diseases",
"microbiology",
"parasitic",
"diseases",
"parasitology",
"retroviruses",
"viruses",
"immunodeficiency",
"viruses",
"preventive",
"medicine",
"research... | 2019 | Long term outcomes and prognostics of visceral leishmaniasis in HIV infected patients with use of pentamidine as secondary prophylaxis based on CD4 level: a prospective cohort study in Ethiopia |
Zinc transporters play a critical role in spatiotemporal regulation of zinc homeostasis . Although disruption of zinc homeostasis has been implicated in disorders such as intestinal inflammation and aberrant epithelial morphology , it is largely unknown which zinc transporters are responsible for the intestinal epithelial homeostasis . Here , we show that Zrt-Irt-like protein ( ZIP ) transporter ZIP7 , which is highly expressed in the intestinal crypt , is essential for intestinal epithelial proliferation . Mice lacking Zip7 in intestinal epithelium triggered endoplasmic reticulum ( ER ) stress in proliferative progenitor cells , leading to significant cell death of progenitor cells . Zip7 deficiency led to the loss of Olfm4+ intestinal stem cells and the degeneration of post-mitotic Paneth cells , indicating a fundamental requirement for Zip7 in homeostatic intestinal regeneration . Taken together , these findings provide evidence for the importance of ZIP7 in maintenance of intestinal epithelial homeostasis through the regulation of ER function in proliferative progenitor cells and maintenance of intestinal stem cells . Therapeutic targeting of ZIP7 could lead to effective treatment of gastrointestinal disorders .
The intestinal epithelium , which renews every 3–5 days , is one of the most rapidly self-renewing tissues in adult mammals [1] . Homeostasis of the intestinal epithelium requires a fine balance between cell proliferation , migration , differentiation , and death [1] . Intestinal epithelial cells ( IECs ) are generated by intestinal stem cells , which are slender columnar cells that are interspersed with Paneth cells at the base of the intestinal crypt . Intestinal stem cells are characterized by expression of specific markers such as Lgr5 , Olfm4 , and Ascl2 [2–5] . They divide to form transit-amplifying ( TA ) cells , which are localized to the lower part of the crypt [2] . TA cells divide continuously , and the daughter cells differentiate into absorptive enterocytes and secretory cell lineages: goblet cells , enteroendocrine cells , and Paneth cells . Secretory epithelial cells have been shown to be sensitive to endoplasmic reticulum ( ER ) stress due to excessive protein synthesis of mucin and antimicrobial products [6 , 7] . Several mouse models with defects in protein folding or the unfolded protein response ( UPR ) exhibit enhanced ER stress in secretory cell lineages , which causes intestinal inflammation [6 , 8] . Furthermore , genetic mutation of the UPR transcription factor Xbp1 , which is required for maintaining secretory cell lineages , is associated with a risk for developing inflammatory bowel disease [6] . UPR may also be important in regulating the differentiation of intestinal epithelial stem cells since it is induced to resolve ER stress during the transition from stem to TA cell . Moreover , inducing excessive ER stress affects stemness [9] . These observations indicate that UPR signals from the ER are critical for maintaining epithelial homeostasis in the intestine . The ER acts as an intracellular store for biological mediators , including zinc , which is released into the cytoplasm in response to extracellular stimuli , such as IgE receptor cross-linking in mast cells [10 , 11] . Zinc deficiency causes ER stress in yeast and mammalian cells [12] , indicating that a subcellular abundance of zinc may be crucial for ensuring cellular homeostasis when ER stress is physiologically induced , particularly in the intestinal epithelium [9] . Specific zinc transporters are responsible for the spatiotemporal regulation of intracellular zinc storage . Zinc transporters are classified into two major families , SLC39A/ZIP and SLC30A/ZnT . The ZIP family of proteins facilitate zinc influx to the cytoplasm from extracellular spaces as well as intracellular compartments , including the ER , whereas the ZnT family of proteins transport zinc in the opposite direction [13] . Recently , we and others have demonstrated the involvement of SLC39A/ZIP family members in a variety of cellular functions , including cell proliferation , differentiation , survival , and migration [14–21] . Among ZIP family members , ZIP7 is an intracellular zinc transporter that localizes to the ER and is a potential target of Wnt/β-catenin [22 , 23] . It is notably upregulated in breast cancer cells as well as a gastric tumor model [22] . Considering that Wnt/β-catenin signaling governs vigorous TA cell proliferation as well as the maintenance of intestinal stem cells , we hypothesized that a ZIP7-mediated zinc signal might contribute to epithelial homeostasis . Here , we demonstrate that the ER-localized zinc transporter , ZIP7 , plays a key role in vigorous IEC proliferation , and in maintaining intestinal stemness , by alleviating ER stress in TA cells . Our data indicate that the fine-tuning of intracellular Zn homeostasis by ZIP7 is indispensable for epithelial proliferation and maintenance of stemness in the intestine .
To characterize ZIP7 distribution in the mouse intestine , we first analyzed the villous and crypt epithelium from the small intestine by quantitative PCR . Consistent with previous reports , Lgr5 [2] and Krt20 [24] were highly expressed in the crypts and the villi , respectively . Zip7 expression was enriched in the crypts ( Fig 1A ) ; this was confirmed by immunoblotting for ZIP7 proteins ( Fig 1B ) . In situ hybridization analysis demonstrated that Zip7 was distributed in the middle and lower crypt regions in a pattern similar to that of TA cells ( Fig 1C and S1 Fig ) . Multi-color FISH analysis demonstrated that Zip7 was positive for the EdU-incorporated TA cells at the lower part of crypt ( Fig 1D ) . Zip7 expression was also detected by the cells with typical Paneth-cell morphology represented by intracellular granules ( Fig 1E , arrows ) and Olfm4-positive cells at the bottom of crypts ( Fig 1F , arrows ) . Collectively , Zip7 was highly expressed in premature proliferative cells , stem cells , and post-mitotic Paneth cells , but its expression was lower in the villous epithelium . To investigate the role of ZIP7 in epithelial homeostasis , we generated a mouse line with floxed alleles of Zip7 ( Zip7flox/flox; S2A Fig ) . Zip7flox/flox mice were crossed with Villin ( Vil ) -CreERT2 Tg mice [25] to generate Zip7ΔIEC mice , in which the Zip7 gene can be deleted in IECs by administering tamoxifen ( Zip7ΔIEC ) . Adult Zip7ΔIEC mice and control littermates carrying a heterozygous floxed allele ( Zip7flox/+/Vil-CreERT2; referred to as Zip7Cont ) were given tamoxifen orally for five consecutive days . About half of the Zip7ΔIEC mice died by day 4 , and all were dead by day 7 ( Fig 2A ) . We subsequently treated the mice with tamoxifen for three consecutive days and performed histological examination at day 4 . In this protocol , more than 80% of Zip7ΔIEC mice remained alive . Deleting Zip7 impaired epithelial integrity and led to the loss of the proliferating compartment ( Fig 2B and 2C ) . TdT-mediated nick end labeling ( TUNEL ) assays revealed increased numbers of apoptotic cells in Zip7ΔIEC intestine ( Fig 2C ) . Crypt base columnar cells ( CBC cells ) marked by Lgr5 or Olfm4 are regarded as mitotically active intestinal stem cells and produce all epithelial cell lineages , including the proliferative progeny . Because of the loss of Ki67-positive cells in the crypts , we speculated that ZIP7 may affect the CBC population . In support of this notion , Olfm4+ stem cells at the bottom of crypts disappeared in Zip7ΔIEC mice within 3 days of tamoxifen treatment ( Fig 2D ) . These observations demonstrate that ZIP7 is essential for intestinal epithelial proliferation and maintenance of intestinal stem cells . To further corroborate the requirement of Zip7 for intestinal epithelial proliferation and maintenance of intestinal stem cells , independent of the niche , we established a crypt-derived organoid from Zip7Cont and Zip7ΔIEC mice , and examined the influence of Zip7 deficiency on organoid growth in an in vitro culture system . Using a conventional method [26] , small-intestinal crypts were isolated from Zip7ΔIEC and Zip7Cont mice , embedded in Matrigel , and cultured with 4-hydroxytamoxifen ( 4-OHT ) for 48 h ( S3 Fig ) . Crypts derived from control mice underwent crypt fission on day 2 , and eventually generated epithelial organoids with multiple budding crypts on day 6 ( Fig 2E ) [26] . In contrast , 4-OHT-treated crypts from Zip7ΔIEC mice failed to undergo crypt fission and died within 6 days ( Fig 2E and 2F ) . Following the 4-OHT treatment , Zip7ΔIEC organoids initially appeared morphologically intact , characterized by epithelial ( Day 1 in Fig 2E , blue asterisk ) and luminal ( Day 1 in Fig 2E , white asterisk ) compartmentalization and the presence of Paneth cells ( Day 1 in Fig 2E green arrowhead ) . Then , Zip7ΔIEC organoids generated cellular debris , which covered large areas of the surface of the organoids ( Affected organoid , Day 2 in Fig 2E , yellow arrowhead ) , and finally , Zip7ΔIEC organoids degenerated ( Day 6 in Fig 2E , yellow arrowhead ) . Ex vivo Zip7-deficient crypt cells are not able to self-renew or survive ( Fig 2F ) . These data indicate that ZIP7 is a prerequisite for the homeostatic regeneration of intestinal crypts . To further analyze the importance of Lgr5+ cell-intrinsic ZIP7 on intestinal stemness , we generated Lgr5EGFP-IRES-CreERT2/Zip7flox/flox ( Zip7ΔLgr5 ) mice . Lgr5+ stem cell-specific deletion of ZIP7 affected neither epithelial homeostasis ( S4A Fig ) nor crypt-derived epithelial organoid formation ( S4B Fig ) . Thus , these results suggest that Lgr5+ stem cell-intrinsic ZIP7 plays a redundant role in the homeostatic self-renewal under physiological conditions . Previous studies have demonstrated that Lgr5+ stem cells are dispensable for homeostatic self-renewal of intestinal epithelium [27 , 28]; however , this cell population is required for regeneration after damage such as irradiation [27] . We therefore asked the importance of ZIP7 in Lgr5+ stem cells in the regenerative process . Zip7ΔLgr5 and Zip7cont mice received 7 . 5-Gy whole body irradiation after tamoxifen treatment ( S4C Fig ) . While half of Zip7cont mice survived over 50 days after the irradiation , all of Zip7ΔLgr5 mice died within 4 days ( S4D Fig ) . Based on these observations , ZIP7 in Lgr5+ stem cells is required for intestinal regeneration after radiation damage . The bottom of small intestinal crypts contains post-mitotic Paneth cells juxtaposed to intestinal stem cells . Because Zip7 expression is higher in Paneth cells than in stem cells ( Fig 3A and 3B ) , we analyzed the influence of ZIP7 deficiency in Paneth cells . In situ hybridization analysis showed that Paneth cell markers , Defa1 and Mmp7 , were detected in Zip7ΔIEC mice after tamoxifen treatment ( Fig 3C ) . However , ultrastructural analysis revealed that Zip7-deficient Paneth cells had an abnormal morphology , characterized by destructive granules and a collapse of ER structure , and a part of Paneth cells underwent apoptosis ( Fig 3D ) . In 100 crypts chosen at random from three Zip7ΔIEC mice , more than 50% crypts displayed the complete loss of normal granule-containing Paneth cells and replacement with broken and/or dying Paneth cells ( Fig 3D ) . This suggests that Zip7 is indispensable for the maintenance of Paneth cells . Paneth cells support the stem-cell niche by providing factors that maintain stemness , such as Wnt3 , EGF and Notch ligands [29] and serve as essential niche cells in in vitro organoid models [30] . Crypt cells cannot form organoids in the absence of Paneth cells . It is therefore postulated that Paneth cell degeneration due to ZIP7 deficiency may affect self-renewal of organoids . To examine this postulation , the organoids are supplemented with Wnt3a , a Paneth cell-derived factor to maintain stem cells , which enable Paneth cell-null crypts to grow into organoids [30] ( S5A Fig ) . In Zip7-sufficient control crypts , exogenous Wnt3a caused the organoids to form enlarged , round cysts due to overgrowth ( S5B Fig ) . However , exogenous Wnt3a did not support the growth of Zip7-deficient crypts , and failed to rescue crypts from cell death ( S5B Fig ) . Therefore , we reasoned that the loss of self-renewal capacity is unlikely to be due to Paneth cell degeneration under physiological conditions in Zip7ΔIEC mice . TA cells localize above the stem cell niche and vigorously proliferate before differentiation [2] . Transmission electron microscopy showed apoptotic bodies in the proliferative crypt compartment in the Zip7ΔIEC small intestine 3 days after tamoxifen injection ( Fig 4A , left panel ) , indicating that Zip7 deficiency induced apoptosis in TA cells . Notably , apoptotic bodies were present inside CBC cells that were morphologically recognized as stem cells by a slender columnar shape interspersed with Paneth cells ( Fig 4A , right panel ) . These observations raised the possibility that CBC cells may engulf neighboring dead cells as reported previously [31] , although further study is necessary to confirm this possibility . To clarify the intracellular events by which Zip7 deficiency causes apoptosis of TA cells , we purified cell populations , based on EGFP expression intensity , from control [Zip7ContLgr5EGFP-IRES-CreERT2] and Zip7ΔIECLgr5EGFP-IRES-CreERT2 mice 2 days after tamoxifen treatment ( Fig 4B ) . The stem cell markers Lgr5 , Ascl2 , and Olfm4 were exclusively expressed in the Lgr5hi stem cell population and were downregulated in the Lgr5int TA cell population ( S6A Fig ) . Comparative gene expression profiling of intestinal TA cells showed that many UPR-related genes were more highly expressed in the Zip7-deficient TA cell population ( Fig 4C and S6B Fig ) . Severe ER stress conditions activate apoptosis signaling through induction of pro-apoptotic factors such as Trib3 and Bcl2l15 . We also detected upregulation of these genes in the Zip7-deficient TA cells ( S6B Fig ) . These results suggest that TA cell apoptosis can be attributed to severe ER stress as a result of Zip7 deficiency . Gene ontology-based functional enrichment analysis also revealed that a group of ER function-related genes was upregulated in the Zip7-deficient TA cell population ( Fig 4D ) . In agreement with these observations , Zip7 ablation led to upregulation of the ER stress-related genes Derl3 and Creld2 , as well as Ddit3 ( encoding CHOP ) , which is responsible for an ER stress-dependent apoptotic pathway in cultured organoids ( Fig 4E ) . Activation of ATF6 and the PERK pathway induces expression of transcription factor XBP1 ( X-box-binding protein 1 ) [32] . Increased expression of Xbp1 was observed in Zip7ΔIEC organoids in response to exposure to 4-OHT ( S7A Fig ) . Notably , Mt1 whose expression depends upon intracellular zinc concentration was downregulated in Zip7ΔIEC organoids ( S7B Fig ) , suggesting that the intracellular zinc may be decreased in the organoid . We therefore examined whether exogenous supplementation of zinc could ameliorate ER stress caused by Zip7 deficiency; however , the zinc supplementation failed to suppress the induction of ER-stress-related genes ( S8A Fig ) . Collectively , these results indicate that ablating ZIP7 augments ER stress and eventually induces apoptosis in TA cells . Therefore , ZIP7 ensures ER function in TA cells , which is essential for maintaining the vigorous proliferative properties of TA cells . Importantly , the same effect was seen in Zip7-deficient Lgr5+ stem cells ( S6C Fig ) , suggesting that ZIP7 is also a key regulator of ER function in intestinal stem cells . We next explored the molecular mechanism by which ZIP7 ameliorates ER stress . ImmunofluoZIP7 was mainly co-localized with the ER-resident protein , PDI ( Protein disulfide-isomerase ) , at the perinuclear region ( Fig 5A ) . Subcellular fractionation by sucrose gradient-based ultracentrifugation confirmed that ZIP7 predominantly fractionated with the ER ( Fig 5B ) ; this localization raised the possibility that ZIP7 was involved in maintenance of ER homeostasis . To examine this possibility , we prepared mouse embryonic fibroblasts ( MEFs ) from Zip7flox/flox/Rosa26-CreERT2 mice ( Zip7-/- in Fig 5C and 5D ) , in which Zip7 can be deleted with 4-OHT treatment ( S2B and S2C Fig ) . Comparative gene expression profiling of MEF cells demonstrated that UPR-related genes , such as Derl3 , Slc2a6 , Creld2 , Herpud1 and Ddit3 ( encoding CHOP ) , were upregulated in Zip7-/- MEF cells ( S9 Fig ) . Quantitative PCR analysis confirmed that the upregulation of the UPR-related genes ( Fig 5C , S8B and S9 Figs ) . Exogenous zinc supplementation did not affect the upregulation of UPR in Zip7-/- MEF cells ( S8B Fig ) as observed in Zip7ΔIEC organoids ( S8A Fig ) . On the other hand , exogenously expressed wild-type ZIP7 rescued UPR caused by the loss of ZIP7 ( Fig 5D ) . Intriguingly , Zip7 expression increased in response to tunicamycin- or thapsigargin-induced ER stress ( Fig 5E ) , suggesting that ZIP7 supports homeostatic ER functions and protects against ER stress by regulating UPR-related genes .
Here , we demonstrate that the ER-localized zinc transporter , ZIP7 , is indispensable for the vigorous proliferation of TA cells , and for maintaining the stemness of intestinal stem cells . Specifically , ZIP7 was abundantly expressed in the crypt epithelium , and deleting Zip7 led to the loss of both TA and stem cells as well as to increased ER stress responses and UPR in crypts . The UPR , which is triggered by high bioenergetic demands , is important in helping TA and stem cells adapt to the demands of biosynthesis and in preventing ER stress-induced apoptosis . Therefore , our study demonstrates that ZIP7 controls crypt homeostasis and strongly contributes to ER function in TA cells ( Fig 6 ) . We observed massive apoptotic cell death in the middle of crypts in Zip7ΔIEC mice , where apoptotic bodies derived from TA cells had been incorporated into stem cells ( Fig 4A ) . Studies show that irradiation- or chemotherapy-induced apoptotic cells are incorporated into neighboring stem cells [31 , 33] , and cells undergoing apoptosis generate signals that induce apoptosis in neighboring cells , leading to cohort cell death [34 , 35] . Given these observations , deleting Zip7 may cause massive apoptosis in crypt cells , particularly TA cells , with the apoptotic bodies being taken up by adjacent viable stem cells . The apoptosis of neighboring cells and accumulation of apoptotic bodies appear to be harmful for stem cells , as evidenced by their upregulation of ER stress molecules . These findings suggest that stem cells may be lost since a large number of neighboring cells are undergoing apoptosis due to Zip7 deletion . Deleting Zip7 induces the expression of UPR genes ( Fig 4C and 4E and S6B Fig ) , indicating that the failure of Zip7-deficient TA cells to resolve ER stress leads to the upregulation of UPR genes and to apoptosis . UPR signaling is associated with IEC proliferation and differentiation . For example , activation of UPR signaling promotes the differentiation of stem cells into TA cells [9] . XBP1 , a key component of the ER stress response , is required for secretory cell development and maintenance [6] . ER stress regulators and UPR signaling molecules are upregulated in TA cells compared to stem cells [9] . Notably , TA cells strongly expressed ZIP7 , which was upregulated by ER stress ( Fig 5E ) . UPR-dependent upregulation of ZIP7 in TA cells might resolve ER stress , thereby securing the highly proliferative status of TA cells . These results suggest that zinc concentration in the ER is tightly regulated by ZIP7 , and in TA cells , this regulation is a prerequisite to resolving ER stress . Lgr5+ stem cells are well documented to show resistance to irradiation , because stem cells can repair DNA damage more efficiently than radio-sensitive TA cells and the other epithelial cell types of the small intestine [24 , 36] . Indeed , repopulation after radiation damage begins with CBCs , including Lgr5+ stem cells [36] . Depletion of Lgr5+ stem cells results in an impaired regenerative response after irradiation [27] . Thus , Lgr5+ stem cells are a prerequisite of the regenerative response for providing TA cells after epithelial damage . Nevertheless , Zip7 deficiency in Lgr5+ stem cells remarkably increased susceptibility to irradiation due to a failure in regenerative responses . Our results suggest that ZIP7 critically contributes to the Lgr5+ cell-dependent regenerative response following irradiation . On the contrary , Zip7 deficiency did not affect stemness without irradiation , implying that Zip7 is dispensable for maintenance of stem cells under the physiological conditions . It is noteworthy that Lgr5+ stem cells play a redundant role in homeostasis of the epithelium under physiological conditions , because Bmi1+ stem cells serve as an alternative stem cell pool[28] . We , therefore , do not formally exclude the possibility that Bmi1+ stem cells may compensate degeneration of Zip7-deficient Lgr5+ stem cells to maintain stemness . We have also demonstrated that ZIP7 is essential for maintenance of Paneth cells that are susceptible to UPR . Ablation of Xbp1 required for expansion and maintenance of the ER results in loss of Paneth cells [6] . We found that Zip7 deletion induced ER stress in intestinal organoids , and Zip7 was upregulated by ER stress induction . Paneth cells contain a large amount of zinc in their granules[37] , and perturbation of zinc homeostasis elicits extrusion of Paneth cells[38] . Furthermore , upregulation of Zip7 expression is associated with the development of Paneth cells [39] . Collectively , ZIP7-mediated regulation of zinc homeostasis seems to be responsible for maintenance of Paneth cells . Because Paneth cells serve as niche cells for intestinal stem cells by providing EGF , Wnt and Notch ligands[29] , this cell population is required for the in vitro organoid culture system that lacks mesenchymal cells , the major producer of Wnt and Notch ligands in the intestine , as evidenced by the observation that the Paneth cell-null crypts fail to develop to organoids[30] . This abnormality can be rescued by exogenous Wnt3a supplementation[30] . We observed that Wnt3a supplementation did not improve the growth arrest in Zip7ΔIEC organoids . This observation supports the notion that loss of intestinal stem cells in Zip7ΔIEC organoid was unlikely attributed to degeneration of Paneth cells . Although the precise mechanisms by which ZIP7 resolves ER stress in the intestinal epithelium are unclear , several key phenomena and predictions may help answer this question . For instance , ZIP7 is needed to maintain sufficient zinc concentration in order to meet the requirements of ER-resident proteins that use zinc as a cofactor , such as ER-localized calreticulin , which binds monoglucosylated carbohydrate on newly synthesized glycoproteins [40 , 41] . In addition , the flexible coordination geometry of zinc [42] and facile ligand exchange allow it to bind proteins stochastically to substitute for a different native metal ion [43] . Therefore , stochastic binding of zinc to ER-resident proteins could negatively affect protein conformation and function . For example , protein disulfide isomerase , which regulates protein folding by catalyzing disulfide bonds in the ER , is oligomerized in the presence of zinc and thus has reduced catalytic activity [44] . Excessive zinc accumulation in the ER may lead to ectopic binding to various ER proteins , thus impairing their structure and function . Misfolded proteins are transported across the ER membrane for cytosolic proteasome degradation in a process known as ER-associated degradation ( ERAD ) [45 , 46] . Key ERAD components include E3 ubiquitin ligases that are embedded in the ER membrane [47] . Most ERAD E3 ligases possess a zinc-coordinating RING domain to facilitate E2-dependent ubiquitylation [47] . The formation of a rigid , globular platform for protein-protein interactions in RING fingers requires zinc [47] , implying that fine-tuning zinc concentration by ZIP7 seems to be essential for normal ER function . UPR activation induces the expression of ERAD components . Similarly , Zip7 expression was elevated in cells treated with tunicamycin or thapsigargin to induce UPR ( Fig 5E ) , suggesting that ZIP7 may be involved in an adaptive program to alleviate ER stress . ZIP7 is reported to be phosphorylated and activated by casein kinase 2 ( CK2 ) , an ER-localized serine/threonine protein kinase [48] . CK2 also plays a key role in the ER stress response [49] . Therefore , the CK2-ZIP7-zinc signaling axis may moderate the strength of UPR signaling . In this study , we illustrate that ZIP7 is fundamental to the homeostatic and active proliferation of the intestinal epithelium and the maintenance of stem cells in the crypt . Our findings also indicate that ZIP7 is a potential therapeutic target for gastrointestinal tumors . Given that various cells proliferate vigorously under certain physiological and pathological conditions , our study represents an important step toward uncovering the mechanisms by which cells adapt their intracellular conditions to maintain a brisk proliferative response through fine-tuning ER function .
All mice were housed and cared for according to guidelines approved by the RIKEN Yokohama institutional Animal Care and Experiments committee ( K24-007 ) , or by the Committee for Institutional Animal Care and Use at the University of Toyama ( A2015MED48 ) . All animal experiments were performed with the approval of the institutional Animal Care and Use committee of RIKEN IMS , and the Graduate School and Pharmaceutical Sciences , University of Toyama . Zip7flox/flox mice were generated as previously described [50] . Briefly , we created a targeting vector to eliminate the genomic region encompassing exons 5 and 6 by inserting a loxP sequence and a Neo cassette into the region between exons 4 and 5 , and a loxP sequence between exons 6 and 7 of Slc39a7 . After this vector was introduced into R1 ES cells , cloned homologous recombinants were selected with antibiotics , and the genotypes were verified . We developed chimeric mice with the targeted ES cell clones and crossed them with E2a-Cre mice to delete the Neo cassette ( S2 Fig ) . The mice used in experiments were backcrossed with C57BL/6J mice . Villin ( Vil ) -CreERT2 transgenic mice were described previously [25] . Lgr5EGFP-IRES-CreERT2 transgenic mice were obtained from Jackson Laboratories . Rosa26-CreERT2 transgenic mice were purchased from Artemis Pharmaceuticals . The transgenic Vil-CreERT2 or Rosa26-CreERT2 mice were crossed with Zip7flox/flox mice . To sort cells from Zip7Cont and Zip7ΔIEC mice based on Lgr5 levels , Lgr5EGFP-IRES-CreERT2 transgenic mice were crossed with Zip7flox/+/Vil-CreERT2 mice or Zip7flox/flox/Vil-CreERT2 transgenic mice . Male and female mice were selected randomly for all experiments . In vivo , the Cre enzyme was activated by orally administering 100 μL tamoxifen ( 50 mg/mL ) dissolved in corn oil/ethanol ( 9:1 ) for the indicated number of consecutive days . Genotyping was done by PCR using KOD-plus or KOD-FX NEO polymerases ( Toyobo ) . The following primers were used for genotyping: 5’-CTTCATGCTTTACTGCCTCCGTTCC-3’ and 5’-ATAAATCCGCCTGCAGTGAA-3’ . We purchased antibodies against Ki67 ( Clone: MM1 ) , Villin ( Santa Cruz , sc-7672 ) , Gapdh ( Clone: 6C5 ) , Calnexin ( Clone: 37 ) , GM130 ( Clone: 35 ) , Adenine nucleotide translocase ( Santa Cruz , sc-9299 ) , Lyn ( Santa Cruz , sc-15 ) , GP130 ( Santa Cruz , sc-9045 ) , Tubulin ( Clone: B-5-1-2 ) , EpCAM-PE ( Clone: 9C4 ) , CD24-APC ( Clone: ML5 ) , CD45-PE-Cy5 ( Clone: 30-F11 ) , and PDI ( Clone , 1D3 ) . We generated the antibody against mouse ZIP7 by immunizing a rabbit with a peptide corresponding to the GNTGPRDGPVKPQSPEE sequence of mouse ZIP7 , which was affinity-purified by using the antigen peptide . Mouse small intestines were isolated , opened longitudinally , and washed with cold PBS . The tissue was chopped into pieces approximately 5 mm in length and washed again with cold PBS . Tissue fragments were incubated in 1 mM dithiothreitol with Hank’s balanced salt solution ( HBSS; Life Technologies ) for 15 min on ice . After removing the dithiothreitol solution , the tissue fragments were washed with cold PBS and then with 2 mM EDTA with HBSS for 5 min , after which the samples were incubated in 2 mM EDTA with HBSS for 30 min on ice . Samples were then shaken vigorously , after which the supernatant was discarded . The sediment was washed into 5 mM EDTA with HBSS for 5 min on ice , incubated in 5 mM EDTA with HBSS for 30 min on ice , and shaken vigorously to yield free crypts . This fraction was passed through 100-μm and 70-μm cell strainers and then centrifuged at 150 g for 3 min . The isolated crypts were plated on Matrigel with culture medium ( Advanced DMEM/F12 , Invitrogen ) containing the growth factors: EGF ( 50 ng/mL , Peprotech ) , R-spondin1 ( 500 ng/mL , R&D Systems ) , and noggin ( 100 ng/mL , Peprotech ) . The growth medium was changed every other day . In some experiments , Wnt3a-conditioned medium was added to the culture medium . For in vitro Cre activation , organoids were cultured with 1 μM 4-OHT for 48 h , after which the culture medium was replaced with fresh medium without 4-OHT . Three or more independent experiments were performed . For quantification , the morphology of organoids was classified as follows: 1 ) Unaffected organoid with clearly distinguishable epithelial and luminal area under bright field; 2 ) Affected organoid that has cellular debris on a large part of the organoid surface and that is difficult to identify the luminal area inside the organoid; 3 ) Degenerated organoid that mostly consists of cellular debris and has no epithelial/luminal compartments . Intestinal samples were fixed and embedded in paraffin using standard protocols . Sections were deparaffinized in xylol and dehydrated in ethanol . Antigens were retrieved with citrate or EDTA buffer for 2 min at 105°C in an autoclave , and endogenous peroxidase was blocked . Primary antibodies were incubated overnight at 4°C , and secondary , horseradish peroxidase-conjugated antibodies were detected using a 3 , 3’-Diaminobenzidine Peroxidase Substrate Kit ( Vector Laboratories ) . In situ hybridization for Zip7 , Olfm4 , and Defa1 was performed by Genostaff ( Tokyo , Japan ) methods or by one of the methods described next . For labeling proliferating cells , mice were administered with EdU ( 5 mg/kg of body weight ) by intraperitoneal injection . After 2 hours , the mice were sacrificed and the intestine was removed . Intestinal epithelial monolayer was isolated by modifying the method described by Kimura et al [51] . Briefly , the intestine were soaked in ice-cold Hank's balanced salt solution ( Life Technologies ) containing 30 mM EDTA and 5 mM dithiothreitol . After incubation with gentle shaking for 20 min on ice , epithelial monolayer was carefully separated from muscle layer by manipulation with a fine needle under stereomicroscopic monitoring . The isolated epithelium was fixed for 20 min in 4% paraformaldehyde in PBS at 4°C . The specimens were washed three times with 50 mM glycine in PBS , and then pretreated with 0 . 3% Triton X-100 in PBS for 10 min . Subsequently , EdU detection was carried out using the Click-iT EdU AlexaFluor488 imaging kit ( Life Technologies ) according to the manufacturer's protocol . Then , FISH was performed with the Quantigene View RNA ISH Cell Assay ( Affymetrix , Santa Clara , CA ) according to manufacturers' protocols . Specific oligonucleotide probe sets against Slc39a7 ( Zip7; VB6-14266 ) were purchased from Affymetrix , Inc . After FISH , samples were treated with an anti-E-cadherin ( dilution 1:400 in PBS; AF748 , R & D systems ) goat polyclonal antibody overnight at room temperature and incubated with Cy3-conjugated anti-goat antibodies ( 1:1000; Life Technologies ) for 2 h at room temperature . Specimens were mounted with SlowFade Gold Antifade Reagent ( Life Technologies ) and examined under a confocal laser microscope ( FV1000 , Olympus , Tokyo , Japan ) . Expression of Zip7 mRNA in the intestinal stem cells were analyzed by dual-color FISH with Slc39a7 and Olfm4 probes ( VB1-11400 ) by similar procedure with the exception of EdU-labeling and detecting step . Mice were placed under deep anesthesia and perfused via the aorta with physiological saline followed by 2 . 5% glutaraldehyde in 0 . 1 M phosphate buffer at pH 7 . 4 . The dissected tissues were immersed in fixative for another 3 h , briefly washed in phosphate buffer , post-fixed in 1% OsO4 for 90 min , dehydrated through a graded ethanol series , and embedded in Epon . Ultra-thin sections were prepared on an ultramicrotome , stained with uranyl acetate and lead citrate , and examined under an electron microscope ( H-7100 , Hitachi , Tokyo , Japan ) . The murine Zip7 gene was cloned by PCR using cDNA from the C57BL/6J mouse strain , sequenced on a 3130xI sequencer ( ABI-PRISM , Applied Biosystems ) , and subcloned into the pcDNA6 . 2/V5-DEST expression vector ( Invitrogen ) to generate C-terminally V5-tagged mouse WT ZIP7 . These constructs were transfected into 293T cells by Lipofectamine LTX reagent ( Life technologies ) according to manufacturer protocols . 293T and MEF cells were maintained in DMEM or RPMI1640 medium containing 10% fetal bovine serum , penicillin , and streptomycin . The transfected cells were fixed , permeabilized , and stained with monoclonal anti-V5 and Dylight 488-conjugated anti-PDI antibodies . Anti-V5 was detected with polyclonal Alexa 546-conjugated goat anti-mouse IgG ( Life technologies ) . Hoechst 33342 was used to visualize nuclei . Images were acquired by LSM780 confocal microscope ( Zeiss ) RNA was extracted with the RNAeasy Kit ( Qiagen , UK ) or Trizol ( Life Technologies ) or Sepazol ( Nacalai Tesque ) according to the manufacturers’ instructions . First-strand cDNA was synthesized by ReverTra Ace ( Toyobo ) according to the manufacturer’s instructions; mRNA levels were quantified by qPCR using SYBR Premix Ex Taq ( Takara ) or PowerUp SYBR Green Master Mix ( ThermoFisher Scientific ) and the ABI3100 or StepOnePlus system ( Applied Biosystems ) or MX3000p system ( Stratagene ) , and were normalized to Gapdh . The sequences of the primers are provided in S1 Table . Isolated crypts , villi , or cells were lysed in radioimmunoprecipitation assay buffer . Lysates were clarified by centrifugation ( 30 min , 12 , 000 rpm , 4°C ) and proteins were resolved by SDS-PAGE , electrotransferred onto polyvinylidene fluoride membranes , and immunoblotted . Horseradish peroxidase-bound secondary antibodies were detected with a chemiluminescence kit . Cells in HTE buffer ( 0 . 25 M sucrose , 10 mM Tris-HCl , 0 . 1 mM EDTA , pH 7 . 4 ) supplemented with protease inhibitor cocktail ( Roche ) were lysed by freeze-thawing . Nuclei and cellular debris were removed by centrifugation at 1 , 400 g for 10 min . Postmitochondrial supernatant obtained by centrifuging at 15 , 000 g for 10 min was layered on a sucrose step-gradient consisting of 1 . 3 M , 1 . 5 M , and 2 . 0 M sucrose in 10 mM Tris-HCl ( pH 7 . 6 ) and banded by centrifugation at 152 , 000 g for 70 min . The ER fraction at the interface between the supernatant and the 1 . 3 M sucrose step was collected , diluted with HTE butter , and pelleted by centrifugation at 126 , 000 g for 45 min . The ER membranes were resuspended in HTE buffer . The Flag-tagged mouse ZIP7 plasmid was inserted into a pMX retroviral vector ( a gift from T . Kitamura , The University of Tokyo , Tokyo , Japan ) . This construct was used to transfect the 293T-based Phoenix packaging cell line ( a gift from G . Nolan , Stanford University , Stanford , CA ) . Lipofectamine 2000 ( Invitrogen ) was used to generate recombinant retroviruses . MEF cells were infected with the retrovirus in the presence of 10 μg/mL polybrene . Small intestinal crypts were isolated by EDTA chelation . Epithelial cells were dissociated using TrypLE express ( Invitrogen ) with 2 , 000 U/mL DNase I for 30 min at 37°C . Dissociated cells were passed through a 40-μm strainer , stained with anti-EpCAM-PE , anti-CD45-PE-Cy5 , and anti-CD24-APC antibodies , and then analyzed by MoFlo . Dead cells were excluded by 7-AAD or SYTOX-AAD staining . Viable single cells were gated , and CD45- CD24+ SSChi , CD45- CD24- EpCAM+ Lgr5-EGFPhi , and CD45- CD24- EpCAM+ Lgr5-EGFPint cells were sorted . RNA was isolated from the sorted Lgr5-EGFPhi EpCAM+ CD45- and Lgr5-EGFPint EpCAM+ CD45- cell fractions from the small intestines of Zip7flox/+/Vil-CreERT2/Lgr5EGFP-IRES-CreERT2 or Zip7flox/flox/Vil-CreERT2/Lgr5EGFP-IRES-CreERT2 mice . Microarray experiments were performed using Affymetrix Mouse Gene 1 . 0 ST Array GeneChips . Datasets were derived from three biological samples of each genotype . All microarray data have been deposited in the reference database of immune cells ( RefDIC , http://refdic . rcai . riken . jp ) under accession numbers: RSM13249 , RSM13250 , RSM13251 , and RSM13252 . Differences in the means were examined by a 2-tailed unpaired Student’s t-test . Analyses were done on GraphPad Prism 6 ( GraphPad Software ) . Results are presented as mean ± SEM or SD for the number of experiments indicated in each figure legend . Kaplan-Meier survival curves were compared using the log-rank test . P < 0 . 05 was considered statistically significant . | Intestinal epithelium undergoes continuous self-renewal to maintain intestinal homeostasis . Given that dysregulation of zinc flux causes intestinal disorders , appropriate spatiotemporal regulation of zinc in the intracellular compartments should be a prerequisite for the intestinal epithelial self-renewal process . Zinc transporters such as Zrt-Irt-like proteins ( ZIPs ) are essential to fine-tune intracellular zinc flux . However , the link between specific zinc transporter ( s ) and intestinal epithelial self-renewal remains to be elucidated . Here , we found that ZIP7 is highly expressed in the intestinal crypts . The finding motivated us to further analyze the role of ZIP7 in intestinal homeostasis . ZIP7 deficiency greatly enhanced ER stress response in proliferative progenitor cells , which induced apoptotic cell death . This abnormality disrupted epithelial proliferation and intestinal stemness . Based on these observations , we reason that ZIP7-dependent zinc transport facilitates the vigorous epithelial proliferation in the intestine by ameliorating ER stress . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"paneth",
"cells",
"cell",
"death",
"organoids",
"medicine",
"and",
"health",
"sciences",
"biological",
"cultures",
"cell",
"processes",
"endoplasmic",
"reticulum",
"epithelial",
"cells",
"physiological",
"processes",
"homeostasis",
"stem",
"cells",
"organ",
"cultures",... | 2016 | Zinc Transporter SLC39A7/ZIP7 Promotes Intestinal Epithelial Self-Renewal by Resolving ER Stress |
An estimated 2 . 7 million new HIV-1 infections occurred in 2010 . `Treatment-for-prevention’ may strongly prevent HIV-1 transmission . The basic idea is that immediate treatment initiation rapidly decreases virus burden , which reduces the number of transmittable viruses and thereby the probability of infection . However , HIV inevitably develops drug resistance , which leads to virus rebound and nullifies the effect of `treatment-for-prevention’ for the time it remains unrecognized . While timely conducted treatment changes may avert periods of viral rebound , necessary treatment options and diagnostics may be lacking in resource-constrained settings . Within this work , we provide a mathematical platform for comparing different treatment paradigms that can be applied to many medical phenomena . We use this platform to optimize two distinct approaches for the treatment of HIV-1: ( i ) a diagnostic-guided treatment strategy , based on infrequent and patient-specific diagnostic schedules and ( ii ) a pro-active strategy that allows treatment adaptation prior to diagnostic ascertainment . Both strategies are compared to current clinical protocols ( standard of care and the HPTN052 protocol ) in terms of patient health , economic means and reduction in HIV-1 onward transmission exemplarily for South Africa . All therapeutic strategies are assessed using a coarse-grained stochastic model of within-host HIV dynamics and pseudo-codes for solving the respective optimal control problems are provided . Our mathematical model suggests that both optimal strategies ( i ) - ( ii ) perform better than the current clinical protocols and no treatment in terms of economic means , life prolongation and reduction of HIV-transmission . The optimal diagnostic-guided strategy suggests rare diagnostics and performs similar to the optimal pro-active strategy . Our results suggest that ‘treatment-for-prevention’ may be further improved using either of the two analyzed treatment paradigms .
HIV-1 infection remains one of the major global health challenges with an estimated 33 million infected and a continuing spread [1] . Currently , an efficient vaccine remains to be developed , while at the same time the complete elimination of replication-competent virus within the host can not be achieved due to the persistence of the virus in inducible , latent cellular reservoirs [2 , 3] , as well as insufficient pharmacological suppression of actively replicating virus in some anatomical/cellular reservoirs [4 , 5] . However , the current situation urges for methods that could bring the epidemic to a halt , or possibly end it . Currently , the most promising strategies are based on the use of antiviral drugs: Pre-exposure prophylaxis ( PrEP ) [6–9] aims to protect uninfected individuals ‘at risk’ by decreasing the probability of infection upon virus exposure , e . g . [10] . PrEP may however be too costly to be broadly implemented in resource-poor countries [11] . Currently , the decision to initiate treatment against HIV is largely guided by CD4+ cell levels [12 , 13] . However , the viral load , which is the primary determinant of infectiousness [14 , 15] , may be very high within the time-window between HIV infection and initiation of treatment . ‘Treatment for prevention’ [16] aims to put infected individuals on therapy as early as possible . This can reduce the infectiousness of a patient by decreasing within-host virus levels , which reduces the amount of transmitted viruses per contact and the probability of infection upon exposure . Analysis of the only completed clinical study to date , HPTN052 [16] , estimated that ‘treatment for prevention’ may reduce the number of linked HIV-1 transmissions by 96% and the number of total HIV-1 transmissions by 89% relative to delayed treatment initiation and subsequently it was nominated as the “breakthrough of the year 2011” by the Science magazine [17] . In the aftermath of the HPTN052 trial , the cost-efficacy of ‘treatment for prevention’ was analyzed by many mathematical modeling approaches ( reviewed in [18] ) . One problem is that most of these approaches focused solely on the epidemic level and did not model drug resistance development within the hosts , which indirectly assumes that the efficacy of ‘treatment for prevention’ is constant over time . However , because viral transmission is strongly correlated with viral levels in the transmitting individual [14 , 15 , 19–21] , it is reasonable to assume that also the efficacy of ‘treatment for prevention’ is intimately connected with viral suppression . One major challenge during HIV treatment lies in the virus’ tendency to develop drug resistance [22] , which in turn can lead to virus rebound and promote HIV transmission for the time it remains unrecognized . An earlier treatment initiation may thus demand an improved therapeutic strategy , that allows long-term control of virus replication ( beyond the typical duration of a clinical trial ) . While sophisticated patient monitoring and timely treatment changes may allow to minimize windows of unrecognized viral breakthrough , they require significant monetary funds , good infrastructure , diagnostic facilities and the availability of alternative treatment options . Only few of these may be available in resource-constrained countries , where the requirement of resources may strongly dominate the possibility to implement a reasonable ‘treatment for prevention’ strategy . Obviously , scaling ‘treatment for prevention’ requires careful examination of various aspects and a policy maker should strike a proper balance between societal and individual perspectives [23] . This work addresses the scaling of ‘treatment for prevention’ by suggesting optimal treatment strategies for the long-term control of HIV within its host ( as recommended by [24] ) . Optimality will be defined from a national economic perspective , taking into account that a diseased individual implies an economic loss . By considering the national economic perspective , we do not evaluate what should be done , but rather what is already worthwhile . However , we also evaluate the derived optimal strategies from an individual perspective and in terms of their utility in prevention , i . e . whether a strategy prolongs the life of an infected person and whether the risk of HIV onward transmission is reduced . We hereby focus on two distinct approaches to handle treatment decisions: The first assumes that treatment decisions ( i . e . when to change therapy ) are made on an individual basis , guided by infrequent diagnostics ( referred to as diagnostic-guided strategy ) . This represents a medical scenario in which a treating physician decides based on the diagnosed status of the patient that he encounters . The second approach suggests pro-active treatment decisions ( referred to as pro-active strategy ) , i . e . does not require diagnostic ascertainment of the patients’ disease status . The two approaches are modeled and solved by two distinct mathematical frameworks . The former is addressed using the recently developed framework of ‘Markov Decision Processes with Rare State Observations’ [25]: For each disease state , it computes the optimal treatment and the next time of medical diagnostics , minimizing viral burden as well as treatment- and diagnostic costs . The latter approach ( the pro-active strategy ) is modeled as an open-loop switched system , where the decision to change the treatment depends on the initial disease state of the patient and the anticipated , ( treatment- ) induced stochastic dynamics up to some time t . The later strategy allows to switch treatment before drug resistance is detectable in the individual ( pro-active ) and may be easier to implement in resource-constrained settings , where poor infrastructure and the costs of diagnostics limit their applicability . By assessing these two distinct frameworks side-by-side , we can rigorously evaluate the different treatment paradigms in terms of their optimality . Algorithms to solve these problems were developed and are stated in the supplementary materials . Several other groups have suggested optimal [26–28] or sub-optimal [29 , 30] treatment strategies to mitigate drug resistance in HIV-1 . All authors treated the underlying system deterministically , which fails to capture the intrinsic stochastic nature of HIV drug resistance development [31] and the time-scales on which drug resistance develops . None of the previous work focused on HIV prevention , and neither work questioned the analyzed treatment philosophy , either focusing on pro-active treatment switching strategies [26–28 , 30] , or diagnostic-driven strategies [29] . In contrast , we used a stochastic model of HIV long-term dynamics after drug application [25] to more realistically capture the underlying dynamics . Also , we evaluate different assumptions for the controllability of the disease dynamics , by evaluating the two different optimal control frameworks , which allows for an objective assessment of alternative treatment philosophies . The manuscript will be organized as follows: We will extend- and parameterize the model introduced in [25] for our needs . After recapitulating essential theory for the diagnostic-guided strategy , we introduce the mathematical concepts behind the pro-active strategy , solve both optimal control problems and evaluate them with respect to monetary costs , patient survival and reduction of onward transmission . All algorithms that we developed to solve the optimal control problems will be provided in the S1 and S2 Text for the interested reader .
The two addressed optimal control approaches share an identical model ( Fig 1 ) that reflects the short-term dynamics of viral decay- and rebound ( Fig 2 ) , as well as the stochastic HIV long-term dynamics after drug application , see Fig 3 . Within this work , we put a focus on viral kinetics and will only indirectly relate to the patient’s health . This is because we are interested in ‘treatment for prevention’ and particularly its efficacy in decreasing onward transmission , which is correlated with the viral load [19–21] and not necessarily with the immune status of the HIV infected patient . The two optimal control problems that we solve , i . e . the diagnostic-guided strategy and the pro-active strategy , differ slightly in the underlying assumption on the controllability of the disease dynamics . Both control strategies will be described in the following , defining in each case a control policy , a performance criterion and an optimality equation . Solving optimal control problems is generally computation intense and may not always be achievable . Our two optimal control scenarios require different algorithms for their solution . For computing the optimal diagnostic-guided strategy , we used an adapted policy iteration algorithm , see S1 Text for details . In order to numerically compute the optimal pro-active strategy , we introduce a dynamic programming technique in S2 Text , which was developed for the considered performance criterion ( expected discounted costs over an infinite time horizon ) . It has some similarity with the algorithm introduced by Hernandez-Vargas [27] , which , however , considers a different performance criterion ( only terminal cost ) . Both algorithms were implemented in MATLAB Version 8 and parallelized , where applicable . For the dynamic programming technique in S2 Text we used the state of art solver cplex from the IBM ILOG CPLEX [56] Optimization Studio to solve embedded linear programs .
The optimal diagnostic-guided strategy is given in S1 Table . In brief , for the considered parameters ( Tables 1 and 2 ) , it is suggested to use the first-line treatment a1 in all states , except those where the virus is resistant against treatment a1 , but susceptible to a2 . In the later case treatment line a2 is suggested . In line with this treatment recommendation , patient monitoring is only suggested as long as the patient is infected with drug-susceptible ( “wild-type” ) virus . If the patient has a high or medium virus load , the next diagnostic test should be within 25 days , if the patient has a ℓow/non-detectable virus load , after 152 days . These results may indicate that the cost for diagnostics is too high in relation to the economic benefit resulting from more close monitoring and informed treatment adaptation ( this will be discussed later in the Discussion ) . An exemplary trajectory that highlights the treatment strategy is shown in Fig 4A . The blue line indicates a patient-specific trajectory . The filled black marks indicate the times when a diagnostic test is performed and the background shading indicates the applied treatment ( white: a1 , gray shading: a2 ) . In the example , the patient initially has a high copy number h of wild type ( WT ) virus , while none of the drug resistant viruses are present . This state is represented by the vector notation X t 0 = [ h , 0 , 0 , 0 ] . For this state , the optimal treatment policy ( see S1 Table ) suggests to use treatment a1 and to perform the next diagnostic test in 25 days ( the second black marking in panel Fig 4A ) . At the next diagnostic test , the patient is in state [ m , 0 , 0 , 0 ] for which continuation of treatment a1 is recommended and the next diagnostic test is scheduled after 25 days ( the 3rd–9th black marking in panel Fig 4A ) . In the following , the virus remains suppressed , with a small detected ‘blip’ after about 500 days . After about 600 days of treatment , during the time lapse between diagnostic tests , the a1 resistant strain R1 emerges . Notice transitions from the state [ m , 0 , 0 , 0 ] → [ m , ℓ , 0 , 0 ] → [ m , m , 0 , 0 ] , then [ ℓ , m , 0 , 0 ] and finally [ ℓ , h , 0 , 0 ] in the Fig 4A , where the copy number of a1 resistant strain R1 increases from a ℓow copy number to a high copy number ( virus rebound after resistance development ) . At the time point of the next diagnostic ( at around 700 days ) , the emergence of resistance is identified [ ℓ , h , 0 , 0 ] and a switch to treatment a2 is suggested ( marked by gray region in Fig 4A ) . After the initiation of treatment a2 , a transition to state [ ℓ , ℓ , 0 , 0 ] can be observed in the trajectory , which implies a decrease in the a1 resistant strain ( viral suppression ) . The optimal pro-active strategy depends on the initial probability state of the patient p0 . We assumed that the patient is treatment naive and has high virus copy numbers , i . e . p [ h , 0 , 0 , 0 ] ( t 0 ) = 1 and p [ x ] ( t 0 ) = 0 for x∈𝒮\[h , 0 , 0 , 0] . For this scenario , it is suggested to start with treatment line a1 and to switch to a2 after 14 days , which is then maintained . The trajectories of the patient probability states are depicted in Fig 4B . For the ease of interpretation , we illustrate only the sets of viral states 𝓛 , ℳ , ℋ and patient death ✠ . 𝓛 denotes an undetectable total viral load . Translated to our model , this is the set of states for which condition nC ( M ) ≤ ℓ for all possible virus mutants M holds , i . e . the current state has to fulfill [ ≤ ℓ , ≤ ℓ , ≤ ℓ , ≤ ℓ ] to belong to this set . Likewise ℋ denotes a high total viral load , i . e . refers to all states for which for at least one viral strain M , nC ( M ) > m is fulfilled . The remaining viral states belong to ℳ . One can nicely see that after approximately 260 days , maximum viral suppression may be achieved in the sense that the probability to have undetectable virus load ( 𝓛 ) is maximal ( 64 . 19% ) , while the patient may have intermediate viral loads ℳ with 15 . 57% probability and high viral loads ℋ with only 14 . 40% probability ( the probability of death is 5 . 84% ) . After this time , it becomes more likely to fail treatment , as indicated by an increase in states ℳ and ℋ relative to 𝓛 . We also assessed the sensitivity of the optimal pro-active strategy to variations in parameter values and found it to be fairly insensitive to parameter perturbations , see S3 Text . For comparison , we also show the dynamics for the case when no treatment switches were conducted in S4 Text in relation to the optimal pro-active strategy . In our model , the cost incurred by a treatment strategy can be divided into two types: The direct costs , which include treatment- and diagnostic costs , and indirect costs incurred by the virologic/health status of a patient ( state costs ) . The pro-active strategy does not comprise diagnostic tests , whereas the protocol for the current standard of care ( S . O . C . ) , as well as the protocol used in the HPTN052 [16] , which we simulate for comparison , require viral load measurements . Currently , the expensive resistance tests are not part of the protocol for the standard of care , nor were they used for treatment decisions in HPTN052 . The protocol for S . O . C . recommends changing treatment , if viral load ( which is measured at month 6 and then every 12 months ) is detectable and confirmed in a follow up testing after 2 months . The protocol for the HPTN052 trial recommends changing treatment , if two consecutive viral load measurements were greater than 1000 copies/mL , 16 weeks after treatment initiation . Viral load was measured at week 2 , at month 1 , 2 , 3 after treatment initiation and then every 3 month . The cost of virologic testing is roughly 30 US$ per test [57 , 58] . In contrast to S . O . C . and HPTN052 , the diagnostic-guided strategy requires drug resistance testing . We set the cost of the diagnostics for the diagnostic-guided strategy to 200 US$ per test , in line with the recent literature [57 , 59] . Table 3 displays the expected discounted costs for an infinite time horizon for different strategies and highlights the direct- and indirect costs of each strategy , respectively . This comparison shows that the pro-active strategy performs best ( 83 , 819 US$ ) , followed closely by the diagnostic-guided strategy ( 83 , 858 US$ ) , the HPTN052 protocol ( 84 , 600 US$ ) and then by the standard of care ( 85 , 641 US$ ) . The total expected discounted costs for the pro-active- and the diagnostic-guided strategy are 2% less than that of the standard of care . The state costs ( indirect cost related to patient-well being ) are the major determinant of the total cost , making up roughly 98% , 97% , 97% and 93% of total cost for the S . O . C . , the HPTN052 protocol , the pro-active—and the diagnostic-guided strategy respectively . In terms of state costs , the diagnostic-guided strategy performs best . The direct costs ( treatment and diagnostic costs ) are highest for the diagnostic-guided strategy ( 5 , 539 US$ ) followed by the pro-active strategy ( 2 , 772 US$ ) , the HPTN052 protocol ( 2 , 390 US$ ) and the standard of care ( 1 , 871 US$ ) . The direct costs make up only 2% , 3% , 3% and 7% of the total costs for S . O . C . , the HPTN052 protocol , the pro-active and the diagnostic-guided strategy respectively . The direct costs of the pro-active and the diagnostic-guided strategy are roughly 48% and 196% more than that of S . O . C . Clearly , the primary goal of any treatment strategy is to improve and prolong the life expectancy of the treated individual . We therefore compare the distinct treatment strategies in terms of patient survival . For that purpose , we define the following term: ℙ ( X s = ✠ | stg ) which denotes the probability of death ✠ at time s given that the patient was treated according to treatment strategy stg . Given two distinct strategies; stg and a reference treatment strategy stgref , the term T 0 → t + ( stg , stg ref ) refers to the expected years of life gained ( life prolongation ) when the treatment strategy stg is used , relative to the reference treatment stgref at time t after initiation of treatment: T 0 → t + ( stg , stg ref ) = ∫ s = 0 t ℙ ( X s = ✠ | stg ref ) - ℙ ( X s = ✠ | stg ) d s ⋅ ( 32 ) In other words , given a patient is treated with stg and another patient is treated with stgref for time t , the terms T 0 → t + ( stg , stg ref ) refers to the expected time that a patient treated with stg will live longer than the patient treated with stgref . We compared all strategies with the following reference strategies stgref: i ) no medical intervention , ii ) the standard of care treatment , iii ) treatment according to the HPTN052 protocol and iv ) the diagnostic-guided strategy . Fig 5A and 5D show the trajectories of expected life prolongation by different strategies in relation to i ) -iv ) . Table 4 displays the expected life-years gained after 1 - , 2 - , 5 - , 8 - , 12—and 13 . 7 years of treatment respectively , where we additionally show the expected life prolongation in relation to the uninfected state . The first five rows of Table 4 show the expected loss-of-life-time of an HIV infected person treated with distinct strategies in relation to an HIV uninfected person . After 13 . 7 years , an HIV patient receiving no treatment lives on average 6 . 2 years less than a healthy person . An HIV patient receiving treatment according to S . O . C . , the pro-active strategy , the diagnostic-guided strategy or according to the HPTN052 protocol lives on average 3 , 2 . 66 , 2 . 3 and 2 . 82 years less than a healthy person . Fig 5A shows that all treatment strategies are better than receiving no treatment at all and prolong the life of an HIV patient by at least 3 . 2 years in relation . Fig 5B shows that the diagnostic-guided , pro-active strategy and the HPTN052 protocol are better at increasing patient survival than the standard of care . Further , Fig 5C shows that the optimal strategies are slightly better than the HPTN052 protocol and Fig 5D shows that the pro-active strategy and the HPTN052 protocol are slightly worse than the diagnostic-guided strategy . Table 4 shows that during the initial 2–3 years of treatment , there is almost no difference between the diagnostic-guided and the pro-active strategy with regard to patient survival . After 13 . 7 years of treatment , the difference between the two optimal strategies is less than 5 month ( 0 . 358 years ) . Besides the primary goal of improving the life of the HIV patient , ‘treatment for prevention’ has gained interest in recent years . ‘Treatment for prevention’ strategies reduce onward transmission of the virus by reducing the infectiousness of HIV positive individuals . In order to measure the efficacy of the treatment strategies in preventing HIV-1 transmission , we estimated the incidence rate per 100 person-years associated with each HIV lumped state ( ℓ , m , h ) from a meta-analysis by Attia et al [14] ( see S5 Text ) . The meta-analysis summarizes the outcome of 11 clinical studies on HIV-1 transmission in heterosexual sero-discordant couples , primarily from Africa . For a strategy stg applied for a time t , the following equation gives a measure of the expected number of secondary cases/transmissions per survivor 𝔼 0 → t ( transm ⋅ | stg ∧ ¬ ✠ ) = ∫ s = 0 t ∑ x ℙ ( X s = x | stg ) · 𝕀ℝ ( x ) 1 - ℙ ( X s = ✠ ) d s ( 33 ) where 𝕀ℝ ( x ) is the incidence rate per 100 person-years for a state x in our virus dynamics model , as explained in S5 Text and given in Table 1 . Given two strategies , stg1 and stgref , the percentage of potential infections prevented by strategy stg1 in comparison to the reference strategy stgref is given by the quotient: % transmissions prevented until t = 100 · ( 1 - 𝔼 0 → t ( transm ⋅ | stg 1 ∧ ¬ ✠ ) 𝔼 0 → t ( transm ⋅ | stg ref ∧ ¬ ✠ ) ) ( 34 ) We computed the expected reduction of secondary cases for different strategies taking either no treatment or the current standard of care as the reference strategy . In comparison to no treatment , the maximal reduction of secondary cases for the pro-active - , the diagnostic-guided strategy , the HPTN052 protocol and S . O . C . are achieved roughly 1 . 5–3 years after treatment initiation with values of 86% , 87% , 82% and 79% respectively , see Fig 6A . The relative reduction of secondary cases per survivor for the diagnostic-guided and the pro-active strategy are very similar , with an increase for the first 2 years , followed by a slow decline ( see Fig 6A and Table 5 ) . The relative reduction of secondary cases per survivor for the HPTN052 protocol is slightly better than that of S . O . C , with a tendency to decline over time , see Table 5 . Note , that the computed relative reduction of secondary cases with the HPTN052 protocol was 82% ( Table 5 ) , which is slightly lower than the reported relative reduction of transmission events in the actual HPTN052 study [16] ( reduction of 96% of linked and 89% of total transmission events ) . We have discussed reasons for this apparent under-prediction later in the manuscript . The difference between the optimal strategies ( diagnostic-guided and the pro-active strategy ) and S . O . C . becomes evident , when looking at the relative risk reduction by the optimal treatment strategies in relation to S . O . C . in Fig 6B . The reduction in secondary cases per survivor by the optimal strategies in comparison to S . O . C . is highest at the beginning and then slowly decreases over time .
The main aim of this work was to develop a rigorous mathematical framework that allows to compare different treatment paradigms in terms of monetary costs , treatment benefit and efficacy for ‘treatment for prevention’ . It was previously stated [60] , that the durability of ‘treatment for prevention’ should be assessed . Our simulations over a long time horizon ( up to 5000 days/13 . 7 years ) indicate that the effect of ‘treatment for prevention’ is significant and remains relatively stable beyond the time horizon typically assessed in clinical studies , see Fig 6A and Table 5 , and that it may even be improved . We estimated that a standard of care therapy in e . g . South Africa can achieve a 66–79% reduction of HIV-1 onward transmission , in comparison to delivering no treatment . We also implemented the HPTN052 protocol , as stated in [16] and predicted that it would achieve up to 82% reduction of HIV-1 transmission , being more effective than the current standard of care , as shown in Fig 6B . Statistical assessment of the actual HPTN052 trial [16] yielded estimates for the relative reduction of transmission of 96% for linked transmission and 89% for any transmission . Our simulated HPTN052 protocol yielded a 82% reduction of onwards transmission , which is within the confidence range of the reported estimates ( CI: 73–99% for linked transmission and CI: 68–96% for any transmission ) [16] . Note , that only one linked transmission event ( 1/1585 person-years ) was observed in the early therapy arm of HPTN052 [16] , giving rise to the statistical uncertainty in the reported estimate . Nevertheless , our simulations may under-predict the efficacy of HPTN052 due to several factors: The reported treatment efficacy in the HPTN052 study was higher than predicted by our model: Virologic failure was only observed in 5% of participants in the early-therapy group of HPTN052 , possibly explaining the difference between the outcome of the simulation vs . the clinical trial . Despite only 5% failing to suppress the virus in the HPTN052 study , 66% initiated a second line therapy [16] , meaning that a significant proportion of patients switched treatment before/without virologic failure . In our simulations of the HPTN052 protocol , patients only switched treatment when they showed signs of virologic failure . However , one may speculate that these treatment switches before/without virologic failure may have an impact on the long-term viral suppression that could be similar to a pro-active treatment switch . The primary measurable endpoint of the HPTN052 study was the infection of the sero-discordant partner . Onward transmissions to other individuals could not be quantified for obvious reasons . While a number of trials are now underway to confirm the results of HPTN052 , see e . g . [61 , 62] , our in silico approach specifically addresses the need for an improved treatment strategy , particularly taking affordability into account , which suggests strategies that are suitable for scaling up . Our work may indicate that if ‘treatment for prevention’ is scaled up and implemented using the current standard of care treatment strategy , its efficacy may not be as high as expected from HPTN052 . Unlike in HPTN052 , where monitoring of treatment success ( viral suppression ) and timely execution of treatment changes were realized , in resource-constrained countries close patient monitoring is currently not implemented in a routine setting and may be difficult to realize due to infrastructural and economic requirements . Two alternative strategies for the immediate initiation of therapy were assessed in our work that take into account the mentioned limitations . Both suggested strategies ( the diagnostic-guided strategy and the pro-active strategy ) yielded better results in our simulations in terms of the reduction of onward transmission ( see Table 5 ) at a lower price ( Table 3 ) . Both optimal strategies could yield a 72–87% reduction in HIV onward transmission in comparison to no treatment , see Fig 6A and Table 5 . In comparison to the standard of care , we estimated that the diagnostic-guided strategy and the pro-active strategy could yield another 33–38% reduction of onward transmission after 2 years of treatment , but the advantages of the diagnostic-guided strategy and the pro-active strategy over the standard of care slowly declined over time , see Fig 6B . This indicates that both optimal strategies have a particular strength in reducing early transmissions ( shortly after treatment initiation ) in comparison to the current standard of care . This may be of particular utility , if transmission occurred primarily during early infection [63 , 64] . In our work , we did not take behavioral factors into account , which would lead to a time-dependency of the infection rate . Rather , we assumed that the infection rate 𝕀ℝ ( x ) is constant over time , but dependent on the total virus load as reported earlier [14 , 15 , 19–21] . If transmission would primarily take place during an early infection , the advantages of the diagnostic-guided strategy and the pro-active strategy over the standard of care would be even more pronounced than indicated in Fig 6B . The optimal diagnostic-guided strategy suggested patient-specific diagnostics , i . e . dependent on the patient’s virologic status ( see S1 Table ) , unlike fixed intervals as in S . O . C , or the protocol stated in [16] . In summary , the optimal diagnostic-guided strategy suggests to take frequent diagnostics ( ≈ every month ) if the patient is infected with a high or medium number of treatment-susceptible virus and less frequent ( ≈ every 5 month ) diagnostics if the patient is infected with a ℓow/undetectable number of virus . No diagnostics are recommended for the remaining virologic states . Altogether , a very sparse diagnostic schedule for individual patients is suggested . Previous work [25] indicated that price reductions for the diagnostic tests would yield a better patient-outcome , which indicates that available drugs may not be utilized optimally in resource-poor settings , because diagnostics are currently too expensive . Of note is the fact that albeit treatment being available at very low expense ( due to negotiations by the Clinton Foundation [65] ) , diagnostics may not be . Furthermore , we suspected that allowing treatment change only after diagnostic confirmation of treatment failure ( i . e . some time after drug resistance has occurred ) may limit future treatment options [34] . Since the optimal diagnostic-guided strategy suggested rare diagnostics , and because it only allows to change treatment after resistance is detectable , we evaluated pro-active switching strategies . Note , that pro-active treatment switching strategies tested in the clinic increased virologic suppression and lowered rates of drug resistance emergence in HIV-1 , when compared to conventional strategies [66 , 67] . Similar strategies are also used against bacterial infections and cancers . The computed pro-active strategy suggests a single treatment change without prior ascertainment of the viral status within a treated patient . Surprisingly , this strategy could yield comparable outcomes in terms of monetary costs , patient health and reduction of onward transmission , see Tables 3–5 and Figs 5 and 6 . Our work thus indicates that pro-active strategies , may be as effective as diagnostically-driven ones , when diagnostics are expensive or inaccessible . Note , that unlike other optimal control approaches , i . e . [28] that suggest infinitely fast switching between regimens to mitigate drug resistance emergence , our predicted pro-active strategy actually only recommends a single treatment change , which is clinically more realistic . We also analyzed the sensitivity of the pro-active strategies with respect to the timing of this switch ( see Fig 7 ) . The graphic illustrates , that the switch is optimal after 14 days , however the difference in the performance measure is marginal , as long as the treatment switch is performed before ≈ 30 days ( 1 month ) after treatment initiation . Obviously , pragmatic and clinical considerations need to be taken into account to translate our results into practice . Also , several assumptions have been made , which require careful evaluation . For example , we used a very coarse-grained model of HIV within-host dynamics , which was required to enable the numerical computation of optimal controls , particularly for the closed-loop system employed in the diagnostic-guided strategy . Most models of viral dynamics , e . g . [33 , 68 , 69] , were developed to accurately predict short-term viral dynamics after drug application and are unable to predict virologic failure after long time intervals , in contrast to our coarse-grained model , which was developed and parameterized in order to predict short-term viral dynamics as well as virologic failure after very long time-intervals , see e . g . Figs 2 and 3 . It is therefore more suitable than existing models in estimating the long-term response to antiviral treatment . However , in the future we will focus on developing more elaborated HIV-models and on algorithms to solve the control problem for the chemical master equation directly , without state-space lumping . Note , that other computationally efficient numerical approaches , such as model predictive control [30] , could be used to approximate the optimal treatment strategies . However , there is no guarantee that the computed control using these approaches will be optimal . In our approach , we modeled treatment change as a switched system , which neglects the pharmacokinetics of the distinct drugs [10 , 70–72] and may only indirectly reflect drug adherence in an average population ( drug efficacy η is a constant term in our model ) . Neglecting pharmacokinetics may , however , be a justifiable step in this modeling exercise , because of the considered time-scales ( on the order of years ) , and also because optimizing e . g . drug adherence was not an objective of this study . However , if the main interest is for example in optimizing the switch between two treatment lines by optimal dosing in order to prevent time frames of insufficient viral suppression or drug over-exposure , or to include patient-specific or time-dependent drug adherence , we advise to consider a different control system , for example [73] . Within such a framework , monitoring ( e . g . viral load assessment ) may also be incorporated as a tool to assess individual drug adherence and to allocate resources to improve it . We did not explicitly consider costs related to contraindications caused by the treatment . For example , the second treatment line a2 may be less tolerable . Mathematically , this can be modeled in terms of increased treatment costs for a2 , in comparison to the first treatment line a1 . In order to test the sensitivity of the optimal pro-active strategy to this parameter , we conducted the necessary computations in S3 Text and found that the computed strategy was fairly insensitive to changes in treatment costs . This may indicate that the benefits of the treatment switch outweigh these potential shortcomings . Also , we did not include screening costs or the costs of the initial virologic assessment , thus our calculations refer to the public health costs that accrue from the start of HIV treatment . These costs will , however , only enter as a constant to each of the tested strategies and will not change the results beyond the addition of this constant to the values stated in Table 3 . Additional costs ( personnel , infrastructure , transportation ) may come along with diagnostic assessments . It is likely that hidden costs for diagnostics are substantial . With a higher cost of diagnosis , the pro-active strategy may outperform the diagnostic-guided strategy , which may suggest an even less frequent diagnostic schedule , supporting our claim that current diagnostics may be too expensive to be appropriately used . We used the price of a drug resistance test ( kdia ≈ 200 US$ [57 , 59] ) to account for diagnostics in the diagnostic-guided strategy . This had the following reason: Current guidelines recommend to measure the total virus load [13] and to switch treatment , if , based on this partial information , virologic failure is anticipated . As reported by others [57] , this may lead to unnecessary treatment switches . In contrast , a resistance test directly informs the physician about the necessity of treatment change . Mathematically , partial information , i . e . the total virus load , would lead to a distinct control framework , namely Partially Observable Markov Decision Processes ( POMDP ) [74] , which are extremely challenging to solve , particularly for larger models like the one used herein ( Fig 1 ) . In POMDPs , partial information may be mapped into a ‘believed’ full virologic status , for example observing a high total virus load may be due to some resistance development , e . g . the viral state [ ℓ , h , 0 , 0 ] with some probability . However , it is hard for us formalize the physicians intuition ( i . e . the relation between observation , belief and truth ) regarding this ‘mapping’ of partial measurements to viral states x . As a primary outcome of our modeling exercise , we estimated the expected relative number of secondary infections prevented ( Table 5 and Fig 6 ) ; -unlike many other approaches ( summarized in [18] ) , which take the absolute number of secondary cases into account . Estimating absolute numbers of secondary cases would require to model complex behaviors , i . e . sexual relationships , etc . over time , for which we do not have data for validation , nor was it the main focus of the current work . This also prevents us from predicting the course of the epidemic or deriving its reproductive number R0 in relation to distinct treatment strategies . However , the primary aim of this study was to compare the efficacy of different treatment strategies , which is nicely quantified in terms of the expected relative number of secondary infections prevented . Note , that this relative estimate requires no assumptions on the underlying transmission dynamics , except that it assumes that these dynamics are similar for a tested strategy versus its comparator . In addition to insights in HIV ‘treatment for prevention’ strategies , the developed mathematical/control theoretic framework may already be applicable to many medical phenomena . Further developments may improve its applicability to even more complex processes , which can be accurately described by intrinsically stochastic dynamics . For example , the open-loop optimal control approach ( used to determine the optimal pro-active strategy ) may be turned into a closed-loop system , if diagnostics are taken from time-to-time to determine the viral state of a patient , i . e . p[x] ( tj ) . Also , the closed-loop system that requires state determination ( the diagnostic-guided strategy ) may be combined with the open-loop system in order to allow for pro-active treatment changes in between diagnostic assessments . | HIV-1 continues to spread globally . Antiviral treatment cannot cure patients , but it slows disease progression and may prevent HIV transmission by decreasing the amount of transmittable viruses in treated individuals . ‘Treatment-for-prevention’ argues for immediate treatment initiation and may reduce transmission by 96% ( CI: 73–99% ) , according to the results of a large clinical study ( HPTN052 ) . In order to ensure long-lasting treatment success , early therapy initiation demands more sophisticated treatment strategies & exceeding funds . However , countries facing the highest HIV burden are among the poorest . Within this work , we provide a mathematical framework that allows assessing different treatment paradigms using optimal control theory together with stochastic modelling of within-host viral dynamics and drug resistance development . We use this framework to compute and evaluate two distinct optimal long-term treatment strategies for resource-constrained settings: ( i ) a diagnostic-guided and ( ii ) a pro-active treatment strategy . The cost of a strategy is evaluated from a national economic perspective , valuating a severe patient health status in terms of an economic loss . The optimal strategies are compared with current clinical treatment protocols and no treatment in terms of costs , life expectation and reduction of secondary cases . Our simulations indicate that the pro-active treatment strategy performs comparably to the diagnostic-guided treatment strategy . Both strategies perform better than current clinical protocols , suggesting directions for improvement . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [] | 2015 | Optimal Treatment Strategies in the Context of ‘Treatment for Prevention’ against HIV-1 in Resource-Poor Settings |
Wolbachia are maternally inherited bacteria that infect arthropod species worldwide and are deployed in vector control to curb arboviral spread using cytoplasmic incompatibility ( CI ) . CI kills embryos when an infected male mates with an uninfected female , but the lethality is rescued if the female and her embryos are likewise infected . Two phage WO genes , cifAwMel and cifBwMel from the wMel Wolbachia deployed in vector control , transgenically recapitulate variably penetrant CI , and one of the same genes , cifAwMel , rescues wild type CI . The proposed Two-by-One genetic model predicts that CI and rescue can be recapitulated by transgenic expression alone and that dual cifAwMel and cifBwMel expression can recapitulate strong CI . Here , we use hatch rate and gene expression analyses in transgenic Drosophila melanogaster to demonstrate that CI and rescue can be synthetically recapitulated in full , and strong , transgenic CI comparable to wild type CI is achievable . These data explicitly validate the Two-by-One model in wMel-infected D . melanogaster , establish a robust system for transgenic studies of CI in a model system , and represent the first case of completely engineering male and female animal reproduction to depend upon bacteriophage gene products .
Wolbachia are the most widespread endosymbiotic bacteria on the planet and are estimated to infect half of all arthropod species [1 , 2] and half of the Onchocercidae family of filarial nematodes [3] . They specialize in infecting the cells of reproductive tissues , are primarily inherited maternally from ova to offspring , and often act in arthropods as reproductive parasites that enhance their maternal transmission by distorting host sex ratios and reproduction [4 , 5] . The most common type of reproductive parasitism is cytoplasmic incompatibility ( CI ) , which manifests as a sperm modification in infected males that causes embryonic lethality or haploidization in matings with uninfected females upon fertilization [6–8] . This embryonic lethality is rescued if the female is infected with the same Wolbachia strain . As such , CI selfishly drives CI-inducing Wolbachia into host populations [9–13] , and the incompatibilities between host populations cause reproductive isolation between recently diverged or incipient species [14–18] . In the last decade , Wolbachia and CI have garnered significant interest for their utility in combatting vector borne diseases worldwide . Two strategies are currently deployed: population suppression and population replacement . The population suppression strategy markedly crashes vector population sizes through the release of only infected males that induce CI upon mating with wild uninfected females [19–22] . In contrast , the population replacement strategy converts uninfected to infected populations through the release of both infected males and females that aid the spread Wolbachia via CI and rescue [23 , 24] . Replacing a vector competent , uninfected population with infected individuals can notably reduce the spread of arthropod borne diseases such as Zika and dengue [25 , 26] because Wolbachia appear to inhibit various stages of viral replication within arthropods based on diverse manipulations of the host cellular environment [27–33] . The combination of Wolbachia’s abilities to suppress arthropod populations , drive into host populations , and block the spread of viral pathogens have established Wolbachia in the vanguard of vector control efforts to curb arboviral transmission [22–25 , 34–36] . An unbiased , multi-omic analysis of CI-inducing and CI-incapable Wolbachia strains revealed two adjacent genes , cifA and cifB , in the eukaryotic association module of prophage WO [37] that strictly associate with CI induction [38] . Fragments of the CifA protein were found in the fertilized spermathecae of wPip infected Culex pipiens mosquitoes [39] , and these genes are frequently missing or degraded in diverse CI-incapable strains [40 , 41] . Dual transgenic expression of cifA and cifB from either of the CI-inducing strains wMel or wPip in uninfected male flies causes a decrease in embryonic hatching corresponding to an increase in CI-associated cytological abnormalities including chromatin bridging and regional mitotic failures [38 , 42] . Single transgenic expression of either cifAwMel or cifBwMel in an uninfected male was insufficient to recapitulate CI , but single transgenic expression of either gene in an infected male enhances wMel-induced CI in a dose-dependent manner [38] . Importantly , dual transgenic CI induced by cifAwMel and cifBwMel expressing males was rescued when they were mated with wMel-infected females [38] . Moreover , transgenic expression of cifAwMel alone in uninfected females rescues embryonic lethality and nullifies cytological defects associated with wild type CI caused by a wMel infection [43] . As such , we recently proposed the Two-by-One genetic model of CI wherein dual expression of cifAwMel and cifBwMel causes CI when expressed in males and expression of cifAwMel rescues CI when expressed in females [43] . However , confirmation of the model’s central prediction requires the complete synthetic replication of CI-induced lethality and rescue in the absence of any Wolbachia infections since it remains possible that other Wolbachia or phage WO genes besides cifA and cifB contribute to wild type CI and rescue by wMel Wolbachia . Moreover , CI induced by dual cifAwMel and cifBwMel expression previously yielded variable offspring lethality with a median survival of 26 . 5% of embryos relative to survival of 0 . 0% of embryos from CI induced by a wild type infection under controlled conditions [38] . The inability to recapitulate strong wild type CI suggests other CI genes are required , other environmental factors need to be controlled , or the transgenic system requires optimization . Here , we utilize transgenic expression , hatch rates , and gene expression assays in Drosophila melanogaster to test if an optimized expression system can generate strong transgenic CI and whether bacteriophage genes cifAwMel and cifBwMel can fully control fly reproduction by inducing and rescuing CI in the complete absence of Wolbachia ( Fig 1 ) . We further assess if both cifwMel genes are required for CI induction in the optimized system and whether cifAwMel in females can rescue transgenic CI . Results provide strong evidence for the Two-by-One model in wMel-infected D . melanogaster , offer context for conceptualizing CI mechanisms and the evolution of bidirectional incompatibilities between different Wolbachia strains , raise points for CI gene nomenclature , and motivate further research in developing these genes into a tool that combats vector borne diseases . To the best of our knowledge , they also represent the first case of completely engineering animal sexual reproduction to depend upon bacteriophage gene products .
Dual transgenic expression of cifAwMel and cifBwMel was previously reported to induce highly variable and incomplete CI relative to CI caused by an age-controlled wMel infection [38] , indicating either the presence of other genes necessary for strong CI , environmental factors uncontrolled in the study , or inefficiency of the transgenic system . Here , we test the latter hypothesis by dually expressing cifAwMel and cifBwMel in uninfected D . melanogaster males under two distinct GAL4 driver lines that express in reproductive tissues: nos-GAL4-tubulin and nos-GAL4:VP16 [44] . Both driver lines contain a nos promoter region , but differ in that nos-GAL4-tubulin produces a transcription factor with both the DNA binding and transcriptional activating region of the GAL4 protein , and nos-GAL4:VP16 produces a fusion protein of the GAL4 DNA binding domain and the virion protein 16 ( VP16 ) activating region [45 , 46] . The GAL4:VP16 transcription factor is a particularly potent transcriptional activator because of its binding efficiency to transcription factors [47 , 48] . Additionally , the nos-GAL4-tubulin driver has a tubulin 3’ UTR , and nos-GAL4:VP16 has a nos 3’ UTR that may contribute to differences in localization within cells or between tissues [44–46] . As such , we predict that differences in the expression level or profile of these two driver lines will lead to differences in the penetrance of transgenic CI . Since CI manifests as embryonic lethality , we measure hatching of D . melanogaster embryos into larvae to quantify the strength of CI . We confirm previous findings [38] that dual transgenic expression of cifAwMel and cifBwMel under nos-GAL4-tubulin in uninfected males yields low but variable embryonic hatching in crosses with uninfected females ( Mdn = 26 . 3% , IQR = 10 . 4–38 . 1% ) that can be rescued in crosses with wMel-infected females ( Mdn = 97 . 5%; IQR = 94 . 2–100% ) ( Fig 2A ) . However , dual cifAwMel and cifBwMel expression under nos-GAL4:VP16 in uninfected males yields significantly reduced embryonic hatching relative to nos-GAL4-tubulin ( p = 0 . 0002 ) with less variability ( Mdn = 0%; IQR = 0 . 0–0 . 75% ) and can be comparably rescued ( Mdn = 98 . 65%; IQR = 95 . 93–100%; p > 0 . 99 ) ( Fig 2A ) . Together , these results support that dual cifAwMel and cifBwMel expression under nos-GAL4:VP16 induces the strongest CI and that the transgenic system , not the absence of necessary CI factors , contributed to the prior inability to recapitulate strong wild type CI . Next , we tested the hypothesis that differences in the penetrance of transgenic CI between the two drivers are due to differences in the strength of expression . To assess this , we used qPCR to measure the gene expression of cifAwMel and cifBwMel under the two drivers relative to a Drosophila housekeeping gene ( rp49 ) in male abdomens ( Fig 2B and 2C ) . Fold differences in RNA transcripts of cifAwMel relative to rp49 reveal nos-GAL4-tubulin ( Mdn = 0 . 0098; IQR = 0 . 0082–0 . 122 ) drives significantly stronger and more variable cifAwMel expression relative to nos-GAL4:VP16 ( Mdn = 0 . 0075; IQR = 0 . 0064–0 . 0090 ) ( p = 0 . 016 , MWU , Fig 2B ) . The same is true for cifBwMel expression where nos-GAL4-tubulin ( Mdn = 0 . 022; IQR = 0 . 0165–0 . 0265 ) drives significantly stronger cifBwMel expression than nos-GAL4:VP16 ( Mdn = 0 . 0168; IQR = 0 . 0135–0 . 0179 ) ( p = 0 . 02 , MWU , Fig 2C ) . Moreover , while cifAwMel and cifBwMel expression significantly correlate with each other under both nos-GAL4-tubulin ( R2 = 0 . 85; p <0 . 0001 ) and nos-GAL4:VP16 ( R2 = 0 . 75; p <0 . 0001; S1A Fig ) , neither cifAwMel ( R2 = 0 . 02; p = 0 . 62; S1B Fig ) nor cifBwMel ( R2 = 0 . 04; p = 0 . 48; S1C Fig ) expression levels under the nos-GAL4-tubulin driver correlate with the strength of CI measured via hatch rates . Notably , cifBwMel is consistently more highly expressed than cifAwMel within the same line ( S1A Fig ) . We predict that expression differences are due to either differences in transgenic insertion sites or more rapid degradation of cifAwMel relative to cifBwMel . Taken together , these results suggest that an increase in CI penetrance in these crosses is not positively associated with higher transgene transcript abundance from different drivers . cifAwMel expression under the maternal triple driver ( MTD ) in uninfected females can rescue CI induced by a wild type infection [43] . MTD is comprised of three drivers in the same line: nos-GAL4-tubulin , nos-GAL4:VP16 , and otu-GAL4:VP16 [44] . We previously reported that cifAwMel expression under the nos-GAL4-tubulin driver alone is rescue-incapable [43] . Here , we test if cifAwMel expression under either of the other components of the MTD driver independently recapitulate rescue of wMel CI . Hatch rate experiments indicate that CI is strong and expectedly not rescued when an infected male mates with a non-transgenic female whose genotype is otherwise nos-GAL4:VP16 ( Mdn = 0 . 0%; IQR = 0 . 0–0 . 0% ) or otu-GAL4:VP16 ( Mdn = 0 . 0%; IQR = 0 . 0–0 . 0% ) ( Fig 3A ) . Transgenic expression of cifAwMel in uninfected females under either of the two drivers rescues CI induced by wMel . However , rescue is significantly weaker under cifAwMel expression with the otu-GAL4:VP16 driver ( Mdn = 70 . 4%; IQR = 0 . 0–90 . 45% ) as compared to the nos-GAL4:VP16 driver ( Mdn = 94 . 2%; IQR = 83 . 3–97 . 1%; p = 0 . 0491 ) which produced strong transgenic rescue ( Fig 3A ) . Gene expression analysis of cifAwMel relative to rp49 in the abdomens of uninfected females reveals that nos-GAL4:VP16 expresses cifAwMel significantly higher ( Mdn = 1 . 08; p < 0 . 0001 ) than otu-GAL4:VP16 ( Mdn = 0 . 03 ) ( Fig 3B ) , suggesting that high expression in females may underpin the ability to rescue . Alternatively , nos-GAL4:VP16 and otu-GAL4:VP16 are known to express GAL4 at different times in oogenesis , with the former in all egg chambers and the latter in late stage egg chambers [44] . With the transgenic expression system optimized for both transgenic CI and rescue , we then tested the hypothesis that the Two-by-One model can be synthetically recapitulated by dual cifAwMel and cifBwMel expression in uninfected males to cause CI and single cifAwMel expression in uninfected females to rescue that transgenic CI . Indeed , dual cifAwMel and cifBwMel expression in uninfected males causes hatch rates comparable to wild type CI ( Mdn = 0 . 0%; IQR = 0 . 0%-2 . 55; p > 0 . 99 ) ( Fig 4 ) . Transgenic CI cannot be rescued by single cifBwMel expression in uninfected females ( Mdn = 1 . 25%; IQR = 0 . 0–3 . 35% ) . Transgenic CI can be rescued by single cifAwMel expression ( Mdn = 98 . 6%; IQR = 97 . 35–100%; p = 0 . 41 ) or dual cifAwMel and cifBwMel expression ( Mdn = 96 . 7%; IQR = 88 . 3–98 . 2%; p > 0 . 99 ) to levels comparable to rescue from a wild type infection ( Mdn = 95 . 6%; IQR = 92 . 5–97 . 4% ) . In addition , cifAwMel rescues a wild type infection at comparable levels to wild type rescue ( Mdn = 96 . 6%; IQR = 93 . 5–98 . 85%; p > 0 . 99 ) . These data provide strong evidence for the Two-by-One model in wMel-infected D . melanogaster , namely that CI induced by transgenic dual cifAwMel and cifBwMel expression is sufficient to induce strong CI , and that cifAwMel alone is sufficient to rescue it . Next we reevaluated if single cifAwMel or cifBwMel expression under the more potent nos-GAL4:VP16 driver in uninfected males can recapitulate CI . Hatch rates indicate that dual cifAwMel and cifBwMel expression induces strong transgenic CI ( Mdn = 0 . 0%; IQR = 0 . 0–1 . 15% ) that can be rescued by a wild type infection ( Mdn = 93 . 8%; IQR = 88 . 2–97 . 4% ) , whereas single expression of cifAwMel ( Mdn = 96 . 1%; IQR = 97 . 78–98 . 55%; p < 0 . 0001 ) or cifBwMel ( Mdn = 92 . 85%; IQR = 84 . 28–96 . 4%; p < 0 . 0001 ) failed once again to produce embryonic hatching comparable to expressing both genes together ( Fig 5 ) . In one replicate experiment , we note a statistically insignificant ( p = 0 . 182 ) decrease in hatching under cifBwMel expression relative to wild type rescue cross ( S1 Data file ) . Thus , both cifAwMel and cifBwMel are required for strong CI . Together , these and earlier results validate the Two-by-One model of CI in wMel whereby cifAwMel and cifBwMel expression are required and sufficient for strong CI , while cifAwMel expression is sufficient to rescue it .
CI is the most common form of Wolbachia-induced reproductive parasitism and is currently at the forefront of vector control efforts to curb transmission of dengue , Zika , and other arthropod-borne human pathogens [22–25 , 34 , 35] . Two prophage WO genes from wMel Wolbachia cause CI ( cifAwMel and cifBwMel ) and one rescues wild type CI ( cifAwMel ) [38 , 43] , supporting the proposal of a Two-by-One model for the genetic basis of CI [43] . However , dual transgenic expression of cifAwMel and cifBwMel recapitulates only weak and highly variable CI as compared to CI induced by a wild type infection [38] . In addition , the Two-by-One model predicts that both CI and rescue can be synthetically recapitulated by dual cifAwMel and cifBwMel expression in uninfected males and cifAwMel expression in uninfected females . Here we optimized the transgenic system for CI and rescue by these genes , further validated the necessity of expressing both cifAwMel and cifBwMel for CI , and synthetically recapitulated the Two-by-One model for CI with transgenics in the absence of Wolbachia . CI induced by wMel Wolbachia can be highly variable and correlates with numerous factors including Wolbachia density [49] , cifAwMel and cifBwMel expression levels [38] , host age [50–52] , mating rate [50] , rearing density [53] , development time [53] , and host genetic factors [52 , 54–56] . Some of these factors , such as age , are known to also correlate with the level of cifwMel gene expression [38] . As such , we hypothesized that prior reports of weakened transgenic CI could be explained by low levels of transgenic cifAwMel and cifBwMel expression in male testes [38] . Indeed , strong CI with a median of 0% embryonic hatching was induced when both cifAwMel and cifBwMel were expressed under the nos-GAL4:VP16 driver . However , contrary to our expectations , nos-GAL4:VP16 generates significantly weaker cifAwMel and cifBwMel expression than the nos-GAL4-tubulin driver previously used to recapitulate weak CI [38] . Thus , the expression data conflict with previous reports in mammalian cells wherein the GAL4:VP16 fusion protein is a more potent transcriptional activator than GAL4 [48] . Other differences between the two driver constructs may explain phenotypic differences , including the presence of different 3’ UTRs that may contribute to differences in transcript localization [44] . While it remains possible , though unlikely , that other Wolbachia or phage WO genes may contribute to CI , the induction of near complete embryonic lethality confirms that cifAwMel and cifBwMel are sufficient to transgenically induce strong CI and do not require other Wolbachia or phage WO genes to do so . Moreover , comparative multi-omics demonstrated that cifA and cifB are the only two genes strictly associated with CI capability [38] . We previously recapitulated transgenic rescue of wMel-induced CI by expression of cifAwMel under the Maternal Triple Driver ( MTD ) [43] , which is comprised of three independent drivers [44] . Expression of cifAwMel using one of the MTD drivers in flies was previously shown to be rescue-incapable [43]; the other drivers had not been evaluated . Here , we tested the hypothesis that expression of cifAwMel using either of the two remaining drivers is sufficient to rescue CI , and we found that cifAwMel expression under both driver lines recapitulates rescue , but at different strengths . Indeed , rescue is strongest when cifAwMel transgene expression is highest . These data are consistent with reports that cifAwMel is a highly expressed gene in transcriptomes of wMel-infected females [57] and the hypothesis that rescue capability is largely determined by the strength of cifAwMel expression in ovaries [43] . These results combined with those for transgenic expression of CI now establish a robust set of methods for future studies of transgene-induced CI and rescue in the D . melanogaster model . The central prediction of the Two-by-One model is that transgenic CI can be synthetically rescued in the absence of Wolbachia through dual cifA and cifB expression in uninfected males and cifA expression in uninfected females . Here , we explicitly validate the model that two genes are required in males to cause CI , and one in females is required to rescue it using wMel cif gene variants . However , to confirm that the optimized expression system does not influence the ability of cifAwMel or cifBwMel alone to induce CI , we singly expressed them with the improved driver and found that embryonic hatching does not statistically differ from compatible crosses . Coupled with prior data in wMel [38 , 43] , these results strongly support the Two-by-One genetic model whereby dual cifAwMel and cifBwMel expression is required in the testes to cause a sperm modification that can then be rescued by cifAwMel expression in the ovaries ( Fig 6A ) . While the genetic basis of unidirectional CI appears resolved , it remains unclear how cifAwMel and cifBwMel functionally operate to generate these phenotypes . Numerous mechanistic models have been proposed over the last two decades [58–64] . We can broadly summarize these models into either host-modification ( HM ) [59] or toxin-antidote ( TA ) [58] models . HM models suggest that CI-inducing factors modify host products in such a way that would be lethal unless they are later reversed by rescue factors [59–64] . Conversely , TA models state that the CI-inducing factor is toxic to the developing embryo unless it is crucially bound to a cognate antidote provided by the female [42 , 58 , 59] . There are numerous lines of evidence in support of both sets of hypotheses and while the Two-by-One genetic model does not explicitly support or favor one set of models over the other , it can be used to generate hypotheses related to the mechanism of CI . HM models [59] predict that CI factors directly interact with host products in the testes , modify them , and are displaced . These modifications travel with the sperm , in the absence of Wolbachia and Cif products , and would induce the canonical cytological embryonic defects including delayed paternal nuclear envelope breakdown , slowed Cdk1 activation , a failure of maternal histones to deposit onto the paternal genome , stalled or failed replication of the paternal DNA , a failure of paternal chromosomes to segregate , and later stage regional mitotic failures [7 , 38 , 60 , 61 , 64–67] , or they are reversed by female-derived rescue factors . Leading HM models are the Mistiming [60 , 61] and Goalkeeper [63] models that leverage findings that male pronuclei are delayed in the first mitosis during embryonic development in CI crosses [61 , 65 , 67] . Since the first mitosis is initiated when the female pronucleus has developed , the delay of the male pronuclei leads to cytological defects [60] . It is thus proposed that rescue occurs through resynchronization of the first mitosis by comparably delaying the female pronucleus [60 , 61] . The Goalkeeper model expands the mistiming model to propose that the strength of the delay is what drives incompatibility between different Wolbachia strains [63] . There are numerous hypotheses to explain the role of the Cif products in these kinds of models . One such hypothesis would be that CifA is responsible for pronuclear delay , thus capable of delaying both the male and female pronuclei , but it requires CifB to properly interact with testis-associated targets . This hypothesis may predict that CifB acts to either protect CifA from ubiquitin tagging and degradation , localize it to a host target , or bind CifA to elicit a conformational change required for interacting with male-specific targets . Alternatively , CI-affected embryos express defective paternal histone deposition , protamine development , delayed nuclear breakdown , and delays in replication machinery [7 , 60 , 61 , 64–67] . Any of these factors could be explained by modifications occurring from HM-type interactions between Cif and host products . TA models [58] contrast to HM models and require that the CI toxin transfers with or in the sperm and directly binds to a female-derived antidote in the embryo . If the antidote is absent , the CI toxin would induce cytological embryonic defects [7 , 38 , 60 , 61 , 64–67] . There is mixed evidence in support of this model . First , mass spectometry and SDS-PAGE analyses in Culex pipiens reveal that CifAwPip peptides are present in female spermatheca after mating , suggesting CifAwPip is transferred with or in the sperm [39] . CifBwPip was not detected in these analyses , curiously suggesting that the CifB toxin was not transferred [39] . These results are inconsistent with the TA model , but the lack of transferred CifB may occur because cifB gene expression is up to nine-fold lower than that of cifA [57] , and the concentration may have been too low to be observed via these methods . Second , CifA and CifB bind in vitro [42] . However , it remains unclear if CifA-CifB binding enables rescue since this binding has no impact on known enzymatic activities of CifB [42] . While the Two-by-One model does not explicitly support or reject the TA model , it does further inform it . Most intriguing is to understand how CifA acts as a contributor to CI when expressed in testes and as a rescue factor when expressed in ovaries . One hypothesis is that CifA and CifB bind to form a toxin complex that is later directly inhibited by female derived CifA [43 , 59] . The difference in function between these two environments could be explained by post-translational modification and/or differential localization of CifA in testes and embryos [43 , 59] . Alternatively , CifB may be the primary toxin , but is incapable of inducing CI unless a CifA antidote is present in both the testes and the ovaries [58] . This hypothesis predicts that male-derived CifA rapidly degrades , leaving CifB with or in the sperm . On its own , CifB would induce lethal cytological embryonic defects [60–62 , 64] unless provided with a fresh supply of CifA from the embryo . It has been suggested that divergence in CI and rescue factors causes the incipient evolution of reciprocal incompatibility , or bidirectional CI , between different Wolbachia strains [38 , 43 , 68 , 69] . Here , we review a non-exhaustive set of hypotheses that we previously proposed to explain the emergence of bidirectional CI and are consistent with the Two-by-One model [43] . First , the simplest explanation for CifA’s role in both CI and rescue is that it has similar functional effects in both testes/sperm and ovaries/embryos . Thus , instead of requiring a separate mutation for CI and another for rescue [69] , bidirectional CI may emerge from a single CifA mutation that causes incompatibility against the ancestral strain while maintaining self-compatibility . Second , CifA in testes and ovaries may also have different functions , localizations , or posttranslational modifications that contribute to CI and rescue . If this occurs , or if CifB is also an incompatibility factor , the evolution of bidirectional CI may require two or more mutations , and the strain may pass through an intermediate phenotype wherein it becomes unidirectionally incompatible with the ancestral variant or loses the capability to induce either CI or rescue before becoming bidirectionally incompatible with the ancestral variant . In fact , some Wolbachia strains are incapable of inducing CI but capable of rescuing CI induced by other strains [70] , and some can induce CI but cannot be rescued [71] . Furthermore , sequence variation in both cifA and cifB from Wolbachia strains in Drosophila [38] and in small regions among strains of wPip Wolbachia [68] have been correlated to incompatibility , suggesting that variation in both genes influence incompatibility . Additionally , it remains possible that significant divergence in cifA , cifB , or both may be necessary to generate new phenotypes . Indeed , comparative genomic analyses reveal high levels of amino acid divergence in CifA and CifB that correlates with incompatibility between strains [38 , 40] . Moreover , some Wolbachia strains harbor numerous phage WO variants , each with their own , often divergent , cif genes , and the presence of multiple variants likewise correlates with incompatibility [38 , 40 , 68] . Thus , horizontal transfer of phage WO [37 , 72–76] can in theory rapidly introduce new compatibility relationships , and duplication of phage WO regions , or specifically cif genes , in the same Wolbachia genome may relax the selective pressure on the cif genes and enable their divergence . Determining which of the aforementioned models best explains the evolution of incompatibilities between Wolbachia strains will be assisted by additional sequencing studies to identify incompatible strains with closely related cif variants . The genetic bases of numerous gene drives have been elucidated in plants [77] , fungi [78–81] , and nematodes [82 , 83] . Some gene drives have also been artificially replicated with transgenic constructs [84–86] . However , to our knowledge , the synthetic replication of the Two-by-One model of CI represents the first instance that a gene drive has been constructed by engineering eukaryotic reproduction to depend on phage proteins . Additionally , vector control programs using Wolbachia rely on their ability to suppress pathogens such as Zika and dengue viruses , reduce the size of vector populations , and spread Wolbachia into a host population via CI and rescue . However , there are limitations to these approaches . Most critically , not all pathogens are inhibited by Wolbachia infection and some are enhanced , such as West Nile Virus in Culex tarsalis infected with wAlbB Wolbachia [87] . Additionally , it requires substantial effort to establish a Wolbachia transinfection in a target non-native species [88] that could be obviated in genetically tractable vectors utilizing transgenic gene drives . The complete synthetic replication of CI and rescue via the Two-by-One model represents a step towards transgenically using the cif genes in vector control efforts . The separation of CI mechanism from Wolbachia infection could theoretically expand CI’s utility to spread ‘payload’ genes that reduce the vectoral capacity of their hosts [89] into a vector population by , for instance , expressing the CI genes and the payload gene polycistronically under the same promoter in the vector’s nuclear or mitochondrial genomes . Moreover , these synthetic constructs have potential to increase the efficiency of Wolbachia-induced CI if they are transformed directly into Wolbachia genomes . For these efforts to be successful , considerable work is necessary to ( i ) generate a constitutively expressing cif gene drive that does not require GAL4 to operate , ( ii ) understand the spread dynamics of transgenic CI , ( iii ) characterize the impact of cif transgenic expression on insect fitness relative to wild vectors , ( iv ) generate and test effective payload genes in combination with cif drive , ( v ) explore and optimize the efficacy of cif drive in vector competent hosts such as mosquitoes , ( vi ) assess the impact of host factors on cif drive across age and development , ( vii ) compare the efficacy of a cif gene drive to other comparable technologies ( CRISPR , homing drive , Medea , etc ) , in addition to numerous other lines of study . For example , while a substantial body of literature exists to describe the spread dynamics of CI [10 , 12 , 13 , 36 , 90 , 91] , none yet describe how the Two-by-One model would translate into nuclear or mitochondrial spread dynamics in the absence of Wolbachia . As such , this study represents an early proof of concept that these genes alone are capable of biasing offspring survival in favor of flies expressing these genes under strictly controlled conditions , and should motivate additional study towards its application in vector control . The generality of the Two-by-One model remains to be tested because it may be specific to certain strains of Wolbachia and/or phage haplotypes . For instance , transgenic expression of cifBwPip from C . pipiens in yeast yields temperature sensitive lethality that can be rescued by dual-expression of cifAwPip and cifBwPip [42] . Moreover , attempts to generate a cifBwPip transgenic line failed , possibly due to generalized toxicity from leaky expression [42] . Therefore , cifBwPip alone could in theory cause CI . However , this model has not been explicitly tested , it has not been explained how cifAwPip and cifBwPip dual-expression induces CI in transgenic Drosophila but prevents CI in yeast , and transgenic wPip CI has not been rescued in an insect . As such , it remains possible that cifBwPip lethality could be explained by artefactual toxicity of overexpression or toxic expression in a heterologous system . Thus , confirmation of an alternative model for CI in wPip is precluded by lack of evidence that cifBwPip alone can induce rescuable lethality in an insect . Since cifBwPip transgenic UAS constructs have not been generated due to toxicity from leaky expression , alternative PhiC31 landing sites or expression systems ( i . e . , the Q System ) could prove valuable in addressing these questions . Finally , these results further validate the importance of cifAwMel as an essential component of CI and underscore a community need to unify the nomenclature of the CI genes . When the CI genes were first reported , they were described as both CI factors ( cif ) and as CI deubiquitilases ( cid ) , both of which are actively utilized in the literature . The cif nomenclature was proposed as a cautious naming strategy agnostic to the varied biochemical functions to be discovered , whereas the cid nomenclature was proposed based on the finding that the B protein is in part an in vitro deubiquitilase that , when ablated , inhibits CI-like induction [38 , 42] . A recent nomenclature proposal suggested that the cif gene family name be used as an umbrella label to describe all CI-associated factors whereas cidA and cidB would be used to describe the specific genes [58] . However , we do not agree with this nomenclature revision despite the appeal of combining the two nomenclatures . CifA protein is not a putative deubiquitilase [40] , does not influence deubiquitilase activity of CifB [42] , functions independently to rescue CI [43] and , as emphasized by the work in this study , is necessary for CI induction and rescue . The competing nomenclature presumes that it is appropriate to name the A protein cid because it could be expressed in an operon with the B protein . However , the evidence for the operon status of the genes is weak , and more work is needed to describe the regulatory control of these genes before they can be categorized as an operon [59] . Moreover , distant homologs that cluster into distinct phylogenetic groups are proposed to be named CI nucleases ( cin ) [42] yet the merger of these two groups into one name lacks phylogenetic rationality as the two lineages are as markedly divergent from each other as they are from cid [59] . In addition , none of these distant homologs have been functionally characterized as CI genes [38 , 40] . As such , it is more appropriate to call these genes “cif-like” to reflect their homology and unknown phenotypes . Thus , the holistic and conservative cif nomenclature with Types ( e . g . , I-IV ) used to delineate phylogenetic clades is appropriately warranted in utilizing and unifying CI gene naming . In conclusion , the results presented here support that both cifAwMel and cifBwMel phage genes are necessary and sufficient to induce strong CI . In addition , cifAwMel is the only gene necessary for rescue of either transgenic or wild type wMel CI . These results confirm the Two-by-One model of CI in wMel Wolbachia and phage WO with implications for the mechanism of CI and for the diversity of incompatibility between strains , and they provide additional context for understanding CI currently deployed in vector control efforts . The synthetic replication of CI in the absence of Wolbachia marks an early step in developing CI as a tool for genetic and mechanistic studies in D . melanogaster and for vector control efforts that may drive payload genes into vector competent populations .
D . melanogaster stocks y1w* ( BDSC 1495 ) , nos-GAL4-tubulin ( BDSC 4442 ) , nos-GAL4:VP16 ( BDSC 4937 ) , otu-GAL4:VP16 ( BDSC 58424 ) , and UAS transgenic lines homozygous for cifA , cifB , and cifA;B [38] were maintained at 12:12 light:dark at 25o C and 70% relative humidity ( RH ) on 50 ml of a standard media . cifA insertion was performed with y1 M{vas-int . Dm}ZH-2A w*; P{CaryP}attP40 and cifB insertion was performed with y1 w67c23; P{CaryP}attP2 , as previously described [38] . UAS transgenic lines and nos-GAL4:VP16 were uninfected whereas nos-GAL4-tubulin and otu-GAL4:VP16 lines were infected with wMel Wolbachia . Uninfected versions of infected lines were produced through tetracycline treatment as previously described [38] . WolbF and WolbR3 primers were regularly used to confirm infection status [38] . Stocks for virgin collections were stored at 18o C overnight to slow eclosion rate , and virgin flies were kept at room temperature . To test for CI , hatch rate assays were used as previously described [38 , 43] . Briefly , GAL4 adult females were aged 9–11 days post eclosion and mated with UAS males . Age controlled GAL4-UAS males and females were paired in 8 oz bottles affixed with a grape-juice agar plate smeared with yeast affixed to the opening with tape . 0–48 hour old males were used since CI strength rapidly declines with male age [50 , 52] . The flies and bottles were stored at 25o C for 24 h at which time the plates were replaced with freshly smeared plates and again stored for 24 h . Plates were then removed and the number of embryos on each plate were counted and stored at 25o C . After 30 h the remaining unhatched embryos were counted . The percent of embryos hatched into larvae was calculated by dividing the number of hatched embryos by the initial embryo count and multiplying by 100 . To assay transgenic RNA expression levels under the various gene drive systems , transgene expressing flies from hatch rates were immediately collected and frozen at -80°C for downstream application as previously described [43] . In brief , abdomens were dissected , RNA was extracted using the Direct-zol RNA MiniPrep Kit ( Zymo ) , the DNA-free kit ( Ambion , Life Technologies ) was then used to remove DNA contamination , and cDNA was generated with SuperScript VILO ( Invitrogen ) . Quantitative PCR was performed on a Bio-Rad CFX-96 Real-Time System in duplicate using iTaq Universal SYBR Green Supermix ( Bio-Rad ) using the cifA_opt and rp49 forward and reverse primers as previously described [43] . Samples with a standard deviation >0 . 3 between duplicates were excluded from analysis . Fold expression of cifA relative to rp49 was determined with 2−ΔΔCt . Each expression study was conducted once . All statistical analyses were conducted in GraphPad Prism ( Prism 8 ) . Hatch rate statistical comparisons were made using Kruskal-Wallis followed by a Dunn’s multiple comparison test . A Mann-Whitney-U was used for statistical comparison of RNA fold expression . A linear regression was used to assess correlations between hatch rate and expression . All p-values are reported in S1 Table . | Releases of Wolbachia-infected mosquitos are underway worldwide because Wolbachia block replication of Zika and Dengue viruses and spread themselves maternally through arthropod populations via cytoplasmic incompatibility ( CI ) . The CI drive system depends on a Wolbachia-induced sperm modification that results in embryonic lethality when an infected male mates with an uninfected female , but this lethality is rescued when the female and her embryos are likewise infected . We recently reported that the phage WO genes , cifA and cifB , cause the sperm modification and cifA rescues the embryonic lethality caused by the wMel Wolbachia strain deployed in vector control . These reports motivated proposal of the Two-by-One model of CI whereby two genes cause lethality and one gene rescues it . Here we provide unequivocal support for the model in the Wolbachia strain used in vector control via synthetic methods that recapitulate CI and rescue in the absence of a Wolbachia infections . Our results reveal the set of phage WO genes responsible for this powerful genetic drive system , act as a proof-of-concept that these genes alone can induce gene drive like crossing patterns , and establish methodologies and hypotheses for future studies of CI in Drosophila . We discuss the implications of the Two-by-One model towards functional mechanisms of CI , the emergence of incompatibility between Wolbachia strains , vector control applications , and CI gene nomenclature . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] | [
"bacteriology",
"invertebrates",
"medicine",
"and",
"health",
"sciences",
"bacteriophages",
"vector-borne",
"diseases",
"microbiology",
"animals",
"wolbachia",
"animal",
"models",
"viruses",
"developmental",
"biology",
"drosophila",
"melanogaster",
"model",
"organisms",
"ex... | 2019 | Two-By-One model of cytoplasmic incompatibility: Synthetic recapitulation by transgenic expression of cifA and cifB in Drosophila |
Flaviviruses deliver their genome into the cell by fusing the viral lipid membrane to an endosomal membrane . The sequence and kinetics of the steps required for nucleocapsid delivery into the cytoplasm remain unclear . Here we dissect the cell entry pathway of virions and virus-like particles from two flaviviruses using single-particle tracking in live cells , a biochemical membrane fusion assay and virus infectivity assays . We show that the virus particles fuse with a small endosomal compartment in which the nucleocapsid remains trapped for several minutes . Endosomal maturation inhibitors inhibit infectivity but not membrane fusion . We propose a flavivirus cell entry mechanism in which the virus particles fuse preferentially with small endosomal carrier vesicles and depend on back-fusion of the vesicles with the late endosomal membrane to deliver the nucleocapsid into the cytoplasm . Virus entry modulates intracellular calcium release and phosphatidylinositol-3-phosphate kinase signaling . Moreover , the broadly cross-reactive therapeutic antibody scFv11 binds to virus-like particles and inhibits fusion .
Many enveloped RNA viruses utilize the endocytic pathway to enter host cells [1] , [2] . Endocytosis begins at the cell membrane , where these viruses bind to their cellular receptors and ends at the lysosome , the “stomach” of the cell . Along the endocytic pathway , changes in the lipid composition and environmental pH provide a series of distinct milieus for specific cellular or viral functions to occur [3] . Enveloped viruses and bacterial toxins enter the endocytic pathway by binding receptors on the cell surface that are coupled to the endocytic machinery , in particular clathrin adaptors . These microbial cargoes undergo sorting at two different checkpoints [4] , [5] , [6] . The first is in early endosomes ( EEs ) where the vesicular contents are either directed back to the cell membrane via tubular structures , or targeted to the trans-Golgi network ( TGN ) . Alternatively , the cargo contents are sorted into intraluminal vesicles and transported to late endosomes via endosomal carrier vesicles ( ECVs ) . ECVs require functional microtubules to be transported to the second sorting station , the late endosomes . In late endosomes , cargo contents can be forwarded to the TGN , the cytoplasm , or for lysosomal degradation . ECVs originate from EEs . Both the EEs and ECVs are rich in cholesterol , phosphatidylserine ( PS ) and phosphatidylinositols ( PI ) [7] , [8] , [9] . The level of cholesterol decreases along the endocytic pathway and is replaced with ceramide in late endosomes and lysosomes , where it maintains membrane fluidity [10] . Unlike cholesterol and PS , the anionic lipid BMP ( bis ( monoacylglycero ) phosphate ) , also known as LBPA ( lysobisphosphatidic acid ) , is abundant in internal membranes of lysosomes and late endosomes , and depleted in the EEs [7] . BMP regulates membrane sorting and dynamics in the late endosome . Autoantibodies against this lipid result in human disorders such as Niemann-Pick type C ( NPC ) syndrome , characterized by dysfunctional sorting and trafficking in late endosomes [11] . The genus flavivirus includes important human pathogens such as dengue , Japanese encephalitis ( JE ) , West Nile ( WN ) and yellow fever ( YF ) viruses . Flaviviruses contain a lipid envelope and a positive-stranded RNA genome encoding for a polyprotein that is processed by the host- and viral proteases to yield the viral proteins . Three structural proteins ( C , M and E ) form the virions; the nonstructural proteins ( NS1-5 ) are required for virus replication , transcription and modulation of the host innate immune system [12] . Flaviviruses assemble in specialized structures within the endoplasmic reticulum and mature in the Golgi network [13] . Glycoprotein E forms the outer protein shell of the virion , mediates cellular attachment , and catalyzes the fusion of the viral and cellular membranes necessary to deliver the genome into the cytoplasm . The E ectodomain contains three domains ( I–III ) connected by hinges [14] , [15] . Conserved histidine residues at the domain I-domain III interface become protonated at the reduced pH of early endosomal compartments ( pH 6–6 . 5 ) , thereby triggering a conformational rearrangement in E that drives membrane fusion [16] , [17] . Although flaviviruses generally follow the clathrin-mediated endocytic pathway , other mechanisms of entry have also been proposed . Dengue virus has been reported to fuse primarily from within Rab7-positive late endosomes . However , certain dengue strains have been reported to infect cell independently of Rab7 [18] and alternative entry pathways have been proposed for specific viruses and cell types [19] , [20] . Moreover , in certain flaviviruses low pH does not appear to be sufficient to trigger fusion suggesting that additional factors may be required [21] . Indeed , different compartment-specific lipids are required for fusion of dengue and Japanese encephalitis viruses [22] , [23] , [24] , [25] . Flaviviruses produce small noninfectious virus-like particles ( VLPs ) during infection [26] . Little is known about the role of these VLPs in infection or in host immunity however , recombinant flavivirus VLPs are the subject of intense study as vaccine candidates and gene delivery vehicles [27] , [28] , [29] , [30] . It is not known whether VLPs have the same requirements for fusion and the same cell entry pathways as full-sized virions . In this study , we use JE-VLPs and YFV virions as model systems to study the cell flavivirus entry pathway . Using a combination of approaches—including single-particle tracking in live cells , a liposome-based membrane fusion assay , a quantitative RT-PCR RNA delivery assay , and viral infectivity assays—we show that our model viruses modulate cellular signal transduction to promote membrane fusion to ECVs , which occurs several minutes before nucleocapsid delivery into the cytoplasm , suggesting that these are two distinct events in virus entry . Our observations are consistent with an entry pathway in which certain flaviviruses fuse with ECVs and require host proteins to deliver the nucleocapsid to the cytoplasm . This pathway has a precedent in vesicular stomatitis virus ( VSV ) [5] , although VSV does not impact PI-3-phosphate kinase activity [31] , [32] nor trigger intracellular calcium release during entry [33] . Moreover , we demonstrated the ability of a broadly cross-reactive therapeutic antibody , scFv11 , to block membrane fusion with the host cells and EE/ECV-like liposomes .
Recombinant Japanese encephalitis virus-like particles ( JE-VLPs ) were obtained by overexpressing the prM and E genes ( see Materials and Methods ) in human HEK 293T cells ( Figure S1A ) , and in insect Tni cells using a baculovirus-based expression system ( Figure S1B ) . JE-VLPs and YFV were purified by precipitation of secreted cellular products with polyethylene glycol , followed by sedimentation in sucrose density gradient ( Figure S2 ) . Since flaviviruses E proteins bind to heparan sulfate [34] , the virus particles could alternatively be purified by affinity chromatography on a heparan sulfate column ( Figure S2A–B ) . Both methods showed highly purified secreted VLPs . Coomassie stained SDS-PAGE confirmed that prM was cleaved to pr and M in the purified particles ( Figure S2D ) , indicating a high degree of maturation in the JE-VLPs . The concentration of the purified VLPs was estimated by enzyme linked immunosorbent assay ( ELISA- see Materials and Methods ) . Negatively stained electron microscopy ( EM ) show that the JE-VLPs have rough surfaces and diameters ranging from 30 to 40 nm ( Figure S2F ) . Purified yellow fever virus ( YFV ) particles had a similar appearance but were 50 nm in diameter ( Figure S2G ) . Dynamic light scattering ( DLS ) analysis indicated an average diameter of approximately 40 nm for the JE-VLPs ( Figure S2E ) , consistent with the EM data . Upon attachment to the plasma membrane , flaviviruses localize to clathrin-coated pits and undergo endocytosis [35] , [36] . Although , VLPs , like full virions , are expected to enter the endocytic pathway , this has not yet been demonstrated experimentally . We treated Vero cells , a commonly used cell line to study flaviviruses , with JE-VLPs and stained fixed cell at different time points with anti-E protein and the endosomal markers Rab5 and Rab7 . VLPs attached to the Vero cells immediately . After 5 minutes , E protein colocalized with Rab5 . At 15 minutes , E protein colocalized with both Rab5 and Rab7 ( Figure 1A ) but more so with Rab7 , as indicated by the Pearson's coefficients of 0 . 19 and 0 . 34 for colocalization with Rab5 and Rab7 , respectively . By 25 minutes , most of the particles colocalized with Rab7 indicating their arrival to late endolysosomal compartments . This confirms that the secreted particles follow the same general cell entry pathway as full mature virions . To dissect the mechanism and kinetics of JE-VLP cell entry , we tracked the membrane fusion step in real time in live cells using confocal microscopy . The lipid envelope of JE-VLPs was labeled with self-quenching concentrations of the hydrophobic dye rhodamine C18 ( R18 ) . Fusion of the VLP membrane with an endosomal membrane was detected as a sudden dequenching of R18 fluorescence as the concentrated dye was diluted with lipids from the host membrane . Fluorescent puncta that showed no dequenching were excluded from our analyses . Fusion events were first detected approximately 250 s after treatment of Vero cells , consistent with fusion occurring in early to intermediate endosomal compartments ( Figure 1B–C ) . Notably , the R18 fluorescence signal for individual fusion events remained at its maximal level for 251±97 seconds ( n = 14 ) before starting to decay and the R18 dye remained concentrated in puncta during this fluorescence plateau ( Figure 1B–D ) . Fluorescence was expected to decay immediately after dequenching due to continued dilution with host lipids . The consistent persistence of fluorescent puncta for several minutes after fusion suggests that the R18 dye becomes trapped in an endosomal subcompartment after the initial membrane fusion event . The rate of fluorescence decay of the puncta after the fluorescence plateau , with a decay half-time of 94±64 ( n = 14 ) , is too great to be attributed to photobleaching alone , suggesting that the decay is due to a second and distinct event leading to dilution of the R18 dye to below detection levels ( Figure 1E ) . Taken together the particle tracking data suggest that JE-VLPs are fusogenic , that fusion occurs in early to intermediate endosomes , and that the viral lipids remain trapped in a small endosomal subcompartment for several minutes , until a distinct event releases the lipids into a much larger compartment . R18-labeled YFV strain 17D particles produced in BHK cells using an NS1 trans-complementation strategy described previously [37] had similar R18 dequenching kinetics , as judged from analysis of a smaller number of YFV particles . Chloroquine is a widely used lysosomotropic drug that acts by inhibiting the acidification of the endocytic pathway . Clinical studies have demonstrated the safety , tolerability , and efficacy of chloroquine as a treatment against enveloped RNA viruses [38] . Treatment with 194 µM ( 0 . 1 g/l ) chloroquine was not toxic to Vero cells ( Figure 2 ) . In the presence of chloroquine , fluorescence dequenching of R18 in R18-labeled JE-VLPs and YFV ( Figure 2B–C ) was completely inhibited , indicating that membrane fusion of JE-VLPs is dependent on the acidic pH of endosomal compartments . This is consistent with the inhibitory effect of chloroquine and other endosomal pH-neutralizing agents reported in other flaviviruses [35] , [39] . To confirm that mature flavivirus virions are also dependent on acidic endosomal pH , and that chloroquine inhibits not only fusion but also nucleocapsid delivery into the cytoplasm , we measured the effect of chloroquine on viral RNA release and infectivity of YFV . To measure delivery of YFV genomic RNA into the cytoplasm , infected Vero cells were fractionated into cytoplasmic and endosomal fractions ( Figure S3 ) , and viral RNA in the cytoplasmic fraction was detected by relative quantitative RT-PCR ( Figure 2D ) . Chloroquine blocked 95–97% of YFV RNA release in Vero cells . Moreover , in a plaque assay for YFV infectivity in BHK cells , chloroquine completely inhibited infectivity ( Figure 2E ) . These results confirm that the acidity of endosomal compartments is required for flavivirus infection , and that infection can be blocked with lysosomotropic drugs such as chloroquine that raise the endosomal pH . In mammalian cells , endosomal carrier vesicles ( ECVs ) are transported from early endosomes ( EEs ) to late endosomes on microtubules . ECVs then dock onto and fuse with late endosomal membrane [40] . Inhibition of microtubule-dependent transport with the microtubule depolymerizing agent nocodazole inhibits West Nile virus infection [35] . Treatment with 20 µM nocodazole was not toxic to the cells , although changes in cell morphology were observed in the treated cells ( Figure 3A ) . In cell pretreated with nocodazole , fluorescence dequenching of R18 in R18-labeled JE-VLPs and YFV occurred with similar kinetics as in untreated cells ( Figure 3 ) , indicating that nocodazole does not affect membrane fusion . However , the R18 fluorescence intensity gradually increased after fusion ( Figure 3B–C ) , rather than decaying after a 3–4 min plateau as in untreated cells ( Figure 1C–D ) . The gradual increase in fluorescence in the presence of nocodazole may be attributed to sequential homotypic fusion events with other non-fluorescent ECVs in early endosomal compartments . The resulting vesicles would be still relatively small ( hence the lack of dilution-dependent decay ) but would allow additional R18 dequenching ( hence the gradual increase in fluorescence ) . Additionally , inefficient lipid mixing in the presence of nocodazole may contribute to the gradual increase in fluorescence . Consistent with this interpretation , YFV RNA release into the cytoplasm , measured by RT-PCR as described above , was inhibited by 80% in the presence of nocodazole ( Figure 3D ) . Moreover , virus infectivity was completely inhibited by nocodazole ( Figure 3E ) . Together , these results indicate that certain flaviviruses fuse with ECVs , and that membrane fusion and RNA delivery into the cytoplasm are two distinct events . Having established that membrane fusion occurs early in the endocytic pathway whereas the YFV nucleocapsid is delivered into a late endosomal compartment , we sought next to determine the importance of factors specific to late endosomal compartments for fusion and nucleocapsid delivery . BMP ( also known as LBPA ) is an anionic lipid that is present in internal membranes , but not the limiting membrane , of late endosomes . An antibody against BMP accumulates on these internal membranes [41] and interferes with the protein sorting and membrane transport functions of the late endosome . Treatment with anti-BMP antibody causes a phenotype characteristic of Niemann-Pick disease type C ( NPC ) [11] , [42] . To assess the role of late endosomal trafficking in flavivirus cell entry , we incubated Vero cells with an anti-BMP antibody and assay membrane fusion activity , RNA delivery and infectivity . The staining patterns of endocytosed anti-BMP antibody and of a total mouse IgG control in BHK cells are shown in Figure 4 . Pretreatment with anti-BMP antibody followed by infection with R18-labeled JE-VLPs or YFV produced similar R18 fluorescence profiles as in cells pretreated with nocodazole , with normal R18 dequenching kinetics but no fluorescence decay ( Figure 4C–D ) . We conclude that membrane fusion is not inhibited by blocking the protein and lipid sorting functions of late endosomes . In contrast , the endocytosed anti-BMP antibody reduced both YFV RNA delivery to the cytoplasm of Vero cells and YFV infectivity in BHK cells by 35% ( Figure 4E–F ) . These data suggest that BMP in internal late endosomal membranes is required for virus infectivity , either to ensure correct ECV trafficking , or possibly to promote “back-fusion” of ECVs to the limiting membrane of the late endosome . In our emerging model of flavivirus cell-entry , virions fuse with ECVs and the nucleocapsid is delivered into the cytoplasm when the ECVs fuse back to the limiting late endosomal membrane . To test this model , we set out to evaluate the importance of factors required for ECV formation for fusion and nucleocapsid delivery and for trafficking of ECV to the late endosome . The lipid phosphatidylinositol-3-phosphate ( PI ( 3 ) P ) is generated by PI ( 3 ) P kinase and is required for endocytic trafficking [43] . Vero cells infected with either fusogenic or chemically inactivated YFV or JE-VLPs induced robust activation of PI ( 3 ) P kinase as indicated by phosphorylation of AKT , also known as protein kinase B ( Figure S4 ) . The PI ( 3 ) P kinase inhibitor wortmannin inhibits ECV formation in mammalian cells [44] . PI ( 3 ) P is abundant in ECVs and early endosomes , but not in the late endosome [45] . To determine whether PI ( 3 ) P kinase activity is required for membrane fusion , we pretreated Vero cells with 60 nM wortmannin and tracked fusion of R18-labeled JE-VLPs and YFV as described above . This concentration of wortmannin was not toxic but did cause vacuoles to form inside the cells as expected ( Figure 5A ) . The resulting R18 fluorescence profiles were similar to those with nocodazole or anti-BMP antibody pretreatment , with the same R18 dequenching kinetics but no fluorescence decay ( Figure 5B–C and Movie S3 ) . Since wortmannin inhibits ECV formation , we attribute the gradual increase in fluorescence in the presence of wortmannin to inefficient lipid mixing . RT-PCR analysis and plaque assay showed that wortmannin pretreatment blocked RNA release into the cytoplasm of Vero cells and virus infectivity in BHK cells , respectively ( Figure 5D–E ) . Collectively , these results suggest that in the absence of ECVs , JE-VLPs and YFV fuse with other as yet unidentified membranous structures or compartments , where the nucleocapsid remains trapped . Alternatively , the lipid composition or curvature of the limiting early endosomal membrane may preclude full membrane fusion of JE-VLPs . The observed R18 dequenching may then be attributed to transient hemifusion of the viral and endosomal membranes , which would allow lipid mixing of the proximal viral and endosomal lipid monolayers , and therefore dilution of R18 , without nucleocapsid delivery into the cytoplasm . In certain flaviviruses low pH does not appear to be sufficient to trigger fusion [21] . We note that YFV fusion is only partially inactivated by a pretreatment under conditions ( pH 6 . 2 ) but that infection is nevertheless completely inhibited in acidic media ( Figure S5 ) . To determine the minimal physicochemical requirements for membrane fusion of JE-VLPs and YFV , we measured fusion of R18-labeled virus particles in vitro in a bulk fusion assay with synthetic liposomes . The liposomes were 0 . 1 µm in diameter ( see Materials and Methods ) and their lipid composition was chosen to correspond to those in EEs/ECVs: cholesterol , phosphatidylcholine ( PC ) , phosphatidylethanolamine ( PE ) , PI ( 3 ) P , and phosphatidylserine ( PS ) at a molar ratio of 3∶4∶1∶1∶1 [7] , [9] . Fusion was measured by R18 dequenching . We found that both JE-VLPs and YFV fused with the liposomes at pH 5 . 5 , but not at neutral pH ( Figure 6A ) . In flaviviruses , a conserved cluster of histidine side chains acts as a “pH sensor” , which triggers the fusogenic conformational change in response to the reduced pH of the endosome [16] , [17] . The histidine modifying agent diethylpyrocarbonate ( DEPC ) inactivates VSV [46] and dengue virus [22] by inhibiting the fusogenic conformational change . We found that DEPC also blocked fusion of R18-labeled YFV with liposomes at pH 5 . 5 ( Figure 6B ) . In conclusion , synthetic liposomes with a lipid composition similar to ECVs are sufficient to induce flavivirus fusion in vitro at low pH . PS and PI ( 3 ) P are abundant in mammalian EEs/ECVs [7] . We found that JE-VLPs and YFV bound to PS-coated beads ( Figure 6C ) . Similarly , heparan sulfate beads also bind the virus particles , consistent with reports that flaviviruses bind heparan sulfate through the viral envelope protein [34] , [47] . However , the virus particles did not bind to PI ( 3 ) P beads ( Figure 6D ) , suggesting that the binding to PS is not due to nonspecific electrostatic interactions and that PS may act as an intracellular receptor or fusion cofactor for flaviviruses . PS and PI ( 3 ) P beads bound with equal affinity to polyarginine peptides , indicating that the surface charges of the two types of beads are comparable ( Figure S6 ) . Calcium released into the cytoplasm during viral infection can result in the translocation of PS from cytoplasmic to extracellular lipid leaflets [48] , [49] , [50] . Imaging of cells with the calcium-dependent dye Fluo-4 showed that calcium is released into the cytoplasm within one minute of infection with YFV or JE-VLPs ( Figure S7 ) . To determine whether intracellular calcium release promotes flavivirus infection we measured the effect of the cell-permeable calcium chelator BAPTA on YFV infectivity . BAPTA reduced YFV infectivity to less than 5% of the untreated infected control ( Figure S7C ) . We propose intracellular calcium release during flavivirus infection may cause a redistribution of PS towards extracellular or luminal leaflets , which may be important for flavivirus infectivity . Antibodies that inhibit fusion by targeting the fusion loop of the E protein are important determinants in the humoral response to flavivirus infection [51] , [52] , [53] . The therapeutic scFv11 antibody fragment , which recognizes the fusion loop , was selected by phage display for binding to West Nile virus E protein [54] , [55] . scFv11 protects mice from a lethal challenge of West Nile virus and also protects against dengue virus types 2 and 4 [54] . To probe the reactivity of scFv11 against other flaviviruses , we used an ELISA assay to measure the binding of scFv11 antibody to either JE-VLPs or YFVs . scFv11 bound tightly to JE-VLPs but did not bind to YFV ( Figure 7A ) . These results were confirmed by co-elution of scFv11-VLP complexes in size-exclusion chromatography ( Figure 7B ) , and by a plaque assay with YFV showing that scFv11 had no effect on YFV infectivity . The location of the scFv11 epitope in the fusion loop of E suggests that scFv11 inhibits viral membrane fusion [55] . To test whether scFv11 inhibits fusion of JE-VLPs , we used the in vivo and in vitro fusion assays described above . Preincubation of R18-labeled JE-VLPs with scFv11 inhibited membrane fusion in Vero cells , as judged by the lack of R18 dequenching . Subsequent addition of untreated JE-VLPs produced R18 dequenching as expected ( Figure 7C ) . In the bulk fusion assay , scFv11 reduced the acid-induced fusion of JE-VLPs with EE/ECV-like synthetic liposomes by 50% ( Figure 7D ) . These data suggest that the fusion loop of JE-VLPs is accessible to scFv11 , which inhibits fusion of JE-VLPs in EEs/ECVs . The dissociation equilibrium constant of scFv11 from JE-VLPs , measured by isothermal titration calorimetry , was at 150 nM ( Figure 7E ) . scFv11 was previously shown to bind soluble form of West Nile E with a 5 nM dissociation constant [54] . The higher affinity for West Nile E could be due to the fusion loop epitope being partially occluded in JE-VLPs , or to differences in the amino acid sequence or structure of the non-cognate JE E and the cognate West Nile E . The stoichiometry of binding was 0 . 648±0 . 013 scFv11 molecules per E protein . Thus , if the JE-VLPs each contain 60 E proteins , as is the case in an electron microscopy structure of tick-borne encephalitis VLPs [56] , each JE-VLP would be capable of binding approximately 40 scFv11 molecules . The observed substoichiometric binding of scFv11 suggests that one third of the E-protein epitopes in VLPs do not bind scFv11 either because they are not fully exposed or because they are clustered too closely together to allow full occupancy by scFv11 without steric clashes . The latter is more likely given the presumed T = 1 icosahedral symmetry of the VLPs , in which each E protein displays identical surface epitopes [56] . In contrast , in mature virions in the E proteins are distributed in three distinct chemical environments with slightly different surface epitopes . Two different neutralizing antibodies against dengue and West Nile viruses bind to only two thirds of the E proteins in their cognate virions [57] , [58] , providing precedents for the stoichiometry reported here for scFv11 binding to JE-VLPs .
Many enveloped viruses enter the endocytic pathway and rely on specific features of the endosomal environment , in particular the reduced pH and the lipid composition , to trigger membrane fusion and productive delivery of the viral genome into the cytoplasm . Although it has been established that flaviviruses generally undergo clathrin-mediated endocytosis [35] , [36] , [59] , the sequence and kinetics of the steps required for cell entry remain unclear . It has been reported that approximately 20% of dengue virions fuse early in the endocytic pathway while the rest fuse in late endosomes [60] , but it is unclear whether all of these fusion events lead to productive infection . We note that diphtheria toxin , a bacterial bipartite toxin complex composed of carrier and toxin subunits , inserts its carrier subunit into early and late endosomal membranes but the active toxin subunit is only delivered to the cytoplasm from early endosomes , and most of diphtheria toxin complexes are degraded in the lysosome [4] . Little is known about the physical and biological properties of flavivirus VLPs- how and when they assemble , and what their possible roles are in infection and in virus evolution . In this study , we have dissected the cell entry steps of VLPs and virions from two different flavivirus species . Tracking of single virus particles in live cells and virus infectivity measurements in the presence of various cell biological inhibitors are consistent with an entry mechanism in which virus particles fuse preferentially with small endosomal carrier vesicles ( ECVs ) , with nucleocapsid delivery into the cytoplasm occurring several minutes later , when the ECVs fuse with the limiting membrane of the late endosome . Alternatively , instead of fusing completely with ECVs , the virus particles may form metastable hemifusion intermediates with the ECVs , with full fusion only occurring in late endosomes , consistent with a previous report that dengue forms ‘restricted hemifusion’ intermediates [22] . Either way , we conclude that flavivirus membrane fusion and nucleocapsid delivery into the cytoplasm are distinct events in space and time ( Figure 8 ) . While novel for flaviviruses , a sequential cell entry mechanism involving delivery into ECVs followed by back-fusion of the ECVs to the limiting late endosomal membrane is not unprecedented . VSV utilizes this mechanism of nucleocapsid delivery into the cytoplasm to reach the cytoplasm [5] although VSV does not impact PI ( 3 ) P signaling or trigger intracellular calcium release [31] , [32] , [33] . Similarly , anthrax lethal toxin ( LT ) from Bacillus anthracis , a bipartite toxin complex composed of carrier and toxin subunits , inserts carrier protein ( protective antigen ) into ECVs in response to endosomal acidification , and delivers its toxin subunit ( lethal factor ) into the cytoplasm upon back-fusion of the ECV with the limiting late endosomal membrane to deliver lethal factor to the cytoplasm [6] . The back-fusion of ECVs with the endosome's limiting membrane depends on the anionic lipid BMP [5] , which is found in internal membranes and vesicles within late endosomes but not in the limiting endosomal membrane [8] . Treatment of cells with anti-BMP antibody did not inhibit membrane fusion of JE-VLPs or YFV but strongly inhibited both YFV infectivity and RNA delivery into the cytoplasm . This suggests that the cell entry mechanism of these viruses is dependent on back-fusion of ECVs to the limiting late endosomal membrane . Since anionic lipids are required for efficient fusion of dengue virus [22] , the presence of the anionic lipid phosphatidylserine ( PS ) in ECVs may be responsible for the preference of JE-VLPs and YFV to fuse with ECV membrane over limiting endosomal membranes . Additionally , the presence of cholesterol in the target membrane promotes fusion [61] , [62] and cholesterol chelation reduces flavivirus infectivity , although addition of exogenous cholesterol at the cellular attachment step has been reported to block JE and dengue virus cell entry [63] . Cholesterol is abundant in early endosomes and ECVs [8] . While anionic lipids are important in a general sense for cell entry of enveloped RNA viruses , PS may perform a more specific function . PS is abundant in early endosomes and ECVs , where PS represents about 9% of all phospholipids [7] , [9] . YFV and JE-VLPs fused with liposomes of this lipid composition . Moreover , PS-coated beads bounds JE-VLPs and YFV whereas beads coated with PI ( 3 ) P , which is also anionic and present in ECVs , did not bind the virus particles . Interestingly , PS is expressed on the plasma membranes of insect cells [64] and malignant and non-apoptotic cells [65] , [66] . The therapeutic antibody fragment scFv11 binds JE-VLPs with reasonably high affinity . Noninfectious flavivirus VLPs produced during infection [26] may thus serve as antibody decoys to promote immune evasion of the infectious virions . VLPs have recently been in focus as vaccine candidates [27] . Our study establishes that flavivirus VLPs can be used as a model for virus entry and for screening of therapeutic antibodies . In summary , our work suggests a novel mechanism for flavivirus cell entry in which the virus fuses to ECVs and depends on a second cell-mediated membrane fusion event to deliver the viral genome from the vesicle lumen to the cytoplasm . We propose that flavivirus infection modulates PI ( 3 ) P-dependent signaling in the host and modifies host phospholipid distribution to promote fusion with endocytic compartments . Our findings provide a framework for future studies to determine the physicochemical basis of the preference for membrane fusion with ECVs , the nature of the contribution of specific lipids ( BMP , PS , cholesterol ) to fusion activity , and the precise sequence and kinetics of the molecular steps required for membrane fusion and nucleocapsid delivery .
Horse anti-WNV E antibody was a generous gift from L2 Diagnostics ( New Haven ) . The scFv11 construct , a kind gift from Erol Fikrig , was expressed and purified as described [54] . The purified protein showed the expected purity and molecular weight in SDS-PAGE and size-exclusion chromatography ( Figure S8 ) . The following reagents were purchased from commercial sources: rabbit anti-human Rab5 , Rab7 , Phospho-AKT and Total-AKT antibodies ( Cell Signaling ) , anti-LPBA ( BMP ) antibody ( Echelon biosciences ) , Fluorescein ( FITC ) -labeled anti-horse antibody ( Bethyl Labs . ) , horseradish peroxidase ( HRP ) -conjugated anti-horse ( Sigma ) , Texas Red anti-rabbit antibody ( Invitrogen ) , total mouse IgG ( Sigma ) , Texas Red anti-mouse antibody ( Invitrogen ) . All inhibitors were freshly prepared before use according to the manufacturers' recommendations . 25 µM BAPTA ( Sigma ) was added to the virus , maintained during cellular attachment , and removed before addition of agarose plugs in plaque assays . Wortmannin ( Sigma ) was used at a final concentration of 0 . 1 µM while nocodazole ( Sigma ) was used at 20 µM final concentration . Chloroquine ( Sigma ) was used at a final concentration of 194 µM ( 0 . 1 g/l ) . All inhibitors were added to the cells before starting the experiments . In the plaque assay , all of the inhibitors were diluted to their half concentrations and kept in the medium after adding the agarose plugs as described in plaque assay method . Diethylpyrocarbonate ( DEPC ) ( Sigma ) was added at final concentration of 2 mM to purified YFVs for 30 min and then removed by buffer exchange using an ultrafiltration unit . The prM-E sequence of JEV strain CH2195LA was cloned into the pAcgp67 vector ( BD Biosciences ) . Sf9 cells were co-transfected with the linearized baculovirus genome and the pAcgp67-prM-E construct . The secreted recombinant baculovirus encoding the prM-E sequences was amplified in Sf9 cells . For protein expression , Tni cells ( Expression Systems ) were infected with the baculovirus in ESF921 medium ( Expression Systems ) at 27°C . Alternatively , JE-VLPs were expressed in HEK293T cells transiently transfected with the pcDNA-prM-E mammalian expression vector . Cell media was clarified by centrifugation at 4°C . We then added 2 . 5% ( w/v ) NaCl slowly with continuous stirring , followed by 10% ( w/v ) polyethylene glycol 6000 ( PEG6000 ) to precipitate the VLPs . VLPs were pelleted at 20 , 000 rpm for 20 min . The pellet was resuspended in 20 ml of buffer ( 50 mM Tris pH 8 . 4 , 0 . 1 mM EDTA , 0 . 15 M NaCl ) and centrifuged at 20 , 000 rpm for 2 min to remove excess PEG6000 . The VLPs were then purified on a 10–40% sucrose gradient by centrifugation for 9 h at 30 , 000 rpm in a SW32 rotor at 4°C . The gradient was then separated into 1-ml fractions and VLPs were detected by immunoblotting . Horse anti-WNV E antibody and anti-horse secondary antibody conjugated to HRP were both used at a dilution of 1∶104 . Fractions containing VLPs were concentrated in ultrafiltration units with a 30-kDa molecular weight cutoff ( Millipore ) . Buffer was exchanged for all positive layers to 50 mM Tris pH 7 . 4 , 0 . 14 M NaCl and the VLPs were checked for the purity by SDS-PAGE on 4–20% acrylamide gels with Coomassie staining . The hydrodynamic radius of the JE-VLPs was determined by dynamic light scattering analysis ( Malvern ) . Generation of the BHK cell line expressing the NS1-GFP fusion protein for trans-complementation of YFV 17D genome lacking NS1 has been described previously [37] . Non-infectious ΔNS1 viruses were collected from the cell culture medium . The YFV particles were purified by PEG6000 precipitation and sucrose gradient as described above for JE-VLPs . In the absence of an effective antibody to detect YFV structural proteins , YFV was quantified in fractions from the sucrose gradient as plaque-forming units in a plaque assay with the BHK-NS1-GFP cells ( see below ) . Serial dilutions of YFV were added to BHK cells ( 5×105 cells/ml ) in DMEM medium . Viruses were allowed to attach to the cells for 1 h at 37°C . A 1∶1 mixture of 2× DMEM medium ( Gibco ) and autoclaved 1 . 6% ( w/v ) agarose ( 37°C ) was then layered onto the cells in 6-well plates . After 2–3 days , the wells were fixed with 7% formalin ( Sigma ) and the agarose plugs were removed . Cells were stained with 0 . 5% crystal violet ( Sigma ) to visualize the plaques . Excess stain was removed with water . For acid pretreatment of YFV , the buffer was exchanged to 50 mM HEPES pH 6 . 2 with an ultrafiltration device ( Millipore ) , viruses were allowed to attach to cells for 15 min in DMEM pH 6 . 5 or 7 . 4 , and cells were then washed with DMEM pH 7 . 4 before proceeding to the next step of plaque assay . 5 µl of JE-VLP suspension was applied for 2 min on the carbon surface of a glow-discharged carbon-coated grid ( Microscopic Science ) . Excess sample was removed using absorbent paper and the grid was air-dried before examination . Data were collected using a Zeiss EM900 electron microscope . JE-VLPs were labeled with Rhodamine C18 ( R18; Invitrogen ) . The dye was added to the PEG-precipitated fraction of either YFV or JE-VLPs at a final concentration of 20 ng/µl and incubated for 15 min before the sucrose gradient centrifugation step . Vero cells were grown on a coverslip petri dish ( MatTek ) overnight at a density of 5×105 cells/ml at 37°C , 5% carbon dioxide . Before microscopic examination , the medium was changed to serum-free OptiMEM ( Gibco ) and cells were stained with Hoechst stain ( Invitrogen ) . JE-VLPs were added from a stock at 17 pM ( 50 ng/ml E protein ) and the particles were kept in the medium during data collection . Time-lapse confocal microscopy was performed using a Zeiss microscope connected to 37°C incubated and buffered with 5% CO2 . Time-lapse images were collected using a slice of 4 µm to avoid changes in confocal planes during data collection . Images were collected every 10 s . Data were analyzed with ImageJ [67] . Immunostaining was performed as described [68] . Briefly , Vero cells were grown on a coverslip overnight at 5×105 cells/ml and treated at different time points with 34 pM JE-VLPs ( 50 ng/ml E protein ) . 1 µl of Hoechst stain ( 10 g/l ) was added to the cells for 10 min at 37°C . Cells were fixed with 4% paraformaldehyde and permeabilized with 1% Triton ×100 . Cells were then blocked for 1 h with 10% fetal bovine serum ( FBS ) and stained with the primary anti-Rab5/7 antibody according to the manufacturer's recommendation ( Cell Signaling ) . The cells were washed 10 times with PBS . For BMP staining , cells were fed 50 µg/ml anti-BMP antibody overnight and the cells were washed and fixed as above . The cells were then permeabilized with 0 . 3% Tween-20 in PBS and blocked for 1 h in 10% FBS . Cells were blocked for 1 h with 10% FBS and then stained with secondary antibody conjugated to Texas Red ( Invitrogen ) . Cells were washed 10 times with PBS and mounted with fluoromount G ( Microscopic Science ) before examination . Semi-quantitative colocalization analysis ( Pearson coefficient calculation ) was performed with ImageJ . Cholesterol , phosphatidylethanolamine ( PE ) , PI ( 3 ) P , phosphatidylserine ( PS ) , and phosphatidylcholine ( PC ) were mixed in chloroform at a molar ratio of 3∶1∶1∶1∶4 , respectively , and then dried with argon gas and under vacuum for 2 h . The lipids were resuspended in 3 ml TEA buffer ( 10 mM triethanolamine pH 8 . 3 , 0 . 14 M NaCl ) and subjected to 10 freeze-thaw cycles using liquid nitrogen and a 37°C water bath . The lipid suspension was then extruded through 0 . 1 µm membranes 21 times with a lipid extruder ( Avanti Polar Lipids ) . The liposome suspension was added to R18-labeled virus particles . After a 5 min incubation , the pH was modified with either sodium acetate pH 5 . 5 buffer or with Tris pH 8 . 4 buffer ( 70 mM final concentration ) . R18 fluorescence was monitored after 1 min . with a QuantaMaster cuvette-based spectrofluorometer ( Photon Technology International ) or a time-domain plate-based fluorimeter ( HoriBa ) . ELISA plates were coated with 0 . 1 M carbonate buffer pH 9 . 6 and either JE-VLPs or YFV overnight at 4°C . 6×1010 JE-VLPs ( 300 ng E protein ) or 6×107 plaque forming units of YFV were added to each well . The coated wells were blocked with 10% FBS in PBS for 1 h at room temperature . Primary antibody was added and plates were incubated for 1 h at room temperature followed by 10 washes with PBS . Secondary antibody was added and the plate was incubated for 30 min at room temperature . Plates were then washed 10 times with PBS . HRP substrate TMB ( Sigma ) was added and stop solution ( Sigma ) was used to stop the color development . Absorbance was measured at 450 nm . Standard curves of JE-VLPs were generated by serial dilution of purified VLPs . The concentration of E protein was estimated by comparing Coomassie staining on SDS-PAGE with staining from known concentrations of bovine serum albumin ( Sigma ) . The E protein concentration was used to determine the concentration of JE-VLPs in this study , assuming 60 copies of E per VLP . 1 ml of media from BHK cells infected with a YFV multiplicity of infection ( MOI ) of 0 . 1 , or insect cells expressing prM-E were mixed with a 50 µl suspension of beads coated with heparan sulfate ( HS; Sigma ) , phosphatidylserine ( PS; Echelon biosciences ) , or phosophoinisitol-3-phosphate ( PI ( 3 ) P ) . The beads were pre-equilibrated with 50 mM Tris pH 7 . 4 and 0 . 1 M NaCl . Uncoated beads ( Echelon biosciences ) were used as a negative control . The beads were collected by centrifugation , washed with equilibration buffer and eluted with 50 mM Tris pH 7 . 4 and 0 . 5 M NaCl . Samples containing JE-VLPs were concentrated and the buffer was exchanged to 50 mM Tris pH 7 . 4 , 0 . 14 M NaCl using an ultrafiltration unit ( MilliPore ) . Samples were then analyzed by SDS-PAGE on 4–20% acrylamide gels and analyzed by Western blotting using the anti-WNV-E antibody . For samples containing YFV , RNA was extracted from the bead eluates with Trizol ( Invitrogen ) in the presence of 50 µg/ml yeast transfer RNA ( RNase and DNase free; Sigma ) as a viral RNA carrier . RNA was quantified by RT-PCR as described below . To control for the surface charges of the PI and PI ( 3 ) P beads , we tested binding of 2 . 5 µg/µl polyarginine ( 5–15 kDa , Sigma ) to each types of bead , using 100 µl beads . The binding , wash , and elution steps for polyarginine were performed as described above for JE-VLPs and YFV . Polyarginine in the eluate was quantified using Bradford reagent . We employed an established protocol to isolate endosomes from the cytosolic fraction [5] , [11] , [41] , [69] . Briefly , Vero cells ( 6×106 cells/ml ) in DMEM were infected with YFV ( MOI = 1 ) and incubated for 1 h . Cells were grown in 2 g/l HRP ( Sigma ) for 40 min post-infection . Cells were washed with PBS and harvested by centrifugation at 1500 rpm for 5 min at 4°C . Cells were suspended in homogenization buffer ( 3 mM imidazole and 8 . 5% sucrose pH 7 . 4 plus protease inhibitors ( Roche ) ) and passed through a steel syringe needle 20 times . The nuclear fraction was isolated by centrifugation for 10 min at 1 kg and 4°C . Sucrose was added to the post-nuclear supernatant ( PNS ) to a final concentration of 40% ( w/v ) . This mixture was overlaid with 35% , 27% , and 8 . 5% sucrose cushions in 10 mM HEPES pH 7 . 4 . Samples were centrifuged for 1 h at 100 kg in a SW60 Ti swinging-bucket rotor . Late endosomes were collected at the 27/8 . 5% sucrose interface while early endosomes were collected at the 35/40% interface . Total RNA was extracted with Trizol ( as described above ) from the load ( cytosolic ) fraction . The purity and integrity of the purified RNA was determined by OD260/280 and by 1% formaldehyde agarose gel electrophoresis . To confirm that the isolated endosomes were intact , infected cells were grown with HRP for the last 20 min of the infection and the HRP activity of the endosomal or cytosolic fractions was measured after lysis with 1% Tween-20 ( Figure S3A ) . Effective separation of the cytosolic and endosomal fractions was also confirmed by Western blotting with antibodies against Rab5 and Rab7 ( Figure S3C ) . The 3′ untranslated region downstream primer ( 3UTR , bases 10109–10128: 5′-AACCCACACATGCAGGACAA-3′ ) and glyceraldhyde-3-phosphatedehydrogenase downstream primer ( GAPDH , bases 1157–1175: 5′-TCCACCACCCTGTTGCTGT3′ ) were used to reverse-transcribe the viral RNA and cellular housekeeping gene GAPDH , respectively . Quantitative real-time PCR ( qRT-PCR ) was performed using the same downstream primers and the 3UTR upstream primer ( 10337–10318 bases , 5′-GTTGCAGGTCAGCATCCACA-3′ ) and GAPDH upstream primer ( bases 724–742: 5′-ACCACAGTCCATGCCATAC-3′ ) . PCR reactions were carried out in triplicate using an RT-PCR kit ( Roche ) and an ABI 9700HT RT-PCR instrument ( Applied Biosystems ) . The amplified products ( 228 bp for 3UTR and 450 bp for GAPDH ) were identified on 2% agarose gels . RT-PCR products were relatively quantitated with the software SDS . Endogenous GAPDH was used as a control for the quality of the total extracted RNAs . Neither 3UTR nor GAPDH formed primer dimers as judged by the dissociation curve . Binding of scFv11 to JE-VLPs was analyzed in 50 mM Tris pH 7 . 4 , 0 . 14 M NaCl , 2 mM β-mercaptoethanol at 25°C , with an iTC200 calorimeter ( MicroCal ) . The sample cell contained 3 . 1 µM JE-VLPs or buffer only , and the titrant syringe contained 40 µM scFv11 . An initial injection of 1 . 5 µl of scFv11 was followed by 20 serial injections of 2 . 0 µl scFv11 , each at 10 min intervals . The stirring speed was 1 , 000 rpm and the reference power was maintained at 11 µcal/s . The net heat absorption or release associated with each injection was calculated by subtracting the heat associated with the injection of scFv11 to buffer . Thermodynamic parameters were extracted from a curve fit to the data to a single-site model with Origin 7 . 0 ( MicroCal ) . Experiments were performed in triplicate . JE-VLPs , scFv11 and VLP-scFv11 complexes were separated on a Superdex 200 10/300 GL column ( GE Healthcare ) in 50 mM Tris pH 7 . 4 , 0 . 14 M NaCl . Vero cells were loaded with 5 µM of Fluo-4 ( Invitrogen ) for 15 min in DMEM medium . Cells were either infected with YFV ( MOI = 5 ) or treated with 200 µl JE-VLPs ( at 17 pM or 50 ng/ml E protein ) . As a positive control we used 10 mM ionomycin ( Invitrogen ) . As a negative control we pretreated Vero cells with 25 mM BAPTA for 30 min before loading the cells with Fluo-4 . Images were collected at one frame every 2 s . Fluo-4 fluorescence was analyzed with ImageJ . Supporting Information includes seven figures and three movies and can be found with this article at the Journal's website . | Many viruses package their genetic material into a lipid envelope . In order to deliver their genome into the host-cell cytoplasm , where it can be replicated , viruses must fuse their envelope with a cellular lipid membrane . This fusion event is therefore a critical step in the entry of an enveloped virus into the cell . In this study , we used various cell biological and biochemical approaches to map precisely the cell entry pathway of two major human pathogens from the flavivirus family , yellow fever virus and Japanese encephalitis virus . We discovered that these viruses co-opt cellular phospholipid signaling to promote the fusion of their envelope with the lipid envelope of small compartments inside the host-cell endosomes . The viral genome remains trapped in these compartments for several minutes until the compartments fuse with the surrounding endosomal membrane . It is this second membrane fusion event that delivers the viral genome into the cytoplasm . We also showed that the antibody fragment scFv11 inhibits the fusion of the viral envelope with small lipid compartments , explaining the therapeutic activity of the scFv11 antibody . Our work identifies new vulnerabilities in the entry pathway of flaviviruses , including the formation of small endosomal compartments and two distinct membrane fusion events involving these compartments . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"medicine",
"viral",
"transmission",
"and",
"infection",
"flavivirus",
"emerging",
"viral",
"diseases",
"immunology",
"microbiology",
"viral",
"structure",
"viruslike",
"particles",
"neglected",
"tropical",
"diseases",
"japanese",
"encephalitis",
"infectious",
"diseases",
... | 2013 | Viral Membrane Fusion and Nucleocapsid Delivery into the Cytoplasm are Distinct Events in Some Flaviviruses |
Mathematical methods of information theory appear to provide a useful language to describe how stimuli are encoded in activities of signaling effectors . Exploring the information-theoretic perspective , however , remains conceptually , experimentally and computationally challenging . Specifically , existing computational tools enable efficient analysis of relatively simple systems , usually with one input and output only . Moreover , their robust and readily applicable implementations are missing . Here , we propose a novel algorithm , SLEMI—statistical learning based estimation of mutual information , to analyze signaling systems with high-dimensional outputs and a large number of input values . Our approach is efficient in terms of computational time as well as sample size needed for accurate estimation . Analysis of the NF-κB single—cell signaling responses to TNF-α reveals that NF-κB signaling dynamics improves discrimination of high concentrations of TNF-α with a relatively modest impact on discrimination of low concentrations . Provided R-package allows the approach to be used by computational biologists with only elementary knowledge of information theory .
Biochemical descriptions of cellular signaling appear to require quantitative support to explain how complex stimuli ( inputs ) are translated and encoded in activities of pathway’s effectors ( outputs ) [1–5] . An attractive approach seems to be offered by probabilistic modeling and information theory [6–10] , which provide a mathematical language to describe input-output relationships of complex and stochastic cellular processes . Specifically , the unique perspective of information theory holds a promise of gaining new insights into functional aspects of signaling , as opposed to biochemical and mechanistic descriptions [1 , 9 , 11–16] . So far , quantification of an overall signaling fidelity and analysis of factors by which it is determined have proven to be useful applications of information theory in studies of cellular signaling [5 , 17–19] . Nevertheless , exploring the information-theoretic approach remains conceptually and technically challenging [7 , 9 , 10] . In particular , for systems with multiple inputs and outputs existing theoretical tools are computationally inefficient and require a large sample size for accurate analysis . Within information theory , regardless of specific details of a signaling pathway , a signaling system can be considered as an input-output device that measures an input signal , x , by eliciting a stochastic output , Y [7 , 9 , 10] . In a typical example , the input , x , is the concentration of a ligand , e . g . , cytokine , that activates a receptor . The output , Y , is an activity of one or more signaling effectors , e . g . , of transcription factors quantified over time . As cellular signaling systems are inherently stochastic , the information about the input contained in the output is imprecise and only a limited number of input values can be resolved [6 , 15 , 20 , 21] . To date , a number of studies have experimentally examined fidelity of various signaling systems , e . g . , [13 , 15 , 19 , 22–24] . In a typical experiment aimed to quantify fidelity , input values , x1 ≤ x2… ≤ xm , ranging from 0 to saturation are considered . In some scenarios , utilization of physiologically arising input concentrations , e . g . , morphogen gradients , is also possible [5 , 16 , 25] . Then , for each input level , xi , cellular responses are quantified in a large number , say ni , of individual cells . Single-cell responses are typically represented as vectors , y l i , that contain entries with quantified activities of signaling effectors , where l varies from 1 to the number of measured cells , ni . Further , cell-to-cell heterogeneity of responses is simplistically used as a proxy of how reproducible signaling output of an individual cell is . Formally , responses corresponding to each of the inputs , xi , are assumed to follow a probability distribution y l i ∼ P ( Y | X = x i ) , ( 1 ) which is reconstructed from the data and serves as a model of the single-cell response to the input xi . Given the above , information theory offers to quantify the overall fidelity of signaling in terms of how many input values , xi , can be resolved based on information contained in the responses . The key factor that determines how many inputs can be resolved is the degree of the overlap between the distributions corresponding to different inputs . For instance , two inputs x1 and x2 can be easily resolved if the corresponding output distributions P ( Y|X = x1 ) and P ( Y|X = x2 ) are completely distinct , non-overlapping . Then , a given response , y , can be assigned without error to the only input for which it can occur . In the scenarios in which P ( Y|X = x1 ) and P ( Y|X = x2 ) are overlapping the inputs x1 and x2 cannot be resolved as a given response y is equally likely to arise for both inputs . The second factor that is essential for formal quantification of the overall fidelity is how frequently the different inputs occur . For illustration , consider a system with three input values , x1 , x2 and x3 . Suppose , x1 and x2 induce very similar output distributions , i . e . , P ( Y|X = x1 ) ≈ P ( Y|X = x2 ) , whereas the input x3 induces distribution P ( Y|X = x3 ) that is distinct , non-overlapping , with the first two . How many inputs are on average resolved in this system depends on how frequently different inputs occur . If for instance , x1 and x3 occur frequently , e . g . , P ( x1 ) = P ( x3 ) ≈ 1/2 , whereas , x2 has a negligible incidence , i . e . , P ( x2 ) ≈ 0 , the fidelity of signaling allows to resolve two inputs on average . This is because the two inputs that can be resolved occur with high probability . On the other hand , if inputs x1 and x2 occur with probability ≈ 1/2 , and x3 has probability close to 0 , then only one state can be resolved on average , i . e . , concatenation of x1 and x2 . The frequencies of inputs are taken into account in the form of the input distribution P ( X ) = ( P ( x1 ) , … , P ( xm ) ) . The degree of the overlap between the output distributions as well as frequencies of inputs are used to evaluate the overall signaling fidelity in the form of the mutual information M I ( X , Y ) = ∑ i = 1 m P ( x i ) ∫ R d P ( y | X = x i ) log 2 P ( y | X = x i ) P ( y ) d y , ( 2 ) where P ( y ) is the overall distribution of the output implied by a given distribution of the input , i . e . , P ( y ) = ∑ i = 1 m P ( y | X = x i ) P ( x i ) . MI is expressed in bits and 2MIcan be interpreted as the number of inputs that the system can resolve on average . Definition of mutual information stems from axioms proposed by C . Shannon [26] and can also be written in a more intuitive form of entropy differences , see supporting S1 Text . Selection of the input distribution that is suitable for quantification of information transfer in a specific application can be problematic and provides a degree of arbitrariness . The uniform distribution , i . e . , one that gives the same weight to all inputs appear to be a suitable choice in some applications [15] . Alternatively , a most favorable input distribution , P ( X ) , i . e . , the one that maximizes information transfer , can be found . The maximization of mutual information with respect to the input distribution defines the information capacity , C* . Formally , C * = max P ( X ) M I ( X , Y ) . ( 3 ) Information capacity is expressed in bits and 2 C * can be interpreted as the maximal number of inputs that the system can effectively resolve . In summary , the overall fidelity of a signaling system depends on the degree of the overlap between distributions corresponding to different inputs . The degree of the overlap is translated into the logarithm of the number of resolvable inputs by mutual information , which takes into account how frequently different inputs are transmitted . On the other hand , the information capacity quantifies the logarithm of the number of resolvable inputs under input frequencies that maximize the information transfer . Information capacity , as opposed to mutual information , does not depend on input distribution and therefore may provide a less arbitrary quantification of the overall fidelity . Existing tools to compute the mutual information and the information capacity [16 , 19 , 22 , 27–29] utilise the data , y l i , to construct approximations , P ^ ( Y | X = x i ) , of the output distributions , P ( Y|X = xi ) , for each input , xi . Thereafter , the approximations , P ^ ( Y | X = x i ) , rather than the exact probabilities , P ( Y|X = xi ) , are used for evaluation of the mutual information and information capacity , according to Eqs 2 and 3 . The available algorithms differ in the way , in which , the approximations of the output distributions , P ^ ( y | X = x i ) , are constructed . Specifically , Blahut—Arimoto ( BA ) algorithm [22 , 27–29] uses a discrete approximation . All possible values of responses are divided into a finite set of intervals and frequencies of responses falling into the same interval as y l i are used as the approximation of P ( y l i | X = x i ) . On the other hand , methods based on the small noise approximation assume Gaussian output with a limited variance [16 , 25 , 30 , 31] . Finally , the approach of [19] , following the earlier work [32] , uses the k-nearest neighbors ( KNN ) method , in which continuous approximations of P ( y l i | X = x i ) are constructed based on the distance of y l i to the k-th most similar response . Each of the above approaches is practically limited by the dimensionality of the output , Y . The BA algorithm can be essentially applied to systems with the one-dimensional output only . On the other hand , for multidimensional outputs , an accurate estimation of P ( Y|X = xi ) using KNN requires a relatively large sample size [33] . Moreover , KNN demands arbitrary specification of the parameter k , which for insufficient data size does not generally guarantee unbiased estimation [32–35] , and yields estimation sensitive to algorithm’s settings , i . e . , is not parameter-free . Also , KNN based approaches , when used to compute capacity , often require solving computationally expensive optimization problems . In summary , for multidimensional outputs , the practical difficulty in calculating mutual information , Eq 2 , results largely from the lack of methods for accurate estimation of multivariate probability distributions , P ( y|X = xi ) . In addition , calculation of C* , Eq 3 , can be problematic , as it requires maximization of the nonlinear function , i . e . , MI , over the input probability distribution , which can be computationally intense . In Section 1 in S1 Text , we provide more background on information theory and existing computational tools . Here further , we propose an alternative framework , statistical learning estimation of mutual information ( SLEMI ) that appears to significantly simplify the calculation of the mutual information and the information capacity , especially for systems with high-dimensional outputs and a large number of input values . Moreover , our framework enables simple quantification of the extent to which different inputs can be discriminated .
In contrast to existing approaches , instead of estimating , possibly highly dimensional , conditional output distributions P ( Y|X = xi ) , we propose to estimate the discrete , conditional input distribution , P ( xi|Y = y ) , which is known to be a simpler problem [36 , 37] . Estimation of the MI using estimates of P ( xi|Y = y ) , denoted here as P ^ ( x i | Y = y ) , is possible as the MI , Eq 2 , can be alternatively written as [38] M I ( X , Y ) = ∑ i = 1 m P ( x i ) ∫ R d P ( y | X = x i ) log 2 P ( x i | Y = y ) P ( x i ) d y . ( 4 ) Although P ( Y|X = xi ) is still present in the above sum , it represents averaging of the term log 2 P ( x i | Y = y ) P ( x i ) with respect to P ( Y|X = xi ) . The experimental data , however , constitutes a sample from the distribution P ( Y|X = xi ) . The average with respect to distribution P ( Y|X = xi ) can be , therefore , approximated by the average with respect to data , which is justified by the law of large numbers . Precisely , for a given P ( X ) and P ^ ( x i | Y = y ) , MI can be approximated with the following formula M I ( X , Y ) ≈ ∑ i = 1 m P ( x i ) ∑ l = 1 n i 1 n i log 2 P ^ ( x i | Y = y l i ) P ( x i ) . ( 5 ) An estimator P ^ ( x i | Y = y ) , can be built using a variety of Bayesian statistical learning methods . For simplicity and efficiency , here we propose to use logistic regression , which is known to work well in a range of applications [39–43] . In principle , however , other classifiers could also be considered . The logistic regression estimators of P ( xi|Y = y ) arise from a simplifying assumption that log-ratio of probabilities , P ( xi|Y = y ) and P ( xm|Y = y ) is linear . Precisely , log ( P ( x i | Y = y ) P ( x m | Y = y ) ) ≈ α i + β i T y . The above formulation allows fitting the logistic regression equations to experimental data , i . e . , finding values of the parameters , αi and βi that best represent the data . Once logistic regression parameters are known , estimates P ^ ( x i | Y = y ) can be constructed . Estimates , P ^ ( x i | Y = y ) , allow , in turn , to calculate mutual information using Eq 5 . The use of logistic regression , therefore , constitutes a simple way to evaluate mutual information for multidimensional data without knowledge of P ( Y|X = xi ) . Moreover , fitting the logistic regression equations to experimental data is efficient and available in most statistical software packages . Even though the assumed linear relationship may seem to be an oversimplification , the logistic regression approach has been shown to work exceptionally well in a variety of scenarios and gained broad applicability [36] . Methods contain more details on the form and estimation of the logistic regression model . In addition to the possibility of effective evaluation of the mutual information for models with the multivariate output , Y , the use of the logistic regression enables to overcome the potentially problematic numerical , typically gradient , maximization of the mutual information with respect to the input distribution , P ( X ) , in computations of the information capacity . Precisely , the numerical optimization can be bypassed , by dividing the maximization with respect to the input distribution , P ( X ) , into two simpler maximization problems , for which explicit solutions exist . Thereafter , a solution of the joint maximization can be obtained from the two explicit solutions in an iterative procedure known as alternate maximization . Compared to gradient optimisation , alternate maximization is superior in terms of computational efficiency , and much less prone to numerical complications . A complete description of the maximization procedure is technically demanding and therefore is provided in Methods section . In particular , the algorithm’s pseudo-code is presented in Box 1 therein . In summary , the use of logistic regression described above allows computing MI without estimation of the possibly highly dimensional output distributions P ( Y|X = xi ) . Moreover , it allows for efficient maximization of MI without gradient-based methods . In Methods and Section 3 in S1 Text , we perform several numerical tests to show how the above design of the algorithm leads to practical benefits in terms of the accuracy of estimation and computational efficiency . Specifically , we demonstrate that SLEMI: ( i ) provides more accurate estimates than the KNN method , especially for small sample size and highly dimensional output; ( ii ) delivers robust estimates , insensitive to algorithm’s settings; and ( iii ) has desired properties in terms of computational cost , especially scales well with the number of input values . The information capacity , C* , tells us how many inputs can be discriminated on average . What it does not tell us directly is which inputs , and to what extent , can be discriminated . Specifically , the same information capacity can result from different patterns of discriminability between input signals . For illustration , consider three inputs , x1 , x2 and x3 . Assume that inputs x1 and x2 induce very similar output distributions , i . e . , P ( Y|X = x1 ) ≈ P ( Y|X = x2 ) , as opposed to x3 that induces a distinct distribution , P ( Y|X = x3 ) . In such a scenario , the information capacity is approximately 1 bit , as two inputs can be discriminated , i . e . , x3 can be resolved from either of the other two . The capacity of 1 bit would also result from a scenario , in which roles of input values are swapped , say , P ( Y|X = x2 ) is distinct from the overlapping P ( Y|X = x1 ) and P ( Y|X = x3 ) . Therefore , it appears that an insight regarding which inputs , and to what extent , can be discriminated can usefully augment computation of the information capacity . Here , we argue that the extent to which different inputs can be discriminated can be conveniently quantified and visualized using the probability of correct discrimination ( PCD ) between input pairs . We define the PCD between a pair of input values , xi and xj , as the fraction of cellular responses that can be assigned correctly to one of the two inputs based on the information contained in the signaling response , Y . If the distributions P ( Y|X = xi ) and P ( Y|X = xj ) are entirely distinct , knowing the response , y , allows assigning each cellular response to the correct input without error . PCD between xi and xj is then equal to 1 . If , on the other hand , these two distributions are completely overlapping knowing the response , y , does not provide any information to assign a cell with a given response to the correct input . In such a case , the discrimination is close to random , yielding half of the cells being assigned correctly , i . e . , PCD equals 0 . 5 . Depending on the degree of the overlap , the PCD varies between 0 . 5 and 1 . Further , calculation of PCDs for all input pairs can provide insight regarding which inputs , and to what extent , can be discriminated . The above intuitions can be mathematically formalized , Fig 1 . For formal quantification , in order to treat both inputs equally , we assume that both have the same frequency , P ( X ) = ( 1/2 , 1/2 ) , or equivalently that half of the considered cells is stimulated with either of the input values , Fig 1A . How many cells can be assigned correctly depends on the overlap between the distributions P ( Y|X = xi ) and P ( Y|X = xj ) , Fig 1B . The conditional input distribution , P ( xi|Y = y ) expresses the probability that a given response y was generated by the stimulation level xi , Fig 1C . Equivalently , P ( xj|Y = y ) is the frequency , at which the given response , y , is generated by the stimulation level xj . The probabilities P ( xi|Y = y ) and P ( xj|Y = y ) tell us , therefore , how often assignment of the observation y to the input xi and xj , respectively , is correct . To maximize the probability of correct assignment , the response y should be assigned to the input for which it is most likely , i . e . to xi if P ( xi|Y = y ) ≥ P ( xj|Y = y ) , or to xj , otherwise . Therefore , the observation y can be assigned correctly with the probability equal to the maximum of P ( xi|Y = y ) and P ( xj|Y = y ) . Precisely , the probability of correct discrimination between input xi and xj for the response y , denoted as P C D x i , x j ( y ) , is calculated as PCD x i , x j ( y ) = max { P ( x i | Y = y ) , P ( x j | Y = y ) } , ( 6 ) which is visualised in Fig 1D . The average of the above probabilities over cellular responses , y l i , corresponding to the input xi is equal to 1 n i ∑ l = 1 n i P C D x i , x j ( y l i ) and quantifies the average probability of correct discrimination of responses induced by the input xi . Then , the overall probability of correct discrimination between xi and xj is given as PCD x i , x j = 1 2 1 n i ∑ l = 1 n i P C D x i , x j ( y l i ) + 1 2 1 n j ∑ l = 1 n j P C D x i , x j ( y l j ) . ( 7 ) In summary , the probability of correct discrimination between inputs xi and xj , PCD x i , x j , quantifies the fraction of cellular responses that can be correctly assigned to either of the inputs based on the output , Y . Therefore , the calculation of PCDs for all pairs of input values reveals the extent to which different inputs are discriminated . From the computational perspective , PCDs are defined in terms of the conditional input probabilities , P ( xi|Y = y ) , Therefore , similarly to the mutual information , these can be calculated using logistic regression . In Methods we provide practical details on how to compute PCDs . In the analysis of the NF-κB signaling data , presented below , we show how quantification of PCDs along with computation of the information capacity C* helps to provide insight regarding how signaling dynamics increases overall signaling fidelity . NF-κB pathway is one of the key biochemical circuits involved in the control of the immune system [44–46] . It is also one of the first cellular signaling systems studied within the framework of information theory [22] . So far , several papers examined its dose dependency , e . g . , [12 , 44 , 47] and quantified its information capacity , e . g . , [13 , 19 , 22] . Interestingly , response dynamics have been shown to have greater signaling capacity compared to time-point , non-dynamic , responses [13 , 19] . To demonstrate what benefits result from efficient calculation of the information capacity and of the probabilities of correct discrimination , we have measured NF-κB responses ( y l i’s in our notation ) to a range of 5 minutes pulses of TNF-α concentrations ( xi’s ) , in single—cells , using life confocal imaging . Experimental methods are described in Sections 4 . 1—4 . 3 in S1 Text . Fig 2A shows temporally resolved responses to representative four concentrations , whereas Fig . IV , in S1 Text , to all ten considered concentrations . Further , we used the data to calculate the information capacity between TNF-α concentration and cellular response for two different scenarios: time-point and time-series . Precisely , for the time-point scenario , we considered single-cell measurements at each time-point separately . In this case , the signaling output of an individual cell at a given time , y , is represented by a single number , which is different for different time-points . On the other hand , for the time-series scenario , we considered single-cell measurements from the beginning of the experiment until an indicated time . In this case signaling output of an individual cell , y , is a vector corresponding to a time window from 0 till a given time . Fig 2B and 2C show information capacity as the function of time for the time-point and time-series scenario , respectively . For the time-point scenario , the capacity increases at early times and reaches the maximum of ≈ 1 bit at ≈ 20 minutes , which coincides with the time of maximum response of trajectories shown in Fig 2A . Interestingly , the second peak of information transfer , of ≈ 0 . 3 bits , appears at ≈ 90 minutes . Inspection of the response trajectories , Fig 2A , around 90 minutes allows for an interpretation of the second peak . Comparison of the response trajectories corresponding to 1 and 100 ng/ml of TNF- α , Fig 2A , indicates the emergence of a fraction of cells that exhibit a second peak in response to the highest considered concentration . The second peak is reminiscent of the oscillatory behavior that is typical for the NF-κB pathway when exposed to continuous , as opposed to 5 minutes , stimulation [44 , 48–50] . The second peak in response trajectories carries some information about TNF-α and , therefore , contributes to the second peak of information transfer . For the time-series scenario the capacity also rapidly increases at early times to reach 1 bit at 20 minutes . For later times , the capacity continues to increase but at a much slower rate , with a modest acceleration around 70 minutes , to reach ≈ 1 . 3 bits at the end of the experiment . As the information contained in shorter time-series is contained in longer time-series , the capacity does not decrease . An increase in the capacity in a given time interval indicates that new information is arriving . Analogously , a time interval with a plateau demonstrates the lack of new information being transmitted at times of that interval . Therefore , our analysis demonstrates that most of the information , ≈ 1 bit , is transferred relatively early , i . e . , within the first 20 minutes after stimulation . Later times provide ≈ 0 . 3 bits of new information . In Section 4 . 5 in S1 Text we use the method proposed in [51] to further examine the redundancy of information contained in responses at individual time-points . The higher information content of the time-series poses the question: what type of information is contained in the time-series responses that is not encapsulated in the time-point responses ? The information capacity per se , being an overall measure of signaling fidelity , does not tell us to what extent specific inputs can be discriminated . Therefore , the information capacity alone cannot reveal which inputs gain discriminability due to signaling dynamics . In order to address the above question in detail , we have calculated the probabilities of correct discrimination , PCDs , for each pair of concentrations , xi , xj . Analogously as in the computation of the capacity , we have considered the time-point and time-series scenarios . PCDs for the two scenarios are shown in Fig 2D and 2E . The filled fractions of the pies mark the PCDs between the corresponding pairs of concentrations , i . e . , the fractions of cells that can be assigned to the correct input . The plots , primarily reveal how the time-point and time-series capacities of 1 and 1 . 3 bits , respectively , translate into discrimination between pairs of inputs . They show that concentrations far apart can be easily discriminated in both scenarios . For instance , PCDs between 0 and 100 ng/ml are close to 1 . The discrimination of closer concentrations is more difficult . For instance , in both scenarios , PCDs between 0 and 0 . 01 ng/ml are approximately 0 . 75 , which corresponds to 0 . 25 probability of incorrect assignment . Interestingly , for time-point responses , PCDs between pairs of concentrations ≥ 1 ng/ml are close to 0 . 5 , which implies lack of discriminability . This is , however , not the case for time-series responses where PCDs are considerably greater than 0 . 5 . Comparison of PCDs between the two scenarios reveals , therefore , which states gain discriminability due to signaling dynamics . Fig 2F presents differences between PCDs for time-series and time-points . The increase in discriminability resulting from signaling dynamics is particularly striking for high concentrations , say > 1 ng/ml . These concentrations are only weakly discriminated for the time-point scenario . Certain low concentrations also gain discriminability , e . g . , 0 . 01 and 0 . 03 . However , overall , the increase in discriminability is not so significant for low concentrations as these are relatively well discriminated in the time-point scenario . The above analysis yields similar conclusions to these presented in [19] . Here , we used 5 minutes TNF-α as opposed to continuous lipopolysacharide stimulation in [19] . Also , we used a more complete , higher dimensional , response trajectories , which allowed to plot the temporal profile of information transfer . Our analysis of PCDs complemented calculation of the capacity by revealing discriminability between different inputs . Indeed , values far apart are discriminated virtually without error . Closer concentration gain discriminability due to dynamics of signaling , whereas the gain is particularly strong for high concentrations . Our algorithm is available as robustly implemented R-package , SLEMI . It is designed to be used by computational biologists with a limited background in information theory . It includes functions to calculate mutual information , information capacity and probabilities of correct discrimination . These functions take as the argument a data frame , data , containing signaling responses stored in the following form input output 1 output 2 output 3 … n 1 { x 1 ⋮ x 1 n 2 { x 2 ⋮ x 2 ⋮ n m { x m ⋮ x m y 1 , 1 1 ⋮ y n 1 , 1 1 y 1 , 1 2 ⋮ y n 2 , 1 2 ⋮ y 1 , 1 m ⋮ y n m , 1 m y 1 , 2 1 ⋮ y n 1 , 2 1 y 1 , 2 2 ⋮ y n 2 , 2 2 ⋮ y 1 , 2 m ⋮ y n m , 2 m y 1 , 3 1 ⋮ y n 1 , 3 1 y 1 , 3 2 ⋮ y n 2 , 3 2 ⋮ y 1 , 3 m ⋮ y n m , 3 m ⋯ . ( 8 ) Each row , l , represents a single cell . The first column contains stimulation levels , xi . Further columns contain entries corresponding to measurements of cellular output , y l , d i , e . g . , subsequent elements of a time-series . Upon download with the function install_github ( ) of the ‘devtools’ package install_github ( “sysbiosig/SLEMI” ) the considered information-theoretic measures can be calculated by running mi_logreg_main ( data ) for MI , with uniformly distributed inputs; capacity_logreg_main ( data ) for information capacity C*; and prob_discr_pairwise ( data ) for probabilities of correct discrimination . More details on installation and applicability are provided in the package’s User Manual . A step-by-step Testing Procedures file is also available to assist with running essential functions . The package includes the NF-κB dataset as well as scripts to reproduce Fig 2 . Computations needed to plot each panel of the figure , without bootstrap , do not exceed several minutes on a regular laptop .
Building upon existing approaches , our framework considerably simplifies information-theoretic analysis of multivariate single-cell signaling data . It benefits from a novel algorithm , which is based on the estimation of the discrete input distribution as opposed to the estimation of continuous output distributions . Conveniently , the algorithm does not involve numerical gradient optimization . These factors result not only in short computational times but , also , in relatively low sample sizes needed to obtain accurate estimates . Therefore , our framework is particularly suitable to study systems with high dimensional outputs and a large number of input values . Also , the approach relates the information capacity to the probability of discrimination between different input values . Information theory seems to offer useful tools to provide a better understanding of how cells transmit information about identity and quantity of stimuli , and further how signaling systems enable cells to perform complex functions . Such tools might have a fair potential to successfully augment more traditional approaches . The latter provided a relatively good understooding of the overall molecular and biochemical mechanisms how individual cells transmit signals to effectors [1] . However , we lack understanding of how the stimuli are translated into distinct responses and , hence , how to effectively control cellular processes and decisions [1 , 3 , 15] . Specifically , the induction of distinct responses in individual cells by means of biochemical interventions in non-trivial settings is most often problematic [52–54] . Results of our work appear to contribute a relevant tool to apply information-theoretic analysis to more complex , particularly highly multivariate , data sets on signaling systems than achievable with available approaches . The multivariate aspect seems to be particularly relevant given the complexity of cellular processes . Hopefully , current and future development of multivariate single-cell measurement techniques , accompanied by computational tools , will enable utilization of the information-theoretic perspective in more complex scenarios . These in turn appear to have a potential to provide a comprehensive insight into how cells can derive a variety of distinct outputs from complex inputs using noisy , cross-wired and dynamic signaling pathways .
The logistic regression model is the state-of-the-art statistical method to estimate the probability of a given observation , i . e . , data vector , y , belonging to one of the m considered classes . In the setting of the paper , classes correspond to input values and observations to signaling responses . The method assumes that the overall frequencies of observations belonging to each class are described by a probability distribution . In the paper’s setting , these probabilities correspond to the input distribution P ( X ) = ( P ( x1 ) , … , P ( xm ) ) . The method is based on the assumption that for a given P ( X ) , the ratio of the probability of the observation y belonging to class i to the same probability for the class m is linear with respect to y . Precisely , denoting the logistic regression estimate of the probability of a given observation , y , belonging to the class i as P ^ l r ( x i | Y = y ; P ( X ) ) , the above assumption writes as follows log ( P ^ l r ( x 1 | Y = y ; P ( X ) ) P ^ l r ( x m | Y = y ; P ( X ) ) ) ≈ α 1 + β 1 T y , ⋮ log ( P ^ l r ( x i | Y = y ; P ( X ) ) P ^ l r ( x m | Y = y ; P ( X ) ) ) ≈ α i + β i T y , ⋮ log ( P ^ l r ( x m − 1 | Y = y ; P ( X ) ) P ^ l r ( x m | Y = y ; P ( X ) ) ) ≈ α m − 1 + β m − 1 T y ( 9 ) and ∑ i = 1 m P ^ l r ( x i | Y = y ; P ( X ) ) = 1 . Given the linear form of the above equations , for a data set given as Eq 8 , estimation of the parameters αi and βi can be done efficiently with state-of-the-art methods [36 , 55] . Further , the above equations imply that estimates , P ^ l r ( x i | Y = y ; P ( X ) ) , can be explicitly written as P ^ l r ( x 1 | Y = y ; P ( X ) ) = exp ( α 1 + β 1 T y ) 1 + ∑ r = 1 m − 1 exp ( α r + β r T y ) , ⋮ P ^ l r ( x i | Y = y ; P ( X ) ) = exp ( α i + β i T y ) 1 + ∑ r = 1 m − 1 exp ( α r + β r T y ) , ⋮ P ^ l r ( x m − 1 | Y = y ; P ( X ) ) = exp ( α m − 1 + β m − 1 T y ) 1 + ∑ r = 1 m − 1 exp ( α r + β r T y ) , P ^ l r ( x m | Y = y ; P ( X ) ) = 1 1 + ∑ r = 1 m − 1 exp ( α r + β r T y ) . ( 10 ) Below , we describe maximization of the mutual estimation , MI , with respect to the input distribution P ( X ) , using the so-called alternate optimization , which bypasses gradient optimization . The proposed algorithm is largely based on the original Blahut-Arimoto approach [27 , 28] . It is adapted to work with logistic regression and , hence , with continuous and multidimensional output , Y . The provided Lemmas are minor modifications of the original Blahut-Arimoto results to account for continuous and multidimensional output . The algorithm is based on the following five key components . One , the maximization of mutual information with respect to input distribution , P ( X ) , is replaced with a double maximization , i . e . , maximization with respect to the input distribution , P ( X ) , and with respect to tailored auxiliary function , Q ( X|Y ) . The function Q ( X|Y ) is introduced to dissect the effects of the input distribution , P ( X ) , and the conditional input distribution , P ( Y|X ) , on the mutual information , MI . Two , explicit solutions of the individual maximizations are found . Three , the individual maximizations are combined in an iterative procedure to provide the solution of the joint maximization . Four , integrals involved in the evaluation of the optimal solutions of individual maximizations are computed through averaging with respect to data . Five , logistic regression is used to evaluate optimal Q ( X|Y ) at each step of the iterative procedure . Each of the above five elements is described in detail below , and the complete algorithm is summarized in Box 1 . Probabilities of correct discrimination , PCDs defined in Eqs 6 and 7 , are expressed in terms of probabilities P ( xi|Y = y ) . Logistic regression model , Eq 10 , on the other hand , provides estimates of these probabilities . Therefore , logistic regression estimates , P ^ l r ( x i | Y = y ; P ( X ) ) , can be used to estimate probabilities of correct discrimination . In order to estimate PCDs , for a given pair of input values xi and xj , the logistic regression model needs to be fitted using response data corresponding to the two considered inputs , i . e . y l r , for r ∈ {i , j} and l ranging from 1 to nr . To ensure that both inputs have equal contribution to the calculated discriminability , equal probabilities should be assigned , P ( X ) = ( P ( xi ) , P ( xj ) ) = ( 1/2 , 1/2 ) . Once the regression model is fitted , probability of assigning a given cellular response , y , to the correct input value is estimated as max { P ^ l r ( x i | Y = y ; P ( X ) ) , P ^ l r ( x j | Y = y ; P ( X ) ) } . Note that P ( xj|Y = y ) = 1 − P ( xi|Y = y ) as well as P ^ l r ( x j | Y = y ; P ( X ) ) = 1 − P ^ l r ( x i | Y = y ; P ( X ) ) . Averaging the above over all responses corresponding to input values xi and xj , i . e . , with respect to the distribution P ( y ) = 1 2 P ( y | X = x i ) + 1 2 P ( y | X = x j ) , yields PCD ( xi , xj ) P C D ( x i , x j ) ≈ 1 2 1 n i ∑ l = 1 n i max { P ^ l r ( x i | Y = y l i ; P ( X ) ) , P ^ l r ( x j | Y = y l i ; P ( X ) ) } + 1 2 1 n j ∑ l = 1 n j max { P ^ l r ( x i | Y = y l j ; P ( X ) ) , P ^ l r ( x j | Y = y l j ; P ( X ) ) } . ( 27 ) To account for the possibility of overfitting of the regression model , the bootstrap procedure needs to be used [36] . For instance , available data should be randomly divided into a training data set , i . e . , data set used to fit logistic regression , and test data set , i . e . , the set to evaluate Eq 27 . The average of the PCDs from the bootstrap procedure should be used as a final estimate of the probability of correct discrimination . In order to validate the accuracy of the proposed information capacity estimators , examine the computational performance of the algorithm , and highlight advantages of the introduced approach , we have designed four test scenarios and carried out a comparison with the KNN method . We have chosen KNN for the comparison as it is virtually the only available technique that allows estimating MI and C* for systems with multidimensional output , Y . One of the test scenarios , Scenario 1 , is presented below whereas the remaining three scenarios , Scenarios 2-4 , are part of S1 Text . Scenario 1 demonstrates that the proposed approach , here further referred to as SLEMI , ( i ) provides more accurate estimates than the KNN method; ( ii ) provides robust , parameter-free estimates; and ( iii ) has desired properties in terms of the computational cost . Scenario 2 replicates diagnostics proposed for the KNN methods in reference [19] and demonstrates that , in contrast to the KNN method , accuracy of SLEMI estimates persists for high dimensionality of output data . Scenario 3 validates the accuracy of SLEMI estimates against several different shapes of the output distribution . Finally , Scenario 4 uses a model of a transcription factor activity [20] to demonstrate that SLEMI can be used to quantify the information capacity of frequency encoded signals . The test model used as Scenario 1 aims to reflect a simple experimental setup in which one-dimensional responses of individual cells to a range of stimuli are quantified . Precisely , the test model considers a channel with the log-normally distributed output , Y . The mean , μ ( x ) and variance σ2 of the log-output are assumed to be the sigmoid function , and a constant , respectively . Precisely , Y | x i ∼ exp ( N ( μ ( x ) , σ 2 ) ) , for μ ( x ) = V · x 1 + x , V = 10 , σ2 = 1 . For the above model , we considered eleven input values , m = 11 . Input vales range from 0 to 100 , xi ∈ [0 , 100] . Sample distributions of output corresponding to considered input values are shown in Fig 3A . To test the estimation accuracy , we computed information capacity estimates using SLEMI and KNN method for different sample size , N , i . e . , the number of data points corresponding to each input value used for estimation . True information capacity was evaluated numerically . Fig 3B presents both information capacity estimates , as well as the true value , as the function of N , for N ranging from 50 up to 2000 . For sample sizes typical for biological experiments , i . e . , tens or hundreds of measured cells , SLEMI clearly provides more accurate estimates than KNN . In contrast to SLEMI , KNN estimates depend on the choice of the parameter k , whereas clear rules how k should be selected are missing [56–58] . For computation of the KNN estimates in Fig 3A we used k = 10 , i . e . , 10 closest points to each observation y were used to approximate the density P ( y|X = xi ) . To highlight the impact of the parameter k , we re-calculated the KNN estimates of Fig 3B using k ranging from 2 to 50 , at fixed N = 1000 . As show in Fig 3C , selection of k has an impact on the value of the capacity estimates that may lead to considerable bias . SLEMI estimates are free from this disadvantage . Further , we have examined how the computational time needed to obtain estimates scales with sample size , N , and the number of input values , m . Fig 3D presents computational times as a function on N . Both methods exhibit linear increase with N ( Fig 3D ) . Computational time of SLEMI increases at the lower rate . Although we optimized implementations of both methods to ensure a fair comparison , the rate of increase may depend on specifics of the code used . Finally , we calculated the computational time as a function of the number of input values . Initially , we considered two input values , m = 2 , and increased , one by one , up to eleven input values , m = 11 . The computational time of SLEMI increases linearly , whereas computational time corresponding to KNN method increases at least quadratically ( Fig 3E ) . Linear scaling with respect to the number of input values is important for the method to be applicable to study complex systems with multiple inputs . In summary , the above test scenario shows that SLEMI provides more accurate estimates of the information capacity , C* , than the KNN method , especially for small sample size . Importantly , estimates are parameter-free , which contributes to robust estimation . Also , computational time scales linearly with respect to the number of input values . The test scenario presented above considered one-dimensional output , Y , as the demonstration of key benefits resulting from using SLEMI did not require a multidimensional complex model . The presented advantages , however , are of particular importance when systems with multivariate output , Y , are studied . For multivariate systems robust estimation of the information capacity , C* , with KNN method can be problematic due to the choice of the parameter k and numerical optimization . Therefore , in the test Scenario 2 , in S1 Text , we have confirmed that , unlike for the KNN method , high accuracy of information capacity estimates persist for multidimensional data . | In light of single-cell , live-imaging experiments understanding of how cells transmit information about identity and quantity of stimuli is incomplete . When exposed to the same stimulus individual cells exhibit substantial cell-to-cell heterogeneity . Besides , stimuli have been shown to regulate temporal profiles of signaling effectors . Therefore , it is , for instance , not entirely clear whether single-cell responses are binary or contain more information about the quantity of stimuli . The above questions resulted in a considerable interest to study cellular signaling within the framework of information theory . Unfortunately , the utilization of the information-theoretic perspective is handicapped in part by the lack of suitable methods that account for multivariate signaling data . Here , we propose a novel algorithm that breaks a considerable computational barrier by allowing the effective information-theoretic analysis of highly-dimensional single-cell measurements . Our approach is computationally efficient , robust and straightforward to use . Moreover , we provide a simple R-package implementation . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [
"applied",
"mathematics",
"cell",
"processes",
"social",
"sciences",
"neuroscience",
"biologists",
"simulation",
"and",
"modeling",
"algorithms",
"optimization",
"scientists",
"probability",
"distribution",
"mathematics",
"cognitive",
"psychology",
"science",
"and",
"techno... | 2019 | Information-theoretic analysis of multivariate single-cell signaling responses |
Chagas' disease , produced by Trypanosoma cruzi , affects more than 8 million people , producing approximately 10 , 000 deaths each year in Latin America . Migration of people from endemic regions to developed countries has expanded the risk of infection , transforming this disease into a globally emerging problem . PGE2 and other eicosanoids contribute to cardiac functional deficits after infection with T . cruzi . Thus , the inhibition of host cyclooxygenase ( COX ) enzyme emerges as a potential therapeutic target . In vivo studies about the effect of acetylsalicylic acid ( ASA ) upon T . cruzi infection are controversial , and always report the effect of ASA at a single dose . Therefore , we aimed to analyze the effect of ASA at different doses in an in vivo model of infection and correlate it with the production of arachidonic acid metabolites . ASA decreased mortality , parasitemia , and heart damage in T . cruzi ( Dm28c ) infected mice , at the low doses of 25 and 50 mg/Kg . However , this effect disappeared when the high ASA doses of 75 and 100 mg/Kg were used . We explored whether this observation was related to the metabolic shift toward the production of 5-lipoxygenase derivatives , and although we did not observe an increase in LTB4 production in infected RAW cells and mice infected , we did find an increase in 15-epi-LXA4 ( an ASA-triggered lipoxin ) . We also found high levels of 15-epi-LXA4 in T . cruzi infected mice treated with the low doses of ASA , while the high ASA doses decreased 15-epi-LXA4 levels . Importantly , 15-epi-LXA4 prevented parasitemia , mortality , and cardiac changes in vivo and restored the protective role in the treatment with a high dose of ASA . This is the first report showing the production of ASA-triggered lipoxins in T . cruzi infected mice , which demonstrates the role of this lipid as an anti-inflammatory molecule in the acute phase of the disease .
American Trypanosomiasis ( Chagas' disease ) is a parasitic illness caused by the flagellate protozoan Trypanosoma cruzi [1] . The area covered by this disease starts in the south of the United States and continues to the central area of Chile and Argentina . It has been present in America for 9 , 000 years [2] . In Latin America , Chagas' disease affects more than 8 million people , causing approximately 10 , 000 deaths each year , which is higher than malaria in the Americas , and covers 89% of the deaths caused by tropical-cluster diseases [3] . In addition , there is an annual productivity loss of US$1 . 2 billion due to Chagas' disease in the 7 endemic countries [4] . Furthermore , the migration of people from endemic regions to developed countries has expanded the risk of infection , especially through blood transfusions and organ transplants . As a consequence , there are currently immigrant infected populations in Japan , Australia , Spain , and in the United States , transforming this disease into an emerging global problem [5] . In addition , the impact of Chagas' disease in U . S . has been recently compared to the first years of the beginning of the VIH/AIDS epidemic [6] . The acute phase of Chagas' disease is characterized by immunosuppression induced by T . cruzi to evade the host immune response . This immunosuppressive state is mediated by prostaglandins [7] , [8] and cytokines , such as transforming growth factor-β ( TGF-β ) [9] . Increased circulating levels of prostaglandin E2 ( PGE2 ) [10] , thromboxane A2 ( TXA2 ) , and prostaglandin F2α ( PGF2α ) have been reported in mice infected with T . cruzi [11] , and during the acute phase , macrophages and spleen cells from T . cruzi-infected mice produce high levels of PGE2 [10] . Thus , as PGE2 and other eicosanoids might contribute to cardiac remodeling and other cardiac functional deficits after infection with T . cruzi , the inhibition of the host cyclooxygenase ( COX ) enzyme emerges as a potential therapeutic target . In infected BALB/c mice , treatment with aspirin , indomethacin or celecoxib decreases parasitemia and delays mortality [7] , [12] . However , some gaps remains in the literature data , since all assays described have been carried out with fixed doses of the COX inhibitor studied . Recently , the effect of ASA has been associated , at least in part , to a metabolic switch towards a pathway linked to the acetylation of the COX-2 isoenzyme . This acetylation enables COX-2 to synthetize other lipid products derived from AA , some of them with anti-inflammatory properties [13] . These metabolic products have been called “ASA-triggered lipoxins” ( ATLs ) . Correspondingly , ASA-triggered 15-epi-Lipoxin-A4 ( 15-epi-LXA4 ) has been described as an anti-inflammatory lipid able to inhibit IL-6 , TNF-α and IL-8 production , as well as NFκB , ERK1/2 and p38 activation [14] . In this report , we explore the relation between the protective role of ASA and the synthesis of 15-epi-LXA4 . In first place , we show that ASA treatment has a protective effect in T . cruzi-infected mice . However , this effect disappears with the higher doses employed . In addition , we found that infected mice treated with the effective ASA doses ( 25 or 50 mg/Kg/day ) produce 15-epi-LXA4 , whereas higher doses inhibit its production . Based upon these data , we propose that the protective role of ASA in experimentally T . cruzi-infected mice is related to the production of 15-epi-LXA4 .
All animal handling protocols were performed according to the “Guide for the Care and Use of Laboratory Animals” , from the National Institute of Health , USA [15] , and approved by the Institutional Ethical Committee at the Faculty of Medicine , University of Chile ( Protocol CBA# 0277 FMUCH ) , associated to FONDECYT-Chile grant number 1090078 . Adult male BALB/c mice ( 20–25 g ) were obtained from the Animal Facility at the Faculty of Medicine , University of Chile . Animals were first infected intraperitoneally with 30 , 000 T . cruzi blood trypomastigotes ( Dm28c strain ) . Afterwards , animals were randomized to receive the different treatments . T . cruzi infection was followed daily by parasitemia through direct microscopic visualization of circulating trypomastigotes from peripheral blood , as previously described [16] . Acetylsalicylic acid ( Sigma , USA ) was given diluted in the drinking water at concentrations that ranged from 19 . 5 to 390 mg/L , to achieve final doses from 5 to 100 mg/Kg/day , based in the observation that mice drank 6 . 4 mL of water daily . The water was available ad libitum . The bottles were replaced every morning with fresh water or drug solution , and the residual volume of water in bottles was measured to assure that the mice drank the intended water volume [17] , [18] . The treatment was initiated 48 hours after parasite inoculation , for 20 days at the doses indicated in each figure . As a measure of the pharmacological effect of orally administered aspirin , we determined the bleeding time in mice . To do this , we made a cut in the tail tip and bleeding was evaluated with filter paper every 15 seconds . These determinations were made at the day 10 post-infection ( p . i . ) . ( Table 1 ) . 5 ( S ) , 6 ( R ) , 15 ( R ) -Lipoxin A4 ( Cayman Chemicals , USA ) , was diluted daily in fresh PBS and administered i . p . for 10 days , starting at the fourth day p . i . , using the doses indicated in each figure . Controls received PBS exclusively . Mortality rate was recorded daily . Hearts were extracted at the moment of death from those mice dead before the end point . Surviving mice were euthanized at day 20 p . i . and their hearts were extracted . Hearts were longitudinally sectioned to further analysis by histopathology and qPCR . Samples were fixed in 10% formaldehyde 0 . 1 M phosphate buffer ( pH 7 . 3 ) for 24 h , dehydrated in alcohol , clarified in xylene , and embedded in paraffin . 5 µm sections were obtained and stained with hematoxylin-eosin for routine histopathological analysis as well as to evaluate the presence of T . cruzi amastigote nests and inflammation of the myocardium [19] . The automated morphometric analysis was performed by methodology developed by us , using the public domain software ImageJ ( ver 1 . 46 ) . The photographs of at least four nonconsecutive slides of tissue from different mice were assessed . The contrast of each photograph was increased automatically . To facilitate visualization of cell nuclei , the color channels ( red , blue and green ) were split . To isolate nuclei marks , the red channel was transformed in a binary image , using the “threshold” tool . To separate particles corresponding to adjacent nuclei , the “watershed” filter was applied . To exclude amastigote nuclei , only particles over 50 pixels were included in the particle count . Results of this procedure were compared with those obtained from manual count to verify the accuracy of the automatic counting . There was less than 10% variation between manual and automated count . To eliminate bias in the application of the methodology , the procedure was kept as a “macro” tool to apply the count automatically without intervention of the researcher . Hearts from infected animals treated with ASA or 15-epi-LXA4 were homogenized , and DNA was isolated using the Wizard Genomic DNA Purification Kit ( Promega , USA ) , following manufacturer's instructions . DNA was quantified through 280 nm absorbance measurements using a Varioskan spectrophotometer ( Thermo Scientific , USA ) . Parasite DNA quantification was performed using the primers TCZ-F ( 5′-GCTCTTGCCCACAMGGGTGC-3′ ) and TCZ-R ( 5′-CCAAGCAGCGGATAGTTCAGG-3′ ) , designed to amplify a 195 bp Satellite-DNA sequence of T . cruzi [20] . We used the TNFα-5241 ( 5′-TCCCTCTCATCAGTTCTATGGCCCA-3′ ) and TNFα-5411 ( 5′-CAGCAAGCATCTATGCACTTAGACCCC-3′ ) primers , which amplify a 170 bp sequence of the Mus musculus TNF-α gene as loading control [20] , [21] . PCR amplifications were carried out in the 7300 Real Time PCR system ( Applied Biosystems , USA ) . All reactions were performed using 10 ng of DNA and using the SensiMix SYBR Hi-Rox Kit ( Bioline , UK ) at a final volume of 20 µL . For both primer pairs , the thermal cycles consisted of one 10 min step of polymerase at 95°C , followed by 40 cycles of 15 s at 95°C , 15 s at 60°C , and 30 s at 72°C . Fluorescence was measured at the end of each amplification cycle . Finally , the melting curve was performed between 60 and 95°C . RAW 264 . 7 cells ( murine macrophages , ATCC number CRL-2922 ) were cultured at a density of 250 , 000 cells/cm2 , in RPMI 1640 medium , supplemented with 5% fetal bovine serum , in humidified air with 5% CO2 , at 37°C . RAW cells were infected with T . cruzi trypomastigotes ( Dm28c strain ) at a 3∶1 ratio ( trypomastigote∶RAW cell ) . Trypomastigotes were allowed to infect cells for 24 hours , after which supernatant was extracted . Cells were then washed twice with sterile PBS ( pH 7 . 4 ) to extract extracellular trypomastigotes . For protein isolation , RAW cells were washed with PBS , scraped , and lysed by sonication . Hearts from control or infected mice were homogenized at 4°C in a Potter-Elvejem homogenizer . All samples were homogenized in lysis buffer at pH 8 , containing Tris 0 . 01 µM , SDS 1% , and protease inhibitor cocktail ( Complete Mini EDTA-free , Roche , USA ) . Total protein was quantified using the bicinchoninic acid method , using the BCA Pierce kit ( Pierce Biotechnology , USA ) , following manufacturer instructions . For electrophoresis , extracted proteins were mixed with loading buffer ( 10% SDS , 50% glycerol , 0 . 5 M Tris , 0 . 1% bromphenol blue , and 1 M dithiothreitol , pH 6 . 8 ) , and 40 µg ( from the in vitro samples ) or 30 µg ( from the in vivo samples ) of total protein was loaded into 8% polyacrylamide gels . After , proteins were transferred to nitrocellulose membranes , and were blocked overnight with BSA ( 3% in PBST 0 . 05% ) . After three washes with PBST , membranes were incubated overnight at 4°C with primary polyclonal antibodies against COX-1 ( ab59964 , Abcam , UK ) , COX-2 ( ab52237 , Abcam , UK ) or 5-LOX ( ab59341 , Abcam , UK ) , diluted in PBST . Membranes were washed with PBST , and incubated with a HRP-conjugated anti-rabbit-IgG secondary antibody ( ab6721 , Abcam , UK ) for 1 hour . Afterwards , membranes were washed with PBST , and developed through chemiluminescence using the Pierce ECL Western Blotting Substrate ( Pierce Biotechnology , USA ) . After developing , membranes were incubated on stripping solution ( 62 . 5 mM Tris-HCl ( pH 6 . 8 ) , 2% SDS and 50 mM of mercaptoethanol ) at 50°C for 30 minutes . Membranes were blocked and incubated with a primary antibody against β-actin ( ab3280 , Abcam , UK ) , overnight at 4°C . We used a polyclonal anti-mouse-IgG Ab ( ab6728 , Abcam , UK ) HRP-conjugated as secondary antibody . Developed films were scanned , and band densitometry analysis was performed using ImageJ software . For PGE2 and LTB4 in vitro determinations , 106 RAW cells were cultured in 24 well plates , and exposed to 3×106 trypomastigotes ( Dm28c strain ) . After 24 hours , supernatant was removed and infected cells were washed twice with PBS ( pH 7 . 4 ) , and treated with ASA at different concentrations . After 48 hours of treatment , PGE2 and LTB4 were assayed in the supernatant , using the Prostaglandin E2 EIA Kit and Leukotriene B4 EIA Kit ( Cayman Chemicals , USA ) immunoassays , following manufacturer instructions . LTB4 from animal plasma was measured using the Parameter ( r ) Kit ( R&D Systems , USA ) , following manufacturer's instructions . 15-epi-LXA4 was assayed in the same model in vitro and in vivo , using the LXA4-15epi BioAssay ELISA Kit ( USBiological , USA ) . For in vivo determinations , plasma from infected mice was collected after 10 days p . i . , and assayed directly following manufacturer instructions . For all experiments , the statistical significance was established at p<0 . 05 . Results represent mean ± SD of triplicates . All statistical analyses were performed using GraphPad Prism ( 5 . 0 ) software . Normal distribution of data was assessed using D'Agostino-Pearsons analysis . One- and two-way ANOVA analyses ( with Tukey post-test ) or non-parametric Kruskal-Wallis analyses ( with Dunns post-test ) were performed when required . For survival analysis , the log rank test was performed .
We evaluated the effect of ASA on BALB/c mice infected with trypomastigotes of T . cruzi ( Dm28c strain ) , at doses ranging from 5 to 100 mg/Kg . Figure 1A shows that treatment with 25 and 50 mg/Kg ASA significantly increased the survival of infected mice ( p<0 . 01 and p<0 . 05 , respectively ) . However , when the highest doses were used , ( 75 and 100 mg/Kg ) , the mortality rate was similar to that observed in the infected control . Similarly , lower ASA doses ( <25 mg/Kg ) did not affect mortality ( data not shown ) . Thus , ASA impact upon survival was only observed at intermediate doses . Parasite DNA content was quantified by qPCR ( Figure 1B ) , no significant differences were observed between the control and the ASA-treated groups . ASA treatment was able to decrease parasitemia peaks at 25 and 50 mg/Kg ( Figure 1C ) . In mice treated with ASA 25 mg/Kg the parasitemia was decreased significantly on days 8 and 14 p . i . ( p<0 . 001 and p<0 . 01 respectively ) , while in the 50 mg/Kg group , parasitemia decreased significantly only on day 14 ( p<0 . 01 ) ( Figure 1C ) . Similarly , at 75 mg/Kg , the parasitemia peak on day 8 was decreased ( p<0 . 05 ) , although at a lower magnitude than that observed at 25 mg/Kg . No differences were observed at 100 mg/Kg when comparing to control . Finally , 5 and 10 mg/Kg of ASA did not show any differences when compared to control group ( data not shown ) . When we evaluated the cardiac structure of mice ( Fig . 1D ) , we found that infected mice exhibited severe inflammatory infiltration , associated edema , and amastigote nests . At 25 mg/Kg ASA , there was less inflammation and edema , and heart tissue histology seemed normal . These protective effects disappeared when the ASA dose was increased , as seen in 100 mg/Kg ASA-treated mice , which showed more edema and inflammation than controls . In addition , amastigote nests were more evident in 75 and 100 mg/Kg treated mice . A quantification of inflammatory infiltrate ( Figure 1E ) showed that ASA 25 , 50 and 75 mg/Kg decreased significantly the number of infiltrate cells . Although ASA 100 mg/Kg seems to have less infiltrate than control , this difference did not have statistical significance with the infected mice without treatment . We have previously reported that the effect of ASA upon in vitro T . cruzi infection is not reversed by exogenous PGE2 administration . Thus , we hypothesized that COX inhibitors , ASA in particular , have alternative mechanisms involved in this phenomenon . One possible mechanism could be ASA induction of a shift in the arachidonic acid metabolic pathway towards the production of 5-lipoxygenase derivatives [22] . Thus , we explored the variation in the metabolic pathway of AA , induced by ASA in an in vitro infection model . We assayed three concentrations of ASA , using 50% of our previously reported effect as a reference [23] . T . cruzi infection increased the PGE2 and LTB4 production in RAW 264 . 7 cells ( Figure 2A and 2B ) . Correlated with the increase of PGE2 production , the COX-2 levels in RAW cells infected with T . cruzi also increased significantly ( Figure 2D and 2E ) . As expected , ASA inhibited PGE2 synthesis at all tested concentrations ( Figure 2A ) . In contrast , LTB4 production in ASA treated cells did not increase . Unexpectedly , we observed a significant low level of LTB4 in the 0 . 5 mM ASA treated cells , when compared with infected cells ( Figure 2B ) . COX-2 acetylation by ASA modifies its activity , promoting the synthesis of 15-epi-LXA4 , a lipid involved in the resolution of inflammation [24] , [25] . Accordingly , we assessed the production of 15-epi-LXA4 in T . cruzi infected RAW cells treated with ASA . 15-epi-LXA4 production was significantly increased by ASA in these cells ( Figure 2C ) . Interestingly , the production of 15-epi-LXA4 was inversely correlated to the concentration of ASA , reaching a level similar to the control at 0 . 5 mM ASA . Based on the previous in vitro results , we evaluated the generation of 15-epi-LXA4 and LTB4 in T . cruzi-infected mice . In agreement with in vitro data , low doses of ASA increased the circulating levels of 15-epi-LXA4 , which decreased in a dose-depended manner ( Figure 3A ) . Unpredictably , infection alone produced an increase in 15-epi-LXA4 in mice . This discrepancy in the 15-epi-LXA4 production observed between infected cells and mice may be due to the lack of cooperative systems in single mammalian cell model . In fact , coincubation of macrophages with polymorphonuclear neutrophils increases the production of 15-epi-LXA4 [26] . Therefore , it is expected that 15-epi-LXA4 production in mice to be more efficient than in RAW cells monoculture . ASA did not modify LTB4 production in infected mice ( Figure 3B ) . Furthermore , T . cruzi infection did not change the basal levels of LTB4 ( Figure 3B ) , contrary to what we observed in infected RAW cells ( Figure 2B ) . In the heart tissue of T . cruzi infected mice , COX-2 levels presented a two-fold increase compared with control ( Figure 3C and 3D ) . These results are in agreement with previously reported data from immunohistochemical determinations [7] . Interestingly , the COX-2 levels appear to decrease with increasing ASA doses ( Figure 3C and 3D ) , indeed the COX-2 levels in the 75 and 100 mg/Kg treated groups were statistically different with infected control . This data could explain why 15-epi-LXA4 levels are decreased in infected mice treated with ASA 75 or 100 mg/Kg . In addition , COX-1 and 5-LOX levels in cardiac tissue did not change with infection or treatment ( Figures 3C , 3E and 3F ) . Thus , the changes in 15-epi-LXA4 production could be related with ASA effects on the COX-2 enzyme . Since ASA treatment can modify synthesis of 15-epi-LXA4 both in vitro and in vivo , we evaluated the effect of exogenous administration of 15-epi-LXA4 in the T . cruzi infection outcome in BALB/c mice . Infected mice were treated with 5 or 25 µg/Kg 15-epi-LXA4 according to previously reported schemes for other murine models [27] . 15-epi-LXA4 at 5 µg/Kg had no effect on the survival rate and cardiac parasite burden when quantified by relative DNA load ( Figures 4A and 4B ) . In contrast , 15-epi-LXA4 at 25 µg/Kg significantly increased survival and decreased cardiac parasite load ( Figure 4A and 4B ) . On the other hand , treatment with 5 and 25 µg/Kg 15-epi-LXA4 significantly decreased the parasitemia peaks observed on days 12 and 14 p . i . ( Figure 4C ) . Cardiac histopathological analysis showed that 25 µg/Kg 15-epi-LXA4 decreased number of amastigote nests and the inflammatory infiltration ( Figures 4D and 4E ) . Nevertheless , at 5 µg/Kg , focal inflammatory infiltration and amastigote nests persisted , as compared with untreated infected controls . Considering the above results , do high ASA doses lose its general protective effect on T . cruzi infected mice due to absence of 15-epi-LXA4 production ? To answer this question , we administered 25 µg/Kg of 15-epi-LXA4 to T . cruzi-infected mice , treated with either 75 or 100 mg/Kg ASA . Although there were no significant effects of 15-epi-LXA4 on survival or cardiac parasite burden ( Figures 5A and 5B ) , parasitemia , inflammatory infiltrate , and amastigote nests decreased when 15-epi-LXA4 was administered to the 75 mg/Kg ASA treated mice ( Figures 5C , 5D and 5E ) . 15-epi-LXA4 did not produce effect in the 100 mg/Kg ASA treated-mice ( Figure 5 ) .
In this report , we showed that ASA decreased mortality , parasitemia , and heart damage in T . cruzi infected mice , at doses of 25 and 50 mg/Kg , doses below 25 mg/Kg did not alter the natural course of the disease in the infected mice , while mice treated with 75 and 100 mg/Kg/day of ASA , showed more intense symptoms . These results could be related to inhibition of COX-2 and consequent decrease in prostaglandin E2 production , with a metabolic shift to 5-LOX derivatives . However , as the expected increase in LTB4 production was not observed , the beneficial effect of this COX-2 inhibitor could be explained by the alternative production of 15-epi-LXA4 . Indeed , this molecule prevented parasitemia , mortality and cardiac changes during the acute infection in vivo . Arachidonic acid cascade research is an open field in the T . cruzi-host interaction studies . In vivo studies about the effect of COX inhibitors upon T . cruzi infection are controversial , because the reported results vary depending upon the mouse or T . cruzi strain used , the parasite inoculum , the type of inhibitor , or the therapeutic scheme [7] , [8] , [12] , [28]–[31] . During early infection , treatment with aspirin or indomethacin dramatically increases parasitemia , and reduces the survival rate of T . cruzi-infected C57BL/6 , C3H/HeN or CD-1 mice , all of which have been described as resistant to the acute infection [8] , [29] , [31] , [32] . By the contrary , there is evidence the points out that COX inhibitors might improve parasitemia , heart damage and survival in T . cruzi-infected BALB/c mice [7] , [12] , [28]–[30] . This contradictory data might be explained by the variation in the production of arachidonic acid derivatives under different experimental conditions . Here , we found that low doses of ASA significantly improved the outcome of T . cruzi-infected mice and demonstrated that this effect is related with the production of 15-epi-LXA4 , a metabolite of AA not previously associated with ASA effect on the Chagas' disease . During acute infection , PGE2 , TXA2 , PGI2 , and PGF2α are increased [11] , [30] , [33] . PGE2 and TXA2 appear to modulate the host response against the parasite , and facilitate the shift to the chronic phase . Parasite-derived TXA2 is essential for parasite survival , regulation of amastigote replication during the acute stage , and modulation of the cardiac disease [33] , [34] . Moreover , knocking out the host TXA2 receptor increases parasite load in cardiac tissue [34] . According to this , COX inhibition by ASA would be expected to produce a similar effect , but we found that aspirin treatment had a protective effect without decreasing parasite load in the heart tissue . Thus , at 25 mg/Kg ASA , cardiac tissue appeared normal; thus becoming evident that prostaglandins are not the only arachidonic acid metabolites involved in T . cruzi infection . In consequence , the parasitological protection provided by ASA , evidenced mainly by increase of survival and histological “normalization” in heart is probably , the result of a better immunological control provided by the metabolic shift toward the 15-epi-LXA4 production and the decrease in prostaglandins and thromboxane levels . The role of 5-lipoxygenase and its metabolites in Chagas' disease has also been previously studied [35]–[37] . It has been reported that leukotrienes participate in parasitemia control , and are important for survival during the early acute phase of the infection , by facilitating the trypanocidal activity of phagocytic cells achieved through NO• production regulation [38] . This is in agreement with the observation that 5-lipoxygenase null-mutant mice infected with T . cruzi show a higher mortality rate than wild-type infected mice [36] . However , leukotriene participation in the overall Chagas' disease physiopathology process has not been described as pivotal , because there could be other mechanisms independent of leukotrienes through which an increase in NO• production could be achieved , as demonstrated by Panis and colleagues [36] . When we investigated the effect of ASA on RAW 264 . 7 cells , we found that prostaglandin E2 production was decreased , which indicates that COX-2 was inhibited . The shift of arachidonic acid metabolism toward leukotriene synthesis was ruled out , because ASA did not increase the LTB4 levels either in vitro or in vivo . Furthermore , in vitro , the level of LTB4 decreased as ASA concentration was increased , indicating that this drug also affected LTB4 metabolism at high doses . In a Toxoplasma gondii infection model , treatment with ASA or LXA4 induces migration of dendritic cells ( DCs ) , and in vivo production of interleukin ( IL ) -12 , through the induction of suppressor of cytokine signaling ( SOCS ) -2 expression , demonstrating a role of ATLs in parasitic infection control [39] . In agreement with those results , we showed that in T . cruzi infected mice , 15-epi-LXA4 decreased inflammatory infiltration in cardiac tissue and improve disease outcome . In addition , increased levels of ATLs in infected mice , and the decrease of parasitic load in cardiac tissue after treatment with 15-epi-LXA4 , might be related to a potential role for this molecule in parasite contention , aside from its role in the resolution of the inflammation . In the context of T . cruzi infection , there is production of 15-epi-LXA4 , especially at low doses of ASA . This result agrees with the anti-inflammatory effect seen in humans when low doses of ASA were administered . This effect was related to ATL production [24] . In addition , a human trial showed that 15-epi-LXA4 production decreased when high doses of ASA were supplied [40] . Thus , the divergent effects observed with low and high doses of ASA observed in our model of murine Chagas' disease are supported by these results . However , the clinical utility of our findings might be limited as the effects are only found in one strain . Conversely , the findings of disease aggravation with ASA are not strain specific and may reflect the more likely clinical scenario in a genetically diverse population , such as the patients likely to encounter the disease In conclusion , this is the first report showing the production of aspirin triggered lipoxins in T . cruzi infected mice suggesting a role of this lipid as an anti-inflammatory molecule in the acute phase of the disease . | Chagas' disease is an infection produced by the parasite Trypanosoma cruzi . This pathology is endemic in Latin America and has become a public health issue in some non-endemic countries like the USA , Spain and Australia . There is no curative treatment against Chagas' disease . NSAIDs , like aspirin , have been assayed as drugs with therapeutic potential in Chagas' disease , but the studies about this issue show contradictory results; also , the mechanism involved in aspirin effect is not yet clear . In this study , we explore a broad range of doses the protective role of aspirin . We found that aspirin has a therapeutic effect at low doses , an effect that disappears when doses are increased . This phenomenon correlates with the presence of 15-epi-LXA4 , a molecule known as an “aspirin-triggered lipoxin , ” which increases at low doses of aspirin , and decreases when aspirin dose is increased . 15-epi-LXA4 has been related with the anti-inflammatory effect of aspirin; in this setting , we found that 15-epi-LXA4 is able to decrease the cardiac inflammation and others parameters related with Chagas' disease . Finally , we present the first study that shows that the protective effect of aspirin on Chagas' disease could be mediated by the synthesis of 15-epi-LXA4 . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [
"medicine",
"infectious",
"diseases",
"chagas",
"disease",
"neglected",
"tropical",
"diseases",
"host-pathogen",
"interaction",
"biology",
"microbiology",
"parasitic",
"diseases",
"parasitology"
] | 2013 | Protective Role of Acetylsalicylic Acid in Experimental Trypanosoma cruzi Infection: Evidence of a 15-epi-Lipoxin A4-Mediated Effect |
In auditory cortex , temporal information within a sound is represented by two complementary neural codes: a temporal representation based on stimulus-locked firing and a rate representation , where discharge rate co-varies with the timing between acoustic events but lacks a stimulus-synchronized response . Using a computational neuronal model , we find that stimulus-locked responses are generated when sound-evoked excitation is combined with strong , delayed inhibition . In contrast to this , a non-synchronized rate representation is generated when the net excitation evoked by the sound is weak , which occurs when excitation is coincident and balanced with inhibition . Using single-unit recordings from awake marmosets ( Callithrix jacchus ) , we validate several model predictions , including differences in the temporal fidelity , discharge rates and temporal dynamics of stimulus-evoked responses between neurons with rate and temporal representations . Together these data suggest that feedforward inhibition provides a parsimonious explanation of the neural coding dichotomy observed in auditory cortex .
Temporal processing is fundamentally important for perceiving and discriminating acoustic stimuli [1 , 2] . Specifically , the timing between successive acoustic events is used by the auditory system in the recognition of musical rhythms [3 , 4] , human speech [5 , 6] and conspecific vocalizations [7 , 8] . When acoustic events occur in a sequence , our perception of this sequence depends on the time interval between successive events . When inter-event time intervals are longer than 25 ms , we perceive a stream of discretely occurring sounds , commonly referred to as acoustic flutter [9 , 10] . At shorter time intervals , this percept changes from flutter to fusion; the sensation of discretized events is lost and the resulting fused percept generally has a pitch equal to the repetition rate of acoustic events [11] . The flutter/fusion perceptual boundary is not unique to the auditory system . An analogous perceptual boundary occurs for both visual stimuli ( flicker/fusion ) [12] and tactile stimuli ( flutter/vibration ) [13] . The dichotomous categorization of a sequence of brief sounds into the perceptions of flutter and fusion is reflected in the corresponding neural representations of these sounds . Within auditory cortex , a sequence of brief sounds , hereinafter referred to as an acoustic pulse train , is encoded with either a temporal or rate representation , for longer and shorter interpulse intervals ( IPIs ) , respectively [14–17] . A temporal representation is provided by neurons with envelope-locked responses , referred to as “synchronized neurons” , reflecting their ability to synchronize their spikes to each acoustic pulse ( Fig . 1a ) . However , the temporal fidelity of this synchronization degrades at shorter IPIs , with an encoding boundary near the flutter/fusion perceptual boundary . In the perceptual range of fusion , synchronized neurons generally elicit only an onset response , and thus cannot be used to discriminate between these shorter IPIs ( Fig . 1b ) . In addition to synchronized responses , neurons can also produce “non-synchronized” responses to acoustic pulse trains [14 , 17–19] . Non-synchronized neurons increase their firing rate monotonically with decreasing IPIs over the perceptual range of fusion without exhibiting envelope-locked responses ( Fig . 1b ) . While non-synchronized neurons are generally unresponsive at IPIs in the range of flutter ( Fig . 1a ) , the combined neural representations from synchronized and non-synchronized neurons are sufficient to encode temporal information across a wide range of IPIs , spanning the percepts of both flutter and fusion . This dichotomy between synchronized and non-synchronized responses is not unique to the auditory system; an analogous dichotomy exists in primary visual cortex for representing spatially modulated stimuli [20–22] . When presented with drifting visual gratings , simple cells synchronize their firing to individual bars of the gratings while complex cells produce a non-synchronized discharge pattern . This difference is reflected in the organization of each cell’s receptive field- excitation and inhibition are spatially segregated in simple cells , but spatially overlapping in complex cells . We reasoned that synchronized and non-synchronized responses in auditory cortex could be generated by a similar relationship between excitation and inhibition , with the degree of segregation between these two inputs varying in the time domain , rather than the spatial domain . To investigate this , we simulated an auditory cortical neuron using an integrate-and-fire computational neuronal model [23–24] , and measured how changing the relative timing between excitatory and inhibitory inputs affected a neuron’s representation of temporal information .
We used two tests to classify neurons as synchronized or non-synchronized [14] . A synchronized neuron was required to have statistically significant vector strength at the longest IPI tested ( Rayleigh statistic>13 . 8 , P<0 . 001 , at IPI = 75ms ) [25] . Non-synchronized neurons were required to have a discharge rate ratio greater than one ( i . e . the discharge rates at shorter IPIs had to be greater than at longer IPIs ) . If a neuron passed both of these criteria , it was classified as having a mixed response ( S1 Fig ) . Mixed response neurons have been previously reported in auditory cortex [14] , but at significantly lower proportions than either synchronized or non-synchronized neurons . If a neuron did not pass either criterion , it was classified as having an “atypical” response . In addition to these criteria , we only included neurons in our analysis with pure tone evoked discharge rates in the range of 1 to 50 spk/s . Our lower bound of 1 spk/s was to prevent the inclusion of neurons that were unresponsive . Our upper bound of 50 spk/s was to only include neurons that had discharge rates representative of a typical auditory cortical neuron [26] and avoid physiologically unrealistic responses . Using these criteria , we observed that the vast majority of neurons ( 98% ) generated by our model could be classified as having a synchronized , non-synchronized , or mixed response ( Fig . 2b ) . In order to directly compare our computational model with real data , we reanalyzed a previously published dataset [15 , 18 , 26] , composed of single-unit responses to acoustic pulse trains from the auditory cortex of four awake marmosets ( Callithrix jacchus ) ( see Methods ) . Using the same criteria as in our computational model , 70% of units responding to our acoustic pulse train stimuli ( 147/210 units ) could be classified as having synchronized , non-synchronized or a mixed response ( Fig . 1 , S1 Fig ) . The remaining neurons were more heterogeneous in their response properties; generally being weakly stimulus synchronized ( but not at an IPI of 75 ms ) and/or having an excited or suppressed response over a range of IPIs , with a tuning curve that was bandpassed , all passed , or high passed ( only IPIs in the range of flutter perception ) . We also observed some of these less common response types in our model , including bandpassed and inhibitory ( S2 Fig ) . Next we compared the rate and temporal representations generated by real and simulated neurons . We observed that synchronized neurons , both real and simulated , were able to represent long IPIs using temporally locked responses ( Fig . 2c ) , although simulated neurons had a substantially better temporal fidelity . Non-synchronized neurons , both real and simulated , represented shorter IPIs using a rate code ( Fig . 2d ) , in which their normalized discharge rate decreased monotonically between IPIs in the range of 3 and 25 ms ( i . e . higher discharge rates at shorter IPIs ) . Thus , the general features of temporal and rate representations produced by synchronized and non-synchronized neurons , respectively , were preserved in our computational model . What determines whether a neuron encodes an acoustic pulse train using a temporal and/or a rate representation ? We observed that synchronized , non-synchronized , and mixed responses were generated within three distinct regions of the model’s parameter space ( Fig . 3 ) . Synchronized neurons were more common when inhibition lagged excitation , and the magnitude of inhibition was at least 40–50% stronger than the magnitude of excitation ( Fig . 4a , model with an I-E delay = 5 ms ) . Comparing synchronized neurons with a fixed I-E delay of 5 ms , we observed a highly significant correlation ( r = 0 . 99 , P<3 . 1x10-87 , Spearman Correlation , Fig . 4b ) between the excitatory input strength and the Rayleigh statistic , the criterion we used to measure the statistical significance of stimulus-synchronization . In other words , as the strength of excitation increased , our confidence in the high fidelity of a synchronized neuron’s temporal representation improved . This can be observed by comparing the responses of two simulated synchronized neurons that differ in their excitation strength parameter ( Fig . 4c-d ) . While both of these examples of simulated neurons differ in the robustness of their temporal representation , they closely match the general properties of real synchronized neurons ( Fig . 4e ) . In contrast to this , non-synchronized neurons were more common when the net excitation was weak , which occurred for I/E ratios close to one ( balanced excitation and inhibition ) or low I/E ratios in combination with a weak excitatory input ( Fig . 5a , model with an I-E delay = 0 ms ) . We observed a statistically significant correlation ( r = 0 . 87 , P<1 . 5 x 10-17 , Spearman Correlation , Fig . 5b ) between the net excitatory input ( excitation-inhibition ) and the discharge rate ratio . In other words , as the magnitude of net excitation increased , short IPIs evoked a higher discharge rate relative to the discharge rate at longer IPIs , effectively increasing the dynamic range of the neuron’s rate code . This effect can be observed by comparing the responses of two simulated non-synchronizing neurons that differ in their net excitation ( Fig . 5c-d ) . While both of these examples of simulated neurons differ in the dynamic range of their rate representation , they closely match the general properties of real non-synchronized neurons ( Fig . 5e ) . When the net excitation increased above a conductance of approximately 0 . 6 nS , non-synchronized responses generally became mixed responses ( Fig . 3 , S1 Fig ) . Mixed responses could occur when the excitatory input of a non-synchronized neuron increased in strength , or alternatively when the inhibitory input of a synchronized neuron decreased in strength . In our computational model , roughly equal proportions of synchronized , non-synchronized , and mixed response neurons were generated ( Fig . 3 ) . This differs from our single-unit recordings , for which mixed response neurons ( 10% , 14/147 neurons ) were less frequently observed than either synchronized or non-synchronized neurons . Because each parameter was uniformly distributed in our model , our results map the possible response types that can be generated , but not their relative proportions . For example , we allowed simulated neurons to have pure tone evoked discharge rates in the range of 1–50 spk/s , reflecting the range of discharge rates observed in our real data . However while evoked discharge rates can reach more than 50 spk/s in auditory cortex , lower discharge rates are more typical ( median discharge rate = 16 . 4 spk/s ) . Reflecting this , we observed that when our criteria for a physiologically realistic response was lowered from 50 spk/s to 20 spk/s , the proportion of mixed response neurons decreased to about 12% , matching the proportion found in our single-unit recordings ( S3 Fig ) . Using our computational model , we could make several predictions concerning how stimulus-evoked responses may differ between synchronized , non-synchronized , and mixed-response neurons , which were testable in our dataset of real neurons . In our model , delayed inhibition in synchronized neurons led to a positive net excitation concentrated at the onset of the synaptic input ( Fig . 6a ) , while balanced excitation and inhibition in non-synchronized neurons led to weaker net excitation that was spread out over a longer time duration ( Fig . 6b ) . Because of this difference , an evoked response from a synchronized neuron only required a single acoustic pulse , while the evoked response of a non-synchronized neuron required the temporal summation of inputs from multiple acoustic pulses . We reasoned that this difference should also be reflected in the temporal dynamics of the neuron’s response to acoustic pulse trains , with a synchronizing neuron having a shorter latency to the onset of its response ( minimum latency ) than a non-synchronizing neuron . We measured the minimum latency of acoustic pulse train responses ( see Methods ) , and found a statistically significant difference between synchronized and non-synchronized neurons , within both our simulated and real neuronal populations ( Fig . 6c , d ) . In the simulated neuronal population , synchronized neurons had a mean minimum latency of 10 . 8 ms , while non-synchronized neurons had a mean minimum latency of 16 . 6 ms ( Wilcoxon rank sum test , P < 1 . 4 x 10-89 ) . A similar difference occurred in the real neuronal population; synchronized neurons had a mean minimum latency of 18 . 1 ms , while non-synchronized neurons had a mean minimum latency of 51 . 1 ms ( Wilcoxon rank sum test , P < 4 . 6 x 10-9 ) . Like synchronized neurons , mixed response neurons are also capable of envelope locking , with only a single acoustic pulse required to evoke a response . Based on this similarity , we reasoned that the minimum latency should be similar between synchronized and mixed response . We observed that mixed neurons had a mean minimum latency ( simulated neurons: 8 . 0 ms , real neurons: 16 . 2 ms ) not significantly different from synchronized neurons ( Wilcoxon rank sum test , P = 0 . 053 ( simulated ) , P = 0 . 30 ( real ) ) and significantly different from non-synchronized neurons ( Wilcoxon rank sum test , P<3 . 1x10-75 ( simulated ) , P<1 . 1x10-5 ( real ) ) . Using a similar reasoning , we hypothesized that the temporal dynamics of pure- tone responses should also differ between synchronized and non-synchronized neurons . Excitation concentrated at the onset of the synaptic input in synchronized neurons should evoke an onset response , while net excitation spread out over a longer time period in non-synchronized neurons should produce a more sustained response . To examine this we calculated the onset/sustained ratio to pure tones in our population of simulated and real neurons ( see Methods ) , where a value of 1 indicated a pure onset response and a value of 0 . 25 indicated a sustained response . We observed a statistically significant difference between synchronized and non-synchronized responses in both simulated and real neurons ( Fig . 6e , f ) . In the simulated neuronal population , synchronized neurons had a mean onset/sustained ratio of 0 . 69 , while non-synchronized neurons had a mean onset/sustained ratio of 0 . 18 ( Wilcoxon rank sum test , P < 6 . 1 x 10-124 ) . A similar difference occurred in the real neuronal population; synchronized neurons had a mean onset/sustained ratio of 0 . 60 , while non-synchronized neurons had a mean onset/sustained ratio of 0 . 25 ( Wilcoxon rank sum test , P < 2 . 3 x 10-9 ) . Thus , in both our real and simulated neuronal populations , synchronized neurons tended to have onset responses to pure tones , while non-synchronized neurons typically had sustained responses . This difference was not due to a slightly longer latency response in non-synchronizing neurons , as we also observed a statistically significant difference between synchronized and non-synchronized neurons when the time window used to calculate the onset discharge rate was lengthened to 100 ms ( simulated neurons: P<1 . 61 x 10-115 , real neurons: P<2 . 3x10-5 , Wilcoxon rank sum test , see Methods ) . These data suggest that the temporal dynamics of a pure tone response ( onset or sustained ) can be used to predict whether a neuron has a synchronized or non-synchronized response to an acoustic pulse train . However , some neurons can change between an onset and a sustained response when a sound’s acoustic parameters are altered; a sustained response is generally evoked by a preferred stimulus ( best frequency and sound level ) while an onset response can be generated by non-preferred stimuli [27] . Thus a neuron that has a non-synchronized response at its preferred sound level could theoretically switch to generate a synchronized response , when the sound level is increased above its preferred level . This can be observed in an example real neuron ( see S4 Fig ) . When the sound level of a pure tone ( at the neuron’s best frequency ) was varied , this neuron responded with a sustained response at 10 dB SPL , and with an onset response at 70 dB SPL , causing a shift in the onset/sustained ratio from 0 . 4 to 0 . 8 respectively ( S4 Fig , a-b ) . In response to acoustic pulse trains , this same neuron produced a non-synchronized response at 10 dB SPL , and was able to synchronize to acoustic pulse trains with IPIs above 20 ms when the sound level was elevated to 70 dB SPL ( S4 Fig , c-d ) . The ability to switch between different neural coding regimes was observed in approximately 44% of neurons ( 8/18 ) tested with multiple acoustic pulse trains differing in sound level , frequency , or pulse width . According to our computational model ( Fig . 3 ) , when either a synchronized neuron’s inhibitory input was reduced or a non-synchronized neuron’s excitatory input was increased , the neural coding regime changed to a mixed response ( i . e . non-synchronized for short IPIs and synchronized for long IPIs ) . This implies that mixed response neurons had a larger net excitation than either synchronized or non-synchronized neurons , which we reasoned should manifest as a larger pure tone evoked response . We observed that for the simulated neuronal population , mixed neurons had a significantly higher discharge rate to pure tones than either non-synchronized and synchronized neurons ( Fig . 7a , mixed = 29 . 7 spk/s , nonsync = 13 . 9 spk/s , sync = 3 . 3 spk/s; Wilcoxon rank sum test , P< 1 . 2 x 10-76 , Bonferroni corrected ) . We also observed a significant difference between non-synchronized and synchronized neurons ( Wilcoxon rank sum test , P< 6 . 9 x 10-96 , Bonferroni corrected ) . We found a similar effect in our real neuronal population- mixed neurons had a significantly higher discharge rate to pure tones than either synchronized or non-synchronized neurons ( Fig . 7b , mixed = 51 . 3 spk/s , nonsync = 22 . 5 spk/s , sync = 18 . 3 spk/s; Wilcoxon rank sum test , P< 0 . 003 , Bonferroni corrected ) . While non-synchronized neurons had a slightly higher pure tone evoked response to pure tones than synchronized neurons , this difference was not statistically significant ( Wilcoxon rank sum test , P = 0 . 58 , uncorrected ) . Compared with real neurons , simulated synchronized neurons generally had lower firing rates . One potential reason for this is a higher percentage of simulated neurons receiving very strong inhibition than in our real neuronal population . However , we still observed both onset and sustained pure tone responses in simulated neurons , representative of typical real synchronizing neurons . Strong delayed inhibition ( e . g . an I/E ratios of 2 ) typically generated onset responses with delayed suppression , while more moderate delayed inhibition ( e . g . an I/E ratio of 1 . 4 ) typically generated onset responses with sustained activity ( S5 Fig ) . We also hypothesized that the strength of a neuron’s inhibitory input would impact its temporal fidelity . Delayed inhibition suppresses random spiking that can occur between acoustic pulses; the resulting stimulus-locked response has a higher vector strength , thus providing a better temporal representation of the acoustic pulse train . As synchronized neurons have stronger delayed inhibition than mixed response neurons , synchronized neurons should therefore have higher vector strengths . However , this comes at a cost , as delayed inhibition prevents the response to a second acoustic pulse during the brief period that the neuron is suppressed . Thus mixed response neurons ( which have less inhibition ) should be able to stimulus synchronize at shorter IPIs than synchronized neurons . As predicted , we observed that for our simulated neuronal population , synchronized neurons had higher maximum vector strengths than mixed response neurons ( Fig . 8a , sync = 0 . 93 , mixed = 0 . 79; Wilcoxon rank sum test , P< 3 . 3 x 10-52 , see Methods ) , while mixed response neurons had lower stimulus synchronization limits ( Fig . 8b , sync = 10 . 2 ms , mixed = 7 . 7 ms; Wilcoxon rank sum test , P< 1 . 3 x 10-43 ) . We observed a similar trend for our real neuron population- synchronized neurons had higher maximum vector strengths than mixed response neurons ( Fig . 8c , sync = 0 . 68 , mixed = 0 . 60; Wilcoxon rank sum test , P = 0 . 16 ) , while mixed response neurons had lower stimulus synchronization limits ( Fig . 8d , sync = 25 . 7 ms , mixed = 13 . 4 ms; Wilcoxon rank sum test , P< 0 . 02 ) . Although we only observed a statistically significant difference between synchronized and mixed response neurons for the stimulus synchronization limit and not maximum vector strength , this may be due to the limited number of mixed response neurons that we were able to record from ( n = 14 ) . These results suggest that blocking inhibition ( e . g . by adding a GABA-A antagonist such as Gabazine [28] ) , effectively decreasing the I/E ratio , should decrease a neuron’s vector strength while increasing its stimulus synchronization limit . However , the relationship between inhibition and temporal fidelity is more complex . While for simulated neurons with the same excitatory input strength , the stimulus synchronization limit of synchronized neurons decreased as the I/E ratio increased ( P<0 . 001 , Spearman correlation coefficient ) , we did not observe a statistically significant trend between the stimulus synchronization limit and I/E ratio in mixed response neurons ( P>0 . 05 , Spearman correlation coefficient ) . In our computational model , the time-varying conductance used to simulate the neuron’s synaptic input was simplified to only approximate the AMPA and GABA-A currents evoked by the acoustic pulse train , with a time-constant of 5 ms [24] ( see Methods ) . The synchronization limit of simulated neurons was also sensitive to this time constant; increasing this time constant ( S6 Fig , a-d ) or adding an additional NMDA-based conductance ( S6 Fig ( e-f ) [29] , see Methods ) shifted the synchronization limit of simulated neurons to longer IPIs ( mean synchronization limit: sync = 15 . 9 ms , mixed = 15 . 3 ms ) . Our computational model operated with a fixed spontaneous rate ( ~4 spk/s ) , close to the median spontaneous rate encountered in our real neuronal population ( 3 . 8 spk/s ) . To generate a spontaneous rate , we added Gaussian noise to the excitatory and inhibitory conductances of the neuron . If the amplitude of this Gaussian noise was increased , the spontaneous rate increased monotonically ( Fig . 9a ) . We next examined how sensitive our computational model was to changes in spontaneous rate . We examined spontaneous rates covering the entire range observed in our real neuronal population ( 0–40 spk/s ) and found that the model parameters for generating synchronized and non-synchronized neurons were similar , albeit with a slight shift in the threshold I/E ratio for observing synchronized responses ( Fig . 9b , c ) . We next examined how robust each neuronal representation ( sync , non-sync , mixed ) was across varying spontaneous rates ( 0–40 spk/s ) , and observed that a large fraction of synchronized ( 67% ) and non-synchronized ( 52% ) neurons did not change their neural coding regime across the entire range of spontaneous rates tested ( Fig . 10 , S7 Fig ) . In contrast to this , only 15% of mixed response neurons showed a similar invariance ( Fig . 10 ) . The addition of internal neuronal noise ( see Methods ) , which generated the neuron’s spontaneous rate , was critical for our computational model . Removing this noise from our model almost completely eliminated non-synchronized responses , although mixed and synchronized responses were still preserved ( S8 Fig ) . Furthermore , the non-synchronized responses generated in the absence of added noise had the unusual behavior of stimulus synchronizing at short IPIs , which was not a property of non-synchronized responses in our real and simulated neuronal populations ( S8 Fig ) . Increasing the temporal jitter of synaptic inputs ( S9 Fig ) generated non-synchronized responses more typical of real neurons , while maintaining synchronized responses . While our model’s ability to generate non-synchronized responses required a source of internal noise ( or sufficient temporal jitter of synaptic inputs ) , other methods of generating internal noise also produced similar results , including injecting noise as a current into the integrate-and-fire model to simulate background synaptic activity [30] ( S10 Fig ) and adding Gaussian noise to the membrane potential spiking threshold [31] ( S11 Fig ) .
Temporal information within an acoustic signal can be represented in auditory cortex by a temporal representation of envelope-locked responses and/or a rate representation where discharge rate varies with interpulse interval [32] . Here we propose that the timing and magnitude of inhibition relative to excitation determines whether a neuron uses a temporal and/or rate representation . Using a computational model of an auditory cortical neuron , we found that synchronized responses were generated when strong inhibition lagged excitation . This created a stimulus-locked response at long interpulse intervals ( IPIs ) , but an onset response followed by suppression at short IPIs . Conversely , non-synchronized responses were generated when excitation and inhibition were concurrent and balanced , which resulted in weak net excitation . This produced an input that was too weak to generate a response to a single acoustic pulse , but for sufficiently short IPIs ( such that two or more pulses occurred within the neuron’s temporal integration window ) , the net excitation was sufficient to evoke a response . As the IPI duration was shortened further , more acoustic pulses occurred within the neuron’s temporal integration window and evoked a higher discharge rate , in turn creating a monotonic relationship between discharge rate and decreasing IPI . Thus non-synchronized neurons act as “integrators” , while the envelope-locked responses of synchronized neurons behave as “differentiators” . It is important to note that although non-synchronized neurons encode temporal rates in the range of pitch perception [33–35] , they are insensitive to periodicity , and thus do not encode pitch salience [18] and may instead be generating a sensation more akin to roughness perception [36] . The parameters of our computational model were based on previously reported intracellular data from rodent auditory cortex [24] , and consistent with intracellular data reported by other laboratories [37–41] . Based on the relationship between excitation and inhibition used to generate temporal and rate representations in our computational model , we were able to make several testable predictions including differences in the temporal fidelity , discharge rates and temporal dynamics of evoked responses , which were subsequently confirmed in our real neuronal population . One important observation supporting our model was that changes in the timing and strength of inhibition , occurring when sound level or frequency were altered [40] , could shift a real neuron’s response between synchronized and non-synchronized ( or vice-versa ) . While this indicates that a neuron’s neural code is not fixed , it is important to note that other potential mechanisms , such as synaptic facilitation and depression [42–44] , recurrent connections [45–46] , lateral connections [30 , 47] , dendritic computation [48] , or other non-linear neuronal properties [49] could also potentially contribute to the generation of synchronized and non-synchronized responses , and are not directly ruled out by our model . Using synaptic depression and facilitation , Rabang and Bartlett [50] have demonstrated an alternative method for generating temporal and rate representations; large inputs with synaptic depression create synchronized responses while weak inputs with mixed plasticity ( synaptic depression of AMPA receptors and synaptic facilitation of NMDA receptors ) generate mixed and non-synchronized responses . While synaptic facilitation and depression likely play a central role in temporal processing , we have demonstrated in our computational model that in the absence of synaptic facilitation and depression , feedforward inhibition is sufficient to generate many of the properties of temporal and rate representations observed in auditory cortex . While non-synchronized responses have been reported in several different species ( e . g . monkey [14 , 17–18] , cat [19] , and rat [51] ) , many previous studies have only reported stimulus-synchronized responses in auditory cortex . Our model provides several suggestions why this may be the case . First , many previous studies have used anesthetized animals to investigate temporal processing in auditory cortex . Because non-synchronized responses are generated by weak net-excitation , usually by concurrent excitation and inhibition , any anesthesia related decrease in excitation or increase in inhibition would further decrease the neuron’s net excitation , potentially silencing non-synchronized responses [52–54] . Because synchronized neurons have delayed inhibition , they would be less affected by anesthesia , as a large variety of I/E ratios and E strengths will still generate a synchronized response ( Fig . 3 ) . The second issue that may reduce the number of non-synchronized responses observed is related to stimulus optimization . For a given neuron , an optimal stimulus can produce a sustained response , while a non-optimal stimulus will often only produce an onset response [27] , the latter likely resulting from delayed inhibition . As delayed inhibition will most likely generate a synchronized response , rather than non-synchronized , there may be a large bias towards observing synchronized response when the non-optimal sound level or frequency is used . The third issue is the choice of temporally modulated acoustic stimuli used by the experimenter . For example , when using sinusoidal amplitude modulated tones [26 , 55–56] the pulse duration changes with modulation frequency , in contrast to acoustic pulse trains that have fixed pulse durations . While non-synchronized neurons do not respond to single pulses at long IPIs for an acoustic pulse train , this would likely change when the pulse duration is sufficiently lengthened , providing enough excitatory drive for the neuron to spike and resulting in synchronization at low modulation frequencies . Finally , it is important to acknowledge several caveats with our computational model . First , our study directly compares single units from awake marmosets with simulated neurons generated by a neuronal model that is based on intracellular recordings from ketamine-anesthetized rats [24] . Because ketamine is an NMDA antagonist , our model is based primarily on AMPA and GABA-A receptors . Using only synaptic inputs with a short time-constant ( 5 ms ) , approximating AMPA and GABA-A receptors , our model was able to simulate the major types of neural representations ( synchronized , non-synchronized , and mixed ) , as well as two atypical types ( inhibitory and bandpassed ) . We observed that adding NMDA receptors to our model , ( see S6 Fig , e-f ) shifted the synchronization boundary to longer IPIs , and closer to the synchronization limit observed in real synchronizing neurons . It’s important to point out that our model was not able to capture the complete diversity of responses observed in auditory cortex . Most notably , we did not observe unsynchronized responses in the flutter range , which have been previously reported in auditory cortex , albeit more commonly in the rostral and rostrotemporal fields [15–16] . These responses are similar to non-synchronized responses described here , except that the encoding range of IPIs is between 20–125 ms ( perceptual range of flutter ) , rather than 3–25 ms ( perceptual range of fusion ) , and that the relationship between IPI and discharge rate can be either positive or negative monotonic . Given that these responses are more common outside of primary auditory cortex , and have a longer latency response [15] , a two-stage model may be required to generate this type of neural representation . Next , our model was only accurate in representing the neural responses observed in auditory cortex to acoustic pulse trains ( Gaussian windowed tones ) , which have a fixed bandwidth over the complete range of IPIs tested . Other acoustic stimuli , such as sinusoidal amplitude modulated ( sAM ) tones , that change their spectral bandwidth and pulse duration with modulation frequency , cannot be represented accurately by our model , however the addition of new parameters that account for the spectrum of the acoustic stimulus [47] and/or the adaptation of synaptic inputs [42] would likely provide further improvements to our description of temporal processing in auditory cortical neurons . In spite of these caveats , our model provides a parsimonious explanation for the neural coding dichotomy of synchronized and non-synchronized responses observed in auditory cortex . One key implication of these results is that if excitation and inhibition can be independently manipulated , which is possible using molecular genetic techniques such as optogenetics [57] , a neuron’s neural representation can theoretically be switched between synchronized and non-synchronized . In a behaving animal , this provides a powerful paradigm to directly test the auditory percepts generated by temporal and rate-based neural representations .
The electrophysiology data in this report comprised of previous published datasets [15 , 18 , 26] collected at Johns Hopkins University ( Laboratory of Xiaoqin Wang ) . All experimental procedures were approved by the Johns Hopkins University Animal Use and Care Committee and followed US National Institutes of Health guidelines . An integrate-and-fire computational neuronal model was simulated in MATLAB using the following equation , using parameters obtained from Wehr and Zador [24]: Vt+1=−dtC[get ( Vt−Ee ) +git ( Vt−Ei ) +grest ( Vt−Erest ) ]+Vt with C = 0 . 25 nF and grest = 25 nS . Each acoustic pulse was simulated as the summation of 10 excitatory and 10 inhibitory synaptic inputs [24] , each temporally jittered ( Gaussian distribution , σ = 1 ms ) . Each synaptic input was modeled as a time-varying conductance fit to an alpha function: g ( t ) =Ate−at with a time constant 5 ms and an amplitude determined by the excitatory input parameter of the model ( ranging from 0 . 3 to 6 nS ) . A 10 ms delay was added to the synaptic input to simulate the delay between peripheral auditory system and auditory cortex . The reversal potential for the excitatory and inhibitory inputs were 0 mV and -85 mV respectively . A timestep of 0 . 1 ms was used for the simulation . Our model consisted of three parameters: 1 ) I-E delay- the temporal delay between inhibitory and excitatory inputs , 2 ) I/E ratio- the ratio between the magnitude of inhibitory and excitatory inputs , and 3 ) Excitatory input- the magnitude of the excitatory input . The first two parameters ( I-E delay and I/E ratio ) spanned the entire range of values measured intracellularly , previously reported by Wehr and Zador [24] [I-E delay: -2 to 7 ms , I/E ratio: 0 to 2] . Physiologically realistic values for excitation magnitude ( 3rd parameter ) [spanning values of 0 . 3 nS to 6 nS] were determined by calculating the discharge rate of simulated neurons to a pure tone stimulus . For a simulated neuron to be included in our analysis , we required an evoked pure tone response ( greater than 1 spk/s ) without having a response greater than 50 spk/s ( to avoid physiologically uncommon or unrealistic levels of excitation ) . The only change to our model if extending this boundary to higher discharge rates would be the inclusion of more synchronized and mixed response neurons , with even higher evoked discharge rates . Gaussian noise was added to the model using three methods: 1 ) noise added to the time-varying excitatory and inhibitory conductances to simulate random channel fluctuations , 2 ) noise added to the current to simulated background synaptic activity contributing to the spontaneous rate , or 3 ) noise added to the membrane potential based spiking threshold . NMDA channels were added to the model only for S6 Fig ( e-f . ) The time constants used for the alpha function governing the time varying conductance for the NMDA channel was 63 ms ( fast component ) and 200 ms ( slow component ) , with a peak amplitude ratio of 0 . 88:0 . 12 ( fast:slow ) . The ratio of the peak amplitude between the AMPA channel ( time constant of 5 ms ) and NMDA channel was 1:0 . 3 ( AMPA:NMDA ) [29] . Our electrophysiology data in this report comprised of previous published datasets [15 , 18 , 26] . For these datasets , we performed single-unit recordings with high-impedance tungsten microelectrodes ( 2–5 MΩ ) in the auditory cortex of four awake , semi-restrained common marmosets ( Callithrix jacchus ) , a new-world primate species . Action potentials were sorted on-line using a template-matching method ( MSD , Alpha Omega Engineering ) . Experiments were conducted in a double-walled , soundproof chamber ( Industrial Acoustic Co . , Inc . ) with 3-inch acoustic absorption foams covering each inner wall ( Sonex , Illbruck , Inc . ) . Acoustic stimuli were generated digitally ( MATLAB- custom software , Tucker Davis Technologies ) and delivered by a free-field speaker located 1 meter in front of the animal . This physiological data was collected at Johns Hopkins University ( Laboratory of Xiaoqin Wang ) . Recordings were made primarily for the three core fields of auditory cortex ( 177/210 neurons ) - primary auditory cortex ( AI ) , the Rostral field ( R ) , and the Rostrotemporal field ( RT ) , with the remaining neurons recorded from surrounding belt fields . For each single unit isolated , the best frequency ( BF ) and sound level threshold was first measured , using pure tone stimuli that were 200 ms in duration . We next generated a set of acoustic pulse trains , where each pulse was generated by windowing a brief tone at the BF by a Gaussian envelope . Interpulse intervals ( IPIs ) ranged from 3 ms to 75 ms ( 3 , 5 , 7 . 5 , 10 , 12 . 5 , 15 , 20 , 25 , 30 , 35 , 40 , 45 , 50 , 55 , 60 , 65 , 70 , 75 ms ) . Acoustic pulse train stimuli were 500 ms in duration , and all intertrial intervals were at least 1 s long . Each stimulus was presented in a randomly shuffled order with other stimuli , and repeated at least five times for all neurons , and at least ten times for about 55% of neurons ( 115/210 ) . Stimulus intensity levels for acoustic pulse trains were generally 10–30 dB above BF-tone thresholds for neurons with monotonic rate-level functions and at the preferred sound level for neurons with non-monotonic rate-level functions . | How does our auditory system represent time within a sound ? Previous work has demonstrated that both the firing rate of neurons ( rate code ) and the timing of their stimulus-evoked responses ( temporal code ) can be used by auditory cortical neurons to represent temporal information . We investigated the underlying mechanisms of these two neural representations using a computational model of a cortical neuron . We found that the timing and magnitude of inhibition determined whether neurons responded to an acoustic stimulus using a rate or temporal code . Our model predicts several differences in the response pattern of neurons using rate and temporal representations , which we next validated with electrophysiological data recorded from the auditory cortex of non-human primates . Together these data suggest that feedforward inhibition provides a parsimonious explanation of how rate and temporal representations are generated in auditory cortex . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [] | 2015 | The Role of Inhibition in a Computational Model of an Auditory Cortical Neuron during the Encoding of Temporal Information |
Type IV secretion systems ( T4SS ) are used by Gram-negative bacteria to translocate protein and DNA substrates across the cell envelope and into target cells . Translocation across the outer membrane is achieved via a ringed tetradecameric outer membrane complex made up of a small VirB7 lipoprotein ( normally 30 to 45 residues in the mature form ) and the C-terminal domains of the VirB9 and VirB10 subunits . Several species from the genera of Xanthomonas phytopathogens possess an uncharacterized type IV secretion system with some distinguishing features , one of which is an unusually large VirB7 subunit ( 118 residues in the mature form ) . Here , we report the NMR and 1 . 0 Å X-ray structures of the VirB7 subunit from Xanthomonas citri subsp . citri ( VirB7XAC2622 ) and its interaction with VirB9 . NMR solution studies show that residues 27–41 of the disordered flexible N-terminal region of VirB7XAC2622 interact specifically with the VirB9 C-terminal domain , resulting in a significant reduction in the conformational freedom of both regions . VirB7XAC2622 has a unique C-terminal domain whose topology is strikingly similar to that of N0 domains found in proteins from different systems involved in transport across the bacterial outer membrane . We show that VirB7XAC2622 oligomerizes through interactions involving conserved residues in the N0 domain and residues 42–49 within the flexible N-terminal region and that these homotropic interactions can persist in the presence of heterotropic interactions with VirB9 . Finally , we propose that VirB7XAC2622 oligomerization is compatible with the core complex structure in a manner such that the N0 domains form an extra layer on the perimeter of the tetradecameric ring .
Bacteria employ multiprotein secretion systems to translocate effector proteins or nucleoprotein complexes across the cell envelope where they modulate the bacterium's interactions with the environment . In Gram-negative bacteria , the secretion systems have been classified into 6 different types [1] , [2] . The type IV secretion systems ( T4SSs ) are ancestrally related to bacterial conjugation machines [3] and are able to translocate proteins and/or protein-DNA complexes to the extracellular milieu or the host interior , in many cases contributing to the ability of the bacterial pathogen to colonize the host and evade its immune system [4] . T4SSs are important in the pathogenic mechanism of many microbes , including the animal pathogens Legionella pneumophila ( Legionnaires' disease; [5] ) , Bordetella pertussis ( whooping cough; [6] ) , Coxiella burnetii ( Q fever; [7] ) , Bartonella henselae ( cat scratch disease , [8] ) , Brucella spp . ( brucellosis; [9] ) and Helicobacter pylori ( gastritis , ulcers , stomach cancer; [10] ) , as well as the plant pathogen Agrobacterium tumefaciens ( crown gall disease; [11] ) which has the prototype T4SS , composed of 12 proteins , VirB1-VirB11 and VirD4 [12] . Xanthomonas citri subsp . citri ( formerly Xanthomonas axonopodis pv . citri ) ( Xac ) is a gammaproteobacterial phytopathogen which causes citrus canker , a disease that affects all citrus plants and has significant economic impact worldwide [13] . The Xac genome was sequenced and many potential virulence-related genes were identified [14] , including an 18 kb chromosomal vir locus that codes for a T4SS [13] , [14] , [15] . Orthologous T4SSs have been identified in the closely related species Xanthomonas campestris pv . campestris ( strains ATCC 33913 [14] , 8004 [16] and B100 [17] ) , Xanthomonas albilineans [18] , Xanthomonas campestris pv . vasculorum NCPPB702 [19] , Xanthomonas campestris pv . musacearum NCPPB4381 [19] and Stenotrophomonas maltophilia ( strains K279A [20] and R551-3 [21] ) . The vir locus is incomplete in Xanthomonas campestris pv . vesicatoria [22] and is absent in four Xanthomonas oryzae strains ( KACC10331 [23] , MAFF311018 [24] , PXO99A [25] and BLS256 ( GenBank database AAQN00000000 ) ) and in two Xanthomonas fuscans subsp . aurantifolii lineages ( 10535 and 11122 [26] ) . The unrelated Gram-negative bacteria Neisseria flavescens SK114 ( GenBank database ACQV00000000 ) and Neisseria sp . oral taxon 014 str . F0314 ( GenBank database ADEA00000000 ) also present a locus highly similar to the T4SS loci found in the Xanthomonadaceae family . Some Xanthomonas genomes , including Xac , carry megaplasmids that code for a second T4SS probably involved in plasmid mobilization and whose structural components exhibit only very low sequence identity to their counterparts in the chromosomally encoded T4SS under study [15] , [27] . One particularly distinctive feature of the Xanthomonad T4SSs is related to the VirB7 component . In well-characterized T4SSs , VirB7 is a lipoprotein attached to the periplasmic side of the outer membrane [28] . It has been shown to interact with several T4SS subunits [29] , including itself [30] and the VirB9 C-terminal region [31] , [32] . The structure of a VirB7-VirB9-VirB10 complex ( TraN-TraO-TraF ) from the T4SS of the conjugative plasmid pKM101 demonstrated that these proteins form a hetero-tetradecameric outer membrane channel through which the substrates pass [33] , [34] . Structures of the complex formed between the TraO C-terminal domain and TraN revealed that the 33-residue TraN forms an extended structure that winds around the TraO β-sandwich [34] , [35] . Members of the VirB7 family are typically 45–65 residues long [36] , becoming 15–20 residues shorter after removal of the N-terminal signal sequence and covalent attachment to lipid molecules . In Xac , several lines of evidence indicate that the gene xac2622 codes for a VirB7 equivalent; specifically , the gene position , the presence of signals for outer membrane localization and lipidation and the interaction of its product with VirB9 [15] . However , XAC2622 does not exhibit any sequence similarity with the VirB7 family and it is much larger than a typical VirB7 protein ( 139 and 118 residues before and after signal sequence removal , respectively ) . In order to understand the unique structural features of Xanthomonad VirB7 , we have solved the solution nuclear magnetic resonance ( NMR ) and X-ray crystal structures of XAC2622 , which we now call VirB7XAC2622 , and studied its interaction with VirB9 . VirB7XAC2622 presents all of the characteristics of VirB7: predicted signal peptide and lipobox sequences followed by a short and extended region that contains the VirB9 binding site . However , unlike other VirB7 proteins , VirB7XAC2622 has an extra domain whose topology is strikingly similar to that of N0 domains found in proteins from different multi-subunit complexes involved in transport across the bacterial outer membrane . We show that VirB7XAC2622 oligomerizes through interactions involving conserved residues in the folded domain and the unfolded N-terminus and that oligomerization and the formation of the VirB7XAC2622-VirB9 complex can occur simultaneously . We propose that these interactions are compatible with the tetradecameric core complex structure , resulting in the formation of an extra ring layer .
The construct used for NMR structure determination corresponds to residues 24–139 of VirB7XAC2622 preceded by a Gly-Ser-His-Met sequence . Bioinformatics analysis predicts that residues 1–21 would be removed and Cys22 lipidated , leading to anchoring in the bacterial outer membrane [15] . Approximately 98% of main chain and aliphatic side chain and 93% of aromatic side chain 1H , 13C and 15N resonances were assigned ( Table 1 and Figure S1 ) . A total of 2176 experimental NMR restraints , consisting of 2056 interproton distances and 120 torsion angles , were used in the structure calculation ( Table 1 ) . A small set of nuclear Overhauser effect ( NOE ) cross-peaks that were not assigned by CYANA [37] were subsequently identified as resulting from intermolecular contacts ( see below ) . The solution structure of VirB7XAC2622_24–139 consists of a globular domain ( residues 52–133 ) , flanked by a long disordered N-terminus ( amino acids 24–51 ) and a short flexible C-terminus ( residues 134–139 ) ( Figure 1A ) . The globular domain is composed of two α-helices sandwiched between a mixed three stranded β-sheet on one side and an antiparallel two-stranded β-sheet plus a short 310 helix on the other ( Figure 1B ) . The NMR structures were independently validated by 1H-15N residual dipolar couplings ( RDCs ) . The average Q factor obtained by fitting RDCs from secondary structure elements to the whole NMR ensemble is 0 . 271±0 . 019 , consistent with a good quality structure [38] , [39] , [40] . To better characterize the globular domain , we cloned and purified a recombinant fragment encompassing residues 51–134 ( VirB7XAC2622_51–134 ) which was submitted to crystallization trials that resulted in large plates ( Figure S2 ) which belong to space group C2221 and diffracted up to 1 . 0 Å . Molecular replacement was performed using the solution structure of the VirB7XAC2622 globular domain as the search model . Details of the data collection and structure refinement are listed in Table 2 and the model is presented in Figure 1C . There was clear electron density for all residues , with the exception of the first amino acid ( threonine 51 ) and residues 122 and 123 in the loop between β4 and β5 . As expected , the crystal structure is very similar to the NMR ensemble ( Figure 1D ) , displaying an average root mean square deviation ( RMSD ) for the backbone heavy atoms and all heavy atoms of 1 . 05±0 . 05 Å and 1 . 61±0 . 06 Å for residues 53–130 , and excellent agreement with the backbone 1H-15N RDCs [Q factor of 0 . 187 ( regular secondary structures only ) and 0 . 272 ( all residues ) ] . Changes in 15N heteronuclear single-quantum coherence ( 15N-HSQC ) cross-peak positions as a function of protein concentration indicated that VirB7XAC2622 oligomerizes in fast exchange on the NMR time scale ( Figure 2A ) . Consistent with this observation , alterations in 15N T1 and T2 relaxation times indicated that the overall tumbling slows significantly at higher protein concentrations ( see below ) . Chemical shift perturbation data showed that two regions are involved in VirB7XAC2622 self-interactions: a region in the unfolded N-terminus ( residues 42–49 ) and a patch on the surface of the globular domain made up of residues 63–65 , 85–93 , 111–119 and 131 ( Figure 2B and C ) . Two possible interaction schemes can be envisaged that would be compatible with the chemical shift perturbation data: i ) a side-by-side arrangement in which the N-terminal unfolded segment and the globular domain of one molecule interact , respectively , with the same regions of a second molecule , or ii ) a head-to-tail complex arrangement where the N-terminus of one molecule recognizes the folded domain of the other . In order to test the hypothesis of the formation of a head to tail complex , we performed a titration of the 15N-VirB7XAC2622_51–134 globular domain with an unlabeled peptide encompassing residues 38–52 from the N-terminal region ( VirB7XAC2622_38–52 ) . During the titration , we observed changes in the same cross-peaks which were perturbed in the entire protein in a concentration dependent manner ( Figure S3 ) . This hypothesis was corroborated by the analysis of a set of 13 NOEs not assigned by CYANA [37] , which showed that they could be accounted for by the formation of head-to-tail dimers . Indeed , those NOEs were observed between proton pairs derived from amino acids whose main-chain 1H-15N chemical shifts are perturbed in a concentration dependent manner ( the NOEs are listed in the Materials and Methods ) . We then used the set of 13 intermolecular NOEs as geometric restraints to drive computational docking simulations of the VirB7XAC2622 dimer using HADDOCK2 . 0 [41] . In one simulation , a fragment corresponding to the disordered N-terminal region ( residues 24–50 ) was docked with a fragment corresponding to the globular domain ( residues 51–139 ) while the second simulation involved the docking between two full length proteins ( residues 24–139 ) . The resulting models from both simulations were highly similar . In the second simulation three of the thirteen NOEs were violated in all solutions by approximately 1 . 4 to 3 Å . They correspond to weak peaks and could have contributions of spin diffusion or chemical exchange . These docking simulations were therefore able to determine the general nature of the oligomerization interface whose resolution is necessarily limited by the small number of restraints and by the absence of structural information for the N-terminal region . One cluster of docking solutions using full length VirB7XAC2622 is shown in Figure 2D . To investigate whether VirB7XAC2622_24–139 forms dimers or higher-order oligomers , glutaraldehyde cross-linking experiments were performed . These assays showed that VirB7XAC2622_24–139 forms dimers , trimers and higher order oligomers ( Figure S4A ) . When the cross-linking experiment was performed in the presence of 1% SDS , no covalently cross-linked oligomers were observed , indicating that oligomerization requires the presence of a correctly folded protein ( Figure S4A ) . As expected , no cross-links were observed when the experiment was performed using VirB7XAC2622_51–134 which lacks the disordered N-terminal region ( Figure S4B ) . To analyze this interaction , VirB7XAC2622_24–139_His was titrated into a solution of VirB9XAC2620_34–255 ( full-length VirB9XAC2620 minus the first 33 residues including a predicted signal peptide ) and changes in intrinsic fluorescence were monitored . A monotonic increase in fluorescence emission was detected until saturation at a 1∶1 VirB7∶VirB9 stoichiometry ( Figure S5 ) . A dissociation constant of 4×10−8 M for the complex was estimated by fitting the observed fluorescence changes to a binding isotherm ( equation 2 , Materials and Methods ) . To study this interaction by NMR , the 15N-HSQC spectrum of 15N-VirB7XAC2622_24–139 was acquired in the presence of unlabeled VirB9XAC2620_34–255 . At a 1∶1 molar ratio , essentially no cross peaks were detected , except for those derived from the flexible N- and C-terminal residues ( amino acids 24–25 and 134–139; data not shown ) . This observation is consistent with the formation of a tight complex with long correlation time , with the majority of cross peaks beyond detection . VirB7XAC2622 was previously shown to interact with the C-terminal domain of VirB9XAC2620 [15] as has been shown for other VirB7-VirB9 homologs [32] , [35] . We therefore produced a fragment , VirB9XAC2620_154–255 , corresponding to the C-terminal domain of VirB9XAC2620 . Changes in the 15N-HSQC spectrum of VirB7XAC2622_24–139 upon adding VirB9XAC2620_154–255 showed that the binding occurs in slow exchange on the NMR time scale , consistent with the sub-micromolar affinity detected by fluorescence spectroscopy for the interaction with full length VirB9 ( VirB9XAC2620_34–255 ) . In order to map the VirB9 binding site on the structure of VirB7 , we assigned the backbone resonances of VirB7XAC2622_24–139 in complex with VirB9XAC2620_154–255 and analyzed the chemical shift differences ( Figure 3A and B ) . This analysis showed that only residues 27–41 , within the disordered VirB7 N-terminus , undergo significant chemical shift perturbations ( Figure 3C and D ) . This region is adjacent to , but does not overlap , the N-terminal region involved in VirB7XAC2622 oligomerization ( residues 42–49 ) . 1H-15N RDCs and backbone chemical shifts for VirB7XAC2622_24–139 alone and in the presence of VirB9XAC2620_154–255 are essentially identical , with the exception of the N-terminal region that is involved in binding to VirB9 ( Figure 3 and Figure S6 ) . Thus , the folded domain does not participate in the recognition of the VirB9XAC2620 C-terminal domain and the linker between the VirB9 interaction site and the VirB7 globular domain is flexible enough to permit the two regions to align independently in the alignment medium . Interestingly , VirB7XAC2622_24–139 is able to simultaneously oligomerize and interact with VirB9XAC2620_154–255 , as the cross-peaks related to the regions involved in VirB7 oligomerization shift even in the presence of VirB9XAC2620_154–255 ( Figure S7 ) . The 15N-HSQC spectrum of VirB9XAC2620_154–255 showed characteristics of poor line shape and chemical shift dispersion ( Figure 4; red ) suggestive of conformational disorder and a probable lack of stable tertiary structure . However , when unlabeled VirB7XAC2622_24–139 was added to the solution containing 15N-VirB9XAC2620_154–255 , the 1H-15N cross peaks became more uniform in shape and more highly resolved ( Figure 4; green ) . These observations suggest that VirB9XAC2620_154–255 on its own is in a dynamic conformational equilibrium or not properly folded in the absence of VirB7XAC2622_24–139 . Nevertheless , it undergoes a significant conformational change which greatly reduces its conformational flexibility when bound to the latter protein . In order to further investigate whether the interaction of VirB9XAC2620_154–255 with the N-terminal domain of VirB7XAC2622 alone is sufficient to drive the formation of a specific VirB7-VirB9 complex , we studied the interaction of 15N-labeled VirB9 with an unlabeled peptide derived from the VirB7XAC2622 N-terminus ( residues 24–46 ) , which includes the VirB7XAC2622 N-terminal segment that interacts with VirB9XAC2620 ( amino acids 27–41 ) . Indeed , the perturbations observed in the 15N-HSQC spectra of VirB9XAC2620_154–255 in the presence of the full-length VirB7XAC2622_24–139 or the VirB7XAC2622_24–46 peptide are essentially the same ( Figure 4 and Figure S8 ) , demonstrating that the VirB7 N-terminal region is sufficient to interact with the VirB9XAC2620 C-terminal domain . To investigate whether aggregation could be responsible for the line broadening observed in the 15N-HSQC spectrum of VirB9 C-terminal domain , we calculated the approximate overall rotational correlation time ( τc ) from estimates of 15N T1 and T2 obtained using 1D versions of the 15N relaxation experiments . The values of τc of VirB9XAC2610_154–255 alone or in complex with the VirB7XAC2622_24–46 peptide are approximately 6 . 5 ns ( data not shown ) , which suggest that the VirB9XAC2620 C-terminal domain is monomeric in both conditions . Measurements of 15N relaxation times ( T1 and T2 ) and heteronuclear {1H}-15N NOE were performed for 15N-VirB7XAC2622_24–139 alone ( at 100 and 800 µM ) and in complex with unlabeled VirB9XAC2620_154–255 . The relaxation data are consistent with the presence of a flexible N-terminal tail and a more rigid globular domain ( Figure 5 ) . It is worth noting that the VirB9 binding surface of VirB7XAC2622_24–139 ( residues 27–41 ) becomes less flexible upon binding to VirB9 ( Figure 5A and C ) . In the absence of VirB9 , residues 29–42 display {1H}-15N heteronuclear NOE values close to zero , indicating that this segment is not fully flexible and may present transient structures poised to interact with VirB9 . Changes in 15N-T1 and 15N-T2 observed upon raising the VirB7XAC2622_24–139 concentration from 100 to 800 µM indicate a significant decrease in the overall rotational correlation rate , consistent with the formation of oligomers ( Figure 5B and C ) . Relaxation parameters for residues involved in oligomerization ( 43–47 and some between positions 86 and 93 ) were not obtained because they were not detectable in the 15N-HSQC spectrum at high protein concentration ( 800 µM ) . Although these cross peaks become visible upon diluting the sample , they display too low amplitude to allow precise measurements . From the 15N T1 and T2 data , the overall rotational correlation times of VirB7XAC2622 at the concentrations of 100 and 800 µM were estimated to be 5 . 8 and 12 . 6 ns , respectively . These values are consistent with the predominance of a monomer at lower concentration and a dimer at higher concentration . This observation , however , does not exclude the possibility of the existence of an equilibrium which includes higher-order oligomers , as shown by glutaraldehyde cross-linking data . In Agrobacterium tumefaciens cells , the VirB7 protein contributes to VirB9 and VirB10 stability [42] . We therefore produced a xac2622 gene knockout strain and performed immunoblot experiments to assess the expression of VirB7XAC2622 , VirB9XAC2620 and VirB10XAC2619 . The production of all three proteins could be detected in wild type cells but none could be detected in the ΔvirB7XAC2622 strain ( Figure 6A ) . To evaluate if the absence of VirB9XAC2620 and VirB10XAC2619 in the mutant strain is due to the lack of the VirB7XAC2622 production or to a polar effect in the T4SS operon , ΔvirB7XAC2622 cells were complemented with a plasmid encoding the VirB7XAC2622 protein in trans . The expression of VirB7XAC2622 from the pUFR-VirB7 plasmid restored VirB9XAC2620 and VirB10XAC2619 protein levels ( Figure 6A ) . Furthermore , no significant differences in virB9XAC2620 transcript levels were detected between the wild-type , ΔvirB7XAC2622 and ΔvirB7XAC2622+pUFR-VirB7 strains in quantitative RT-PCR experiments ( data not shown ) . These results show that VirB7XAC2622 is necessary for the stability of VirB9XAC2620 and VirB10XAC2619 proteins in vivo . The VirB7 , VirB9 and VirB10 orthologs TraN , TraO and TraF form a stable trimeric complex [33] , [34] . In order to determine whether the VirB7XAC2622 , VirB9XAC2620 and VirB10XAC2619 form a stable complex in Xac cells , we performed immunoprecipitation experiments using antisera for each of the three proteins . Results described in Figure 6B show that VirB7XAC2622 , VirB9XAC2620 and VirB10XAC2619 are present in the material immunoprecipitated by each of the three antisera in wild-type cells . Furthermore , none of the three proteins were immunoprecipitated from ΔvirB7XAC2622 cells while all three proteins were detected in immunoprecipitation experiments using ΔvirB7XAC2622+pUFR-VirB7 cells ( Figure 6B ) . Altogether , these results show that a complex between VirB7XAC2622 , VirB9XAC2620 and VirB10XAC2619 is formed in vivo in Xac cells . In order to test the role of VirB7XAC2622 in Xac virulence , sweet orange leaves were infiltrated with the wild type and ΔvirB7XAC2622 strains . Both strains presented essentially the same infection phenotypes , including water-soaking , hyperplasia and necrosis ( Figure 6C ) [13] . In planta growth curves of the wild type and ΔvirB7XAC2622 cells revealed that the two strains replicate at a similar rate ( Figure 6D ) . These data indicate that the VirB7XAC2622 protein does not participate in the Xac's ability to induce canker symptoms in orange leaves under the experimental conditions tested . Similarly , a T4SS knockout failed to affect the pathogenicity of X . campestris pv . campestris on several host plants [43] . Xac pathogenicity in citrus and canker symptoms are strongly dependent on the action of virulence factors secreted by the Type III secretion system [13] . The T4SS is probably involved in other , as yet unidentified , cellular functions . As noted in the Introduction , the Xac pXAC64 megaplasmid codes for a second T4SS probably involved in plasmid mobilization [15] , [27] and whose structural components exhibit only very low sequence identity to their counterparts in the chromosomally encoded T4SS under study . For example , its VirB9 and VirB10 homologs ( XACb0039 and XACb0038 ) are only 23–24% identical to VirB9XAC2620 and VirB10XAC2619 [15] . Furthermore , the pXAC64 plasmid does not code for any proteins with similarity to VirB7XAC2622 or any other known VirB7 proteins [15] . Therefore , it is unlikely that the two Xac T4SSs exercise redundant functions .
VirB7XAC2622 is structurally distinct from all VirB7 homologs characterized to date . Its structural motifs are illustrated in Figure 7A . Like all other VirB7 proteins , it has a signal peptide , a conserved cysteine residue ( Cys22 ) within a lipobox for covalent attachment to outer membrane lipids and a short extended region involved in interaction with VirB9 . These motifs are all contained within the first 41 residues , which corresponds relatively well with the size of the majority of unprocessed VirB7 proteins ( 45–65 amino acids ) [36] . VirB7 proteins are very poorly conserved though it has been observed that many proteins of this family have in common a PVNK motif that is involved in the interaction with VirB9 [35] . The VirB9 binding site in VirB7XAC2622 includes a HVNH sequence ( residues 36–39 ) found in almost all Xanthomonadaceae orthologs ( PVNR in X . albilineans and Stenotrophomonas maltophilia ) which aligns with a perfect PVNK motif in the Neisseria sp . oral taxon 014 str . F0314 ortholog ( Figure 7B ) . The side chains of the two most conserved residues of this motif , Val and Asn , make intimate contacts with TraO/VirB9 in the pKM101 outer membrane complex structure [35]: the Val side chain is inserted between the two TraO β-sheets while the Asn amide Oδ1 and Nδ2 atoms makes a set of three H-bonds with the TraO main chain . Furthermore , the PVNK motif in TraN is preceded by an approximately 10 amino acid segment that makes contacts with TraO , including a 3 residue β-strand that adds to the edge of one of the TraO β-sheets [34] . The corresponding regions in VirB7XAC2622 are precisely those that demonstrate the greatest chemical shift perturbations upon binding to VirB9XAC2620 ( Figure 3 ) . This suggests that interactions between VirB7XAC2622 and VirB9XAC2620 are very similar to that observed in the TraN-TraO complex . The VirB7-VirB9 interactions described are most likely responsible for the loss of conformational flexibility observed for the N-terminal tail of VirB7XAC2622 and the C-terminal domain of VirB9XAC2620 . Furthermore , interaction with VirB7XAC2622 is likely to be accompanied by a significant change in the structure of VirB9XAC2620_154–255 , as revealed by the very large chemical shift perturbations observed for the majority of the 1H-15N cross-peaks ( Figure 4 ) . These changes in VirB9XAC2620 structure and dynamics may be related to its instability in vivo in the absence of VirB7XAC2622 . Our immunoblot and coimmunoprecipitation data suggest that VirB7XAC2622 , VirB9XAC2620 and VirB10XAC2619 form a stable trimeric complex in vivo and that the stabilities of VirB9XAC2620 and VirB10XAC2619 are dependent on VirB7XAC2622 ( Figure 6 ) . This may be a physiologically relevant mechanism by which excess VirB9 and VirB10 are degraded to maintain proper VirB7-VirB9-VirB10 stoichiometry . VirB7XAC2622 has a mosaic structure ( Figure 7A ) . In addition to the canonical VirB7-like N-terminal region , VirB7XAC2622 has a C-terminal region that includes a globular domain ( residues 52–133 ) . This globular domain does not interact with VirB9XAC2620_154–255 but does interact with an extended region ( residues 42–49 ) that immediately precedes the globular domain in another VirB7XAC2622 molecule ( Figure 2D ) , leading to the formation of homo-oligomers . The unique structural features of VirB7XAC2622 raise the question of how they may be incorporated into the model for the T4SS outer membrane complex tetradecamer previously solved for the pKM101 conjugation machine [34] . Specifically: can VirB7XAC2622 oligomerization occur while maintaining VirB7-VirB9 interactions in the context of the pore ? And how are the VirB7XAC2622 globular domains oriented with respect to the pore ? In order to investigate these questions , we used the VirB7XAC2622-TraN alignment depicted in Figure 7B to manually place 14 different NMR models of residues 37–139 of VirB7XAC2622 ( VirB7XAC2622_37–139 ) around a tetradecameric ring made up of TraOCT subunits derived from the pKM101 TraN-TraOCT-TraFCT structure [34] . The VirB7XAC2622_37–139 subunits were placed so that Val37 and Asn38 occupied positions similar to Val33 and Asn34 of TraN . This ( VirB7XAC2622_37–139-TraOCT ) 14 complex was used as a starting model in a molecular dynamics simulation in which the following restraints were used: i ) The 14 TraOCT subunits were fixed in space , ii ) VirB7XAC2622 residues 37 and 38 were fixed in space and iii ) intermolecular NOE derived distance restraints were placed on 5 hydrogen pairs between neighboring VirB7XAC2622 subunits . During the initial cycles of the molecular dynamics simulation , the distance observed between the 13 intermolecular VirB7XAC2622 hydrogen pairs for which NOEs were measured ( Figure 2D and Results above ) decreases from the 12–25 Å range observed in the starting model to a 4–7 Å range , consistent with observation of NOE signals . These short contact distances are then observed throughout the simulation for all 14 VirB7XAC2622 pair interfaces made up of residues A43 , T45 , E46 , I47 , L49 in one VirB7XAC2622 molecule and residues T63 , S85 , D86 , Y87 , T88 , I90 in a neighboring VirB7XAC2622 . This is a consequence of the strong distance restraints included in the potential energy function . The final molecular dynamics configuration is shown in Figure 7C . Adding models of TraN residues 19–32 to represent VirB7XAC2622 residues 22–36 ( see alignment in Figure 7B ) and TraFCT ( to represent VirB10XAC2619 ) results in the structure shown in Figure 7D made up of 14 repeats of VirB7XAC2622_37–139 , TraN19–32 , TraOCT and TraFCT . This can be seen as putative model for the VirB7XAC2622-VirB9XAC2620-VirB10XAC2619 outer membrane complex . Note that in Figures 7C and 7D , the VirB7 N0 domains ( dark blue ) adopt a variety of orientations with respect to the central ring and the membrane normal . This flexibility is derived from the conformational freedom of the VirB7XAC2622 regions immediately before and after residues 42–49 ( magenta ) involved in VirB7-VirB7 interactions . Our proposal of a physiological role for VirB7XAC2622 oligomerization is supported by an analysis of conserved residues in the small group of full-length VirB7XAC2622 homologs found in Xanthomonadaceae and the two Neisseria species Neisseria flavescens SK114 and Neisseria sp . oral taxon 014 str . F0314 . An alignment of VirB7XAC2622 and the two Neisseria proteins is shown in Figure 7B . The three proteins have only 21% sequence identity . However , 6 out of 11 of the residues involved in intermolecular NOEs at the B7-B7 interface ( E46 , I47 , L49 , T63 , D86 and T88 ) are absolutely conserved in all Xanthomonadaceae , Neisseria str . F0314 and Neisseria flavescens and is either Phe or Tyr at position 87 . Thus , the oligomerization interface is well conserved in an otherwise highly diverse protein family . VirB7XAC2622 oligomerization is different from the dimerization observed in the canonical VirB7 protein from A . tumefaciens [30] which occurs via the formation of disulfide bonds involving residue 24 . The structure of the outer membrane complex [34] however , is not compatible with the maintenance of these disulfide bonds in a fully assembled T4SS . Furthermore , VirB7XAC2622 does not have any cysteine residues other than at the N-terminal lipidation site . The VirB7XAC2622 globular domain has no significant sequence similarity to proteins of known structure . We therefore used the Dali Server [44] to search for proteins with similar topology to this domain . This analysis revealed that the VirB7XAC2622 folded region resembles domains found in the following proteins ( Table 3 and Figure 8 ) : i ) the TonB-dependent receptors ( periplasmic signaling domain; Figure 8A ) [45] , [46] , ii ) the outer membrane secretin channel GspD from the type II secretion system ( T2SS; Figure 8B ) [47] , iii ) the secretin EscC from the Type III secretion system ( T3SS; Figure 8C ) [48] , iv ) the needle-like cell-puncturing device components gp27 and gp44 from long-tailed phages like T4 and Mu ( Figure 8D ) [49] , [50] and v ) the Type VI secretion system ( T6SS ) protein VgrG ( superposition not shown ) [51] . These domains are found at the N-termini of T2SS and T3SS secretins and have been denominated N0 domains which have also been identified in the Type IV pilus secretin PilQ [47] and the filamentous phage secretin pIV [52] . It is striking that in spite of its small and compact nature , domains with the VirB7XAC2622 topology are found only in a restricted number of proteins , all of which are involved in the transport of molecules across bacterial outer membranes . These observations suggest that all of these proteins may be distantly related and have evolved in the periplasm or outer membrane to adopt a variety of functions , from structural modules in outer membrane pores ( secretins from type II and type III secretion systems , type IV pili and filamentous phages ) to membrane-penetrating devices in T6SS and long-tailed bacteriophages , and signal-transduction modules in TonB-dependent receptors . Nakano et al . [53] have recently described the crystal structure of the DotD lipoprotein of the Type IV-B secretion system from Legionella pneumophila . Type IV-B secretion systems have weak or no sequence similarity with the larger group of Type IV-A secretion systems [54] to which the Xanthomonad T4SS pertains . Interestingly , the DotD crystal structure presents a C-terminal region with an N0 domain topology similar to that of VirB7XAC2622 structure ( Figure 8E ) , though the proteins present no significant sequence similarity . An interesting difference between the two structures is that a 6 amino acid sequence ( called the “lid” ) of the otherwise disordered N-terminal region is visible in the crystal structure and makes β-strand addition to β1 . This interaction is different from that observed in the VirB7XAC2622 oligomer described here . Since no electron density for the 23 residues between the lid and the N0 domain were visible , it was not clear whether they come from the same molecule in the crystal lattice . We therefore predict that DotD will eventually be shown to exhibit the same characteristics as VirB7XAC2622 , namely to interact with both a VirB9-like protein and to form head-to-tail polymeric rings based on interactions between the flexible N-terminal region and globular N0 domains of neighboring subunits . This analysis points to previously unrecognized common structural features between the outer membrane complexes of Type IV-A and Type IV-B secretion systems . The novel VirB7XAC2622 structure , its oligomerization and its interaction with VirB9 described here point to a possible structural variation in the Xanthomonas T4SS that could result in the formation of an extra ring layer in the core complex . While the VirB7-VirB7 interactions observed in the NMR experiments were of relatively low affinity ( fast exchange chemical shift changes were observed in the 7–850 µM range ) , these interactions would be strengthened by the increased effective concentration afforded by the fourteen neighboring VirB7XAC2622 binding sites on the outer surface of the VirB9/VirB10 ring . Several hypotheses regarding the physiological role of this VirB7XAC2622 ring can be proposed . One possibility is that the weak B7-B7 interactions break and reform again to accommodate different conformational states of the outer pore during substrate translocation or to permit allosteric communication between the outer and inner membrane core complexes . Another possibility is that the VirB7XAC2622 N0 domains could act as a conduit for signal transmission between substrates or signaling molecules in the periplasm and the VirB9 and VirB10 subunits . Whether the VirB7XAC2622 N0 domain carries out a purely structural function or is involved in substrate recognition or signal transduction will have to be tested in the future .
Plasmids , oligonucleotides and bacterial strains used in this study are listed in Table S1 . The DNA fragments encoding residues 24–139 of VirB7XAC2622 ( GenBank AAM37471 ) , 154–255 of VirB9XAC2620 ( GenBank AAM37469 ) and 85–389 of VirB10XAC2619 ( GenBank AAM37468 ) were amplified by PCR from Xanthomonas citri subsp . citri strain 306 genomic DNA and inserted into the pET28a expression vector ( Novagen ) using the NdeI and XhoI ( VirB7XAC2622 ) or NdeI and BamHI ( VirB9XAC2620 and VirB10XAC2619 ) cloning sites , to express the recombinant proteins fused with an N-terminal His-tag . DNA fragments encoding residues 51–134 of VirB7XAC2622 and 34–255 of VirB9XAC2620 were cloned into the pET11a vector ( Novagen ) using NdeI and BamHI sites . All constructs were confirmed by DNA sequencing . E . coli BL21 ( DE3 ) RP and BL21 ( DE3 ) Star strains ( Novagen ) were transformed with the recombinant plasmids for expression of the protein products which are referred to as VirB7XAC2622_24–139_His , VirB9XAC2620_154–255_His , VirB7XAC2622_51–134 , VirB9XAC2620_34–255 and VirB10XAC2619_85–389_His . Cells were grown in 2XTY for expression of unlabeled proteins or in M9 minimal media containing 15NH4Cl and 12C-glucose or 13C-glucose ( Cambridge Isotope Laboratories ) for production of 15N or 15N and 13C isotopically labeled proteins . Recombinant protein expression was induced at midlog phase by the addition of 0 . 5 mM isopropyl β-D-1-thiogalactopyranoside and the cells were grown for four hours at 37°C . Cells were then harvested and lysed by French Press . In the case of VirB7XAC2622_24–139_His and VirB9XAC2620_154–255_His , the lysis supernatants were separated by passage through Q-Sepharose ( VirB7XAC2622_24–139_His ) or SP-Sepharose ( VirB9XAC2620_154–255_His ) ion exchange columns ( GE Healthcare ) followed by affinity chromatography using a nickel affinity column . The N-terminal His-tags were removed using the Thrombin CleanCleave kit ( Sigma ) , creating the proteins VirB7XAC2622_24–139 and VirB9XAC2620_154–255 . A final purification step consisted of passage through a Superdex 75 26/60 size exclusion column ( GE Healthcare ) . VirB7XAC2622_51–134 was purified using Q-sepharose anion exchange and Superdex 75 gel filtration chromatography . VirB9XAC2620_34–255 and VirB10XAC2619_85–389_His are expressed as insoluble proteins present in inclusion bodies . After lysis , VirB9XAC2620_34–255 was solubilized in the presence of 8 M urea and purified by SP-Sepharose cation exchange and Superdex 75 size exclusion chromatographies using buffers containing 8 M urea . VirB10XAC2619_85–389_His was purified by affinity chromatography using buffers containing 8 M urea . Purified proteins were refolded by successive dialysis against decreasing concentrations of urea in 10 mM sodium acetate ( pH 5 . 0 ) for VirB9XAC2620_34–255 or 10 mM Tris-HCl ( pH 8 . 0 ) , 150 mM NaCl for VirB10XAC2619_85–389_His . Synthetic peptides corresponding to VirB7XAC2622_24–46 and VirB7XAC2622_38–52 were purchased from Biomatik Corporation ( Cambridge , Ontario , Canada ) and EZBiolab ( Carmel , Indiana , USA ) , respectively . NMR samples containing 15N or 15N/13C labeled protein were prepared in 10 mM 2H-sodium acetate ( pH 5 . 0 ) , containing 50 mM sodium chloride , 0 . 05% ( w/v ) sodium azide , 7% ( v/v ) 2H2O and 1 mM trimethylsilyl-2 , 2 , 3 , 3-tetradeuteropropionic acid as internal standard for 1H chemical shift referencing . NMR spectra were acquired on a 600 MHz Varian Unity-Inova spectrometer equipped with inverse triple resonance ( 1H , 15N , 13C ) cold probe , or on BRUKER 500 MHz DRX or 800 MHz Avance III spectrometers equipped with inverse triple resonance ( 1H , 15N , 13C ) room temperature probes . Unless otherwise mentioned , all NMR experiments were performed at 40°C . The backbone resonance assignment of 15N/13C-VirB7XAC2622_24–139 was achieved by analyzing the following three-dimensional spectra: HNCA , HN ( CO ) CA , HNCO , HN ( CA ) CO , HNCACB and CBCA ( CO ) NH [55] . Side-chain resonance assignments were determined by analyzing the following spectra: 2D 13C-HSQC , 2D TOCSY , 2D NOESY , 3D HNHA , 3D HBHA ( CBCACO ) NH , 3D 15N-TOCSY-HSQC , 3D H ( CCO ) NH-TOCSY , 3D ( H ) CC ( CO ) NH-TOCSY , 3D 15N-NOESY-HSQC , 3D H ( C ) CH-TOCSY and 3D 13C-NOESY-HSQC ( specific for the aliphatic or the aromatic regions ) . The 3D H ( C ) CH-TOCSY and 3D 13C-NOESY-HSQC spectra were recorded in >99% ( v/v ) 2H2O . NOESY experiments were carried out using 0 . 2–0 . 3 mM protein samples and all NOESY mixing times were 100 ms . Backbone resonance assignments for 15N/13C-VirB7XAC2622_24–139 in complex with unlabeled VirB9XAC2620_154–255 were obtained from the analysis of three-dimensional HNCA , HN ( CO ) CA , HNCO , HNCACB , CBCA ( CO ) NH , HBHA ( CBCACO ) NH and 15N-NOESY-HSQC experiments recorded on a sample of 15N/13C-VirB7XAC2622_24–139 in the presence of a 20% molar excess of unlabeled VirB9XAC2620_154–255 . The backbone resonance assignment ( 15N and 1H ) for VirB7XAC2622_51–134 was performed by comparison with the 15N-HSQC spectrum of the larger construct , VirB7XAC2622_24–139 . Residual 1DNH dipolar couplings for VirB7XAC2622_24–139 ( 0 . 25 mM ) , alone or in complex with VirB9XAC2620_154–255 , were determined from the differences in 1H-15N splittings obtained in isotropic and anisotropic media . The 1H-15N splittings were extracted from a series of J-modulated 2D 1H-15N-HSQC experiments acquired at 30°C on the 500 MHz spectrometer , and analyzed essentially as described [56] . Residual alignment was induced by a liquid crystalline medium consisting of 5% penta-ethyleneglycol monododecyl ether ( C12E5 ) /n-hexanol ( r = 0 . 96 ) [57] . Fittings of RDCs to the NMR and X-ray structures were carried out using PALES [58] or MODULE [59] . All NMR spectra were processed with NMRPipe [60] and analyzed with CCPN Analysis [61] . Resonance assignments were deposited in the Biological Magnetic Resonance Data Bank ( BMRB accession code 17257 ) . Structure calculation and automated NOE assignment were performed using CYANA 2 . 1 [37] . Chemical shifts of 1Hα , 13Cα , 13Cβ , 13C′ and 15NH were used as input for TALOS [62] to predict φ and ψ dihedral angles that were subsequently used as restraints in the structure calculation . A total of 3471 NOE cross peaks were manually picked in the 15N-NOESY-HSQC and 13C-NOESY-HSQC ( acquired specifically for the aliphatic or the aromatic regions ) spectra and used as input for CYANA . The CYANA protocol consisted of 7 cycles of simulated annealing in torsion angle space with 10000 integration steps . In each cycle 300 conformers were calculated . The intra-residue , sequential and 10 unambiguous long range NOE cross peaks were assigned manually and imposed during all stages of the protocol . Chemical shift tolerances for automated NOE assignment were set to ±0 . 025 ppm in the direct 1H dimension , ±0 . 030 ppm in the indirect 1H dimension and ±0 . 25 ppm for the heteronuclei . The best 50 CYANA solutions were selected based on a target function criteria , and further refined in explicit water using CNS2 . 1 and HADDOCK2 . 0 [41] . The 1616 NOEs assigned by CYANA , the manually assigned NOEs , and the TALOS dihedral angle restraints were used as input during water refinement in HADDOCK . The best 20 lowest-energy conformers after water refinement were deposited in the Protein Data Bank ( PDB code 2L4W ) . The quality of the final NMR structures was investigated by PROCHECK-NMR [63] and PSVS [64] . The structures were visualized using MOLMOL [65] or PYMOL ( http://www . pymol . org ) . Experiments for measuring backbone 15N longitudinal ( T1 ) and transverse ( T2 ) relaxation times , and the heteronuclear {1H}-15N NOE , were recorded at 40°C on a Varian Inova 600 MHz spectrometer using standard Biopack pulse sequences ( Varian , Inc . ) . The inversion recovery delays used for sampling T1 relaxation were 50 , 250 , 450 , 650 , 850 , 1050 , 1250 , 1450 and 1650 ms . The CPMG delays used for sampling T2 relaxation were 10 , 30 , 50 , 70 , 90 , 110 , 130 , 150 , 170 , 190 and 210 ms . The 1H saturation period in the heteronuclear {1H}-15N NOE experiment was 3 s . All 2D spectra were acquired sequentially as matrices of 512 ( 1H ) ×128 ( 15N ) complex points , and inter scan delay of 3 s . Relaxation rates were obtained by fitting the time decay of peak intensities to an exponential decay function with three fitted parameters , the relaxation rate , the intensity at time zero and an offset , using the CCPN Analysis software [61] . Overall rotational correlation times ( τc ) were obtained either from the mean ratios of 15N T1/T2 relaxation rates of 1H-15N bond vectors located in regions of secondary structure using TENSOR2 [66] , or from estimates of 15N T1 and T2 relaxation times obtained from 1D 15N-edited relaxation experiments , as previously described [67] , [68] . Unlabeled VirB9XAC2620_154–255 or VirB9XAC2620_34–255 were titrated into a solution of 15N-labeled VirB7XAC2622_24–139 and a 15N-HSQC spectrum was acquired after the addition of each aliquot until a molar excess of 1 . 5 of unlabeled protein with respect to 15N-VirB7XAC2622_24–139 . In analogous experiments 15N-labeled VirB9XAC2620_154–255 was titrated with unlabeled VirB7XAC2622_24–139 or VirB7XAC2622_24–46 . VirB7XAC2622_24–139 oligomerization was studied by monitoring changes in the 15N-HSQC spectra as a function of protein concentration in the range of 7 to 850 µM . 15N-labeled VirB7XAC2622_51–134 was also titrated with unlabeled VirB7XAC2622_38–52 . Chemical shift perturbations due to sample dilution or protein-protein interaction were obtained from the weighted chemical shift changes in the 15N-HSQC spectra calculated according to Eq . 1 [69]: ( 1 ) where ΔδHN and ΔδNH are the chemical shift changes ( in ppm ) of the amide proton and nitrogen resonances , respectively . Docking simulations were used to build a docking model for the dimer of VirB7XAC2622_24–139 . The calculations were performed with HADDOCK2 . 0 using CNS2 . 1 [41] , and the first model of the NMR ensemble as starting structure . The docking was driven by unambiguous intermolecular NOEs and by ambiguous interaction restraints ( AIR ) between active residues and active plus passive residues . The following 13 NOE restraints were used: Ala43Hβ - Ile90Hδ1; Thr45Hα - T88Hγ2; Thr45Hβ - T88Hγ2; Glu46Hβ - Tyr87Hε; Ile47Hδ1 - Thr63Hβ; Ile47Hδ1 - Thr63Hγ2; Ile47Hδ1 - T88Hα; Ile47Hδ1 - T88Hγ2; Leu49Hγ - Ser85Hα; Leu49 Hδ1 - Asp86Hα; Leu49Hδ1 - Asp86Hβ1; Leu49Hδ1 - Asp86Hβ2 and Leu49Hδ2 - Asp86Hα . Active residues were identified as those showing chemical shift perturbations of at least one standard deviation above the mean , for 15N-VirB7XAC2622_24–139 as a function of protein concentration ( Figure 2 ) . Active residues were E42 , A43 , T45 , E46 and I47 at the N-terminal tail of the first subunit and D86 , T88 , I90 , G91 , Q115 of the C-terminal globular domain of the second subunit . Passive residues were F40 , D41 , P44 , P48 and L49 in the N-terminal segment , and P84 , S85 , Y87 , P92 , A113 and A114 in the C-terminal domain . The N-terminal tails ( residues 24–50 ) were kept fully flexible during all stages of the docking , while C-terminal domain residues at the interface , 84–92 , 61–63 and 113–115 , were maintained semi-flexible , being allowed to move only during semi-flexible refinement . The rigid body stage consisted of the calculation of 1000 solutions , from which 100 with the lowest HADDOCK scores were selected for semi-rigid simulated annealing in torsion angle space including backbone and side chain flexibility , and final refinement in Cartesian space with explicit water . The refined solutions were then clustered based on the backbone RMSD at the interface . An analogous protocol was used for docking the N-terminal ( residues 24–50 ) and C-terminal ( residues 51–139 ) fragments of VirB7 , except that in this calculation NOE restraints were not included and flexibility was introduced only at the interface . Models of residues 37–139 of VirB7XAC2622 ( VirB7XAC2622_37–139 ) were positioned around a tetradecameric ring made up of TraOCT subunits derived from the pKM101 T4SS TraN-TraOCT-TraFCT structure [34] . The VirB7XAC2622_37–139 subunits were placed so that Val37 and Asn38 occupied positions similar to Val33 and Asn34 of TraN . Molecular mechanics optimization and dynamics were then carried out from this initial model . The OPLS/AA force-field [70] for the protein and an implicit solvent representation of the generalized Born-formalism [71] were used . In addition to the molecular force-field , distance restraints with a flat-bottom harmonic functional form were included such that an energy penalty was added to the potential when the distance between specified pairs of atoms was less than a minimum threshold value of 3 Å or exceeded a maximum of 6 Å . A harmonic force constant of 5000 kJ⋅mol−1⋅Å−2 was used . The following 5 NOE proton pairs were specified for each VirB7XAC2622 pair interface: Ala43Hβ - Ile90Hδ1 , Thr45Hβ - T88Hγ2 , Ile47Hδ1 - Thr63Hβ; Leu49Hγ - Ser85Hα; and Leu49Hδ2 - Asp86Hα . These pairs correspond to a subset of the pairs for which intermolecular NOEs were detected . Thus , a total of 70 ( 14×5 ) distance restrains were included in the potential . All ionizable residues besides histidine were treated in their charged forms . Molecular dynamics were carried out at a temperature of 300 K for a total time of 2 ns . All computations were carried out with the GROMACS 4 . 5 . 3 software [72] . A construct corresponding to the VirB7 C-terminal region , VirB7XAC2622_51–134 , at 14 mg/mL in 5 mM Tris-HCl pH 7 . 5 and 25 mM sodium chloride , was submitted to vapor diffusion sitting-drop crystallization trials at 18°C . Large plates appeared after one day over a reservoir solution comprising 1 . 4 M ammonium sulfate and 4% ( v/v ) isopropyl alcohol . Reservoir solution supplemented with 25% ( v/v ) glycerol was used as cryoprotectant . Crystals were flash frozen at 100 K and submitted to X-ray diffraction at beam line W01B-MX2 of the Brazilian Synchrotron Light Laboratory ( LNLS ) coupled to a Marmosaic-225 CCD detector ( Mar , USA ) . The diffraction data were indexed , integrated and scaled using HKL2000 [73] . The crystal structure of VirB7XAC2622_51–134 was determined by molecular replacement by Phaser [74] using residues 51–134 of the lowest energy NMR structure as the search model . The model was iteratively refined using the graphics program Coot [75] with rounds of restrained refinement in Refmac5 [76] with individual anisotropic B-factor for all non-hydrogen atoms . All programs used for refinement were from the CCP4 suite [77] . The quality of the structure was analyzed by the programs Coot , Refmac5 , Procheck [78] , Rampage [79] and MolProbity [80] . The coordinates of the crystal structure were deposited in the PDB ( PDB code 3OV5 ) . Fluorescence experiments were conducted in an AVIV ATF-105 spectrofluorimeter ( AVIV Instruments ) . Fluorescence emission was collected after successive addition of 0 . 1 µM VirB7XAC2622_24–139_His aliquots into a sample containing 1 µM VirB9XAC2620_34–255 in 5 mM sodium acetate ( pH 5 . 0 ) . The final VirB7XAC2622_24–139_His concentration in the titration was 2 . 2 µM . Samples were pre-equilibrated for 2 min at 25°C , excited at 280 nm ( 2 nm bandwidth ) and fluorescence emission was detected from 334–342 nm ( 7 nm bandwidth ) at 2 nm intervals . The dissociation constant was calculated from a nonlinear regression fit of fluorescence titration data to Eq . 2 [81] , [82] , using the SigmaPlot program ( SPSS , Inc . ) : ( 2 ) where F is the fluorescence intensity at any given point of the titration curve , v is the initial volume , V is volume at any given point of the titration , [X] and [Y] are the VirB9XAC2620_34–255 and VirB7XAC2622_24–139_His concentrations at any given point of the titration , respectively , Kd is the dissociation constant and α is the ratio between the maximum fluorescence intensity and the initial fluorescence intensity . Samples with different concentrations of VirB7XAC2622_24–139 or VirB7XAC2622_51–134 were pre-equilibrated for 15 min , and then incubated with 0 . 01% ( v/v ) glutaraldehyde . After 20 min , the cross-linking reaction was stopped by the addition of SDS-PAGE sample buffer ( 100 mM Tris-HCl pH 6 . 8 , 3 . 7% ( w/v ) SDS , 18 . 7% ( v/v ) glycerol , 140 mM 2-mercaptoethanol and 0 . 01% ( w/v ) bromophenol blue ) . The cross-linking products were analyzed by 16% Tricine SDS-PAGE [83] . Approximately 1 kb of the upstream and downstream regions of the virB7XAC2622 gene were amplified by PCR from Xac genomic DNA and the two fragments were ligated to produce an in frame deletion , leaving only the region coding for the first three and last five codons . This sequence was then cloned into the BamHI restriction site of the pNPTS138 suicide vector ( M . R . Alley , unpublished ) , thereby generating the plasmid pNPTS-Δxac2622 . This vector was introduced into Xac by electroporation and replacement of the wild-type copy by the deleted version was obtained after two recombination events as described [84] . In order to complement the virB7XAC2622 knockout , a fragment including the virB7XAC2622 gene plus 1000 pb of upstream sequence was amplified by PCR from Xac genomic DNA and inserted into the pUFR047 vector [85] at the BamHI restriction site , creating the plasmid pUFR-VirB7 . This plasmid was then transferred to the ΔvirB7XAC2622 strain by electroporation and selection of gentamicin resistance . Virulence assays were performed by infiltration of the Xac strains into sweet orange leaves ( Citrus sinensis L . Osbeck ) . The cell cultures were adjusted to an optical density of 0 . 1 at 600 nm and leaves were inoculated by syringe infiltration with needle . The plants were maintained at 28°C with a 12 h photoperiod and the development of the symptoms was regularly observed . Bacterial growth was measured by harvesting 10 mm2 citrus leaf discs for each inoculated Xac strain , and the leaf discs were macerated in 150 mM sodium chloride . The solution was serially diluted and spread onto LBON plates [1% ( w/v ) tryptone , 0 . 5% ( w/v ) yeast extract and 1 . 5% ( w/v ) agar] containing ampicillin . The mean number of colony forming units per square centimeter ( CFU/cm2 ) was calculated by counting individual colonies obtained from each dilution . Rabbit polyclonal antibodies were raised against VirB7XAC2622_24–139 , VirB9XAC2620_34–255 and VirB10XAC2619_85–389_His . To analyze protein levels in Xac , cells were grown in XVM2 medium [86] to an optical density ( 600 nm ) of 1 . 0 . Recombinant proteins and Xac cellular extracts were separated by 18% SDS-PAGE followed by immunoblot analysis . The rabbit sera were used at 1∶1000 dilutions and the antibodies were detected with staphylococcal protein A conjugated to horseradish peroxidase ( Sigma ) . Immunoblots were developed with the ECL Advance Western Blotting system and exposed to Amersham Hyperfilm ECL film ( GE Healthcare ) . Strains of Xac were grown in XVM2 to an optical density ( 600 nm ) of 1 . 0 and equal numbers of cells were harvested by centrifugation . Cells were resuspended in 500 µl of 50 mM Hepes ( pH 8 . 0 ) containing 5 mM EDTA . Coimmunoprecipitation experiments were performed after solubilization of cells with deoxycholate ( DOC ) and N , N–Dimethyldodecylamine N-oxide ( LDAO ) detergents , as described [87] . Bacterial suspensions were treated with 200 µg/ml lysozyme for 1 h at 4°C . Cells were lysed by sonication in the presence of 100 mM NaCl , 1% ( w/v ) DOC and protease inhibitor cocktail ( Sigma Aldrich ) . Volumes were adjusted to 1 mL with 50 mM Hepes ( pH 8 . 0 ) and 2% ( w/v ) LDAO and lysates were solubilized by incubation for 16 h at 4°C with agitation . Solubilized material was isolated by centrifugation at 15 , 000×g for 20 min and incubated with Protein G agarose beads ( Millipore ) for 4 h at 4°C , removing nonspecifically bound proteins . For immunoprecipitation , supernatants from this pre-clearing step were incubated with protein G agarose beads in the presence of pre-immune serum ( controls ) or anti-VirB7 , anti-VirB9 or anti-VirB10 antibody , for 16 h at 4°C , with agitation . The unbound material was discarded after centrifugation ( 400×g , 5 min ) and beads were washed three times with decreasing concentrations of LDAO and DOC ( final wash contained 0 . 1% DOC and 0 . 1% LDAO ) , each time for 10 min at 25°C , with agitation . Immunoprecipitates were eluted by addition of SDS sample buffer and boiling for 10 min . Samples were analyzed by SDS-PAGE followed by immunoblot . RNAs from Xac cultures grown in XVM2 medium to an optical density ( 600 nm ) of 1 . 0 were extracted using Illustra RNAspin Mini kit , according to manufacturer's instructions ( GE Healthcare ) . Purified RNA was treated with 1 U/µg DNaseI , RNase-free ( Fermentas ) and successful removal of contaminating DNA was confirmed by PCR . Reverse transcription was performed using 1 µg of DNaseI-treated RNA and RevertAid H Minus First Strand cDNA Synthesis Kit , following manufacturer's protocol ( Fermentas ) . Quantitative amplification of the resulting cDNA ( 40 ng ) was performed using 0 . 3 µM of each primer ( F_virB9_RT and R_virB9_RT; Table S1 ) and SYBR Green/ROX qPCR Master Mix in the ABI7300 Real-Time System ( Applied Biosystems ) . Relative quantification of gene expression was performed using gene xac1631 ( that codes for subunit A of DNA gyrase ) as an endogenous control and the 2−ΔΔCT method [88] . Primers were designed using the PrimerExpress Software ( Applied Biosystems ) . Triplicates of two independent biological samples were used . 1H , 13C and 15N resonance assignments of VirB7XAC2622 were deposited in the BMRB: entry 17257 . NMR and X-ray crystallography coordinates have been deposited in the PDB with accession codes 2L4W and 3OV5 , respectively . | Many aspects of bacterial life require that they translocate proteins to the cell exterior . To do this , different macromolecular secretion systems of varying complexity have evolved ( Type I–VI secretion systems ) . These secretion systems are often at the front lines of pathogen-host interactions and are important for the development of disease . In this work , we have determined the structure and studied the interactions of an unusually large VirB7 subunit ( VirB7XAC2622 ) of the outer membrane pore of the Type IV secretion system found in the Xanthomonas genera of phytopathogens . Its mosaic structure combines a canonical VirB7 N-terminal region with a C-terminal globular domain whose topology is observed in a relatively limited set of proteins , all involved in molecular transport across outer membranes . Our results lead to the hypothesis that the VirB7XAC2622 globular domains can form an extra ring around the perimeter of the outer membrane pore and reveal deeper structural and evolutionary relationships among bacterial macromolecular secretion systems that have evolved to adopt a variety of functions , including structural modules in outer membrane pores ( secretins from Type II , III and IV secretion systems , Type IV pili and filamentous phages ) , signal-transduction modules in TonB-dependent receptors and membrane-penetrating devices in T6SS and long-tailed bacteriophages . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"macromolecular",
"assemblies",
"microbiology",
"host-pathogen",
"interaction",
"protein",
"structure",
"bacterial",
"pathogens",
"proteins",
"lipoproteins",
"biology",
"recombinant",
"proteins",
"biophysics",
"biochemistry",
"gram",
"negative",
"lipoprotein",
"structure"
] | 2011 | A Component of the Xanthomonadaceae Type IV Secretion System Combines a VirB7 Motif with a N0 Domain Found in Outer Membrane Transport Proteins |
Animals have many ways of protecting themselves against stress; for example , they can induce animal-wide , stress-protective pathways and they can kill damaged cells via apoptosis . We have discovered an unexpected regulatory relationship between these two types of stress responses . We find that C . elegans mutations blocking the normal course of programmed cell death and clearance confer animal-wide resistance to a specific set of environmental stressors; namely , ER , heat and osmotic stress . Remarkably , this pattern of stress resistance is induced by mutations that affect cell death in different ways , including ced-3 ( cell death defective ) mutations , which block programmed cell death , ced-1 and ced-2 mutations , which prevent the engulfment of dying cells , and progranulin ( pgrn-1 ) mutations , which accelerate the clearance of apoptotic cells . Stress resistance conferred by ced and pgrn-1 mutations is not additive and these mutants share altered patterns of gene expression , suggesting that they may act within the same pathway to achieve stress resistance . Together , our findings demonstrate that programmed cell death effectors influence the degree to which C . elegans tolerates environmental stress . While the mechanism is not entirely clear , it is intriguing that animals lacking the ability to efficiently and correctly remove dying cells should switch to a more global animal-wide system of stress resistance .
In nature , animals are constantly exposed to changing environmental conditions . In order to survive , organisms must weather normal and oftentimes extreme variations in temperature , water availability , salt levels , xenobiotics and other environmental factors . To cope , animals have developed mechanisms for stress protection . At the cellular level , DNA damage and other forms of stress can induce apoptosis to remove damaged cells; at the organismal level , stressful conditions can sometimes induce responses that make the entire animal more stress resistant [1] , [2] . The nematode C . elegans has evolved several stress-protective responses . These include aversive behaviors , such as avoidance of noxious stimuli , and the activation of alternative developmental programs , such as entry into the dauer state of diapause [3] , [4] . The animal can also induce environmental stress resistance by turning on gene transcription to manage the stressor using , for example , the heat-shock transcription factor HSF-1 to combat heat stress [5] , SKN-1/Nrf2 to combat xenobiotic stress [6] and the hypoxia-inducible factor HIF-1 to combat hypoxia [7] . In addition to coordinated stress responses at the organismal level , C . elegans , like other organisms , can protect itself against stress at the cellular level . For example , in C . elegans , germ cells undergo apoptosis in response to DNA damage from ionizing radiation [8] , [9] . In general , these types of single-cell , live-or-die decisions may be made to sacrifice a part for the betterment of the whole . How these decisions are made and the mechanistic and molecular relationship , if any , between animal-wide stress responses and programmed cell death are , however , poorly understood . As a fundamental process by which organisms remove unnecessary , abnormal or damaged cells , programmed cell death involves both cell killing via apoptosis and cell corpse removal via phagocytosis and degradation [10] . Although once considered a disinterested second party that simply removes the dead cell , the engulfing cell is now known to be an active participant in the cell death program . For example , in weak C . elegans caspase mutants , in which decisions about whether to complete the cell death program are made stochastically , a second mutation in an engulfment gene further reduces cell death [11] , [12] . Likewise , in mammals , mutations affecting either the dying or engulfing cell can disrupt tissue homeostasis and produce developmental disorders , autoimmune disease , cancer and neurodegeneration [13] . Genes responsible for carrying out apoptosis and apoptotic-cell engulfment were first described in C . elegans ( Figure S1 ) [14] , [15] . In the apoptotic cell , the canonical programmed cell death pathway involves the Apaf-1 like protein CED-4 , which is inhibited by the BCL-2-like protein , CED-9 [16]–[19] . When disinhibited by developmental cues , CED-4 activates the C . elegans executioner caspase CED-3 [20] , [21] . In the engulfing cell , several partially redundant pathways govern the membrane and cytoskeletal rearrangements required for phagocytosis of the dying cell ( Figure S1 ) [22]–[25] . C . elegans was instrumental in illuminating the core features of programmed cell death and clearance because it is highly amenable to genetic and experimental manipulation . Recently , we implicated the human disease gene progranulin in the regulation of programmed cell death using a C . elegans mutant [26] . The regulation and function of progranulin are particularly interesting because of its links to disease . Progranulin haploinsufficiency causes the human neurodegenerative disease frontotemporal lobar degeneration while homozygous null carriers develop neuronal ceroid lipofuscinosis [27]–[29] . Allelic variations in the gene have also been linked to Alzheimer Disease , Parkinson Disease and amyotrophic lateral sclerosis ( ALS ) [30]–[34] , and altered progranulin levels have been implicated in autoimmune disease [35] , [36] , cancer [37]–[42] and ischemic injury [43] , [44] . Thus , precise regulation of progranulin levels is important for maintaining health and homeostasis . Previously , we showed that progranulin normally functions to regulate the rate of apoptotic cell engulfment during the process of programmed cell death [26] . In pgrn-1 ( - ) mutants , apoptotic cells are cleared approximately twice as fast as normal . We also showed that macrophages from progranulin null-mutant mice are able to engulf apoptotic cells more rapidly than are wild-type macrophages . Thus , progranulin , like mtm-1 , abl-1 and srgp-1 , is a negative regulator of programmed cell death clearance [25] , [26] , [45]–[47] ( Figure S1 ) . Given the close relationship between environmental stress and age-related disease , we asked whether pgrn-1 ( - ) mutants exhibited an altered response to cellular stressors . We found that they did . However , unexpectedly , they demonstrated increased stress resistance . Even more surprisingly , we found that the same was true of mutations that perturb cell death in other ways , suggesting that a stress response pathway is activated when any part of the programmed cell death pathway does not proceed normally . Our findings reveal an unexpected link between mechanisms that control life-or-death decisions at the level of the individual cell and at the level of the entire animal .
We tested the resistance of progranulin mutants to several environmental stressors . We found that compared to wild-type controls , pgrn-1 ( tm985 ) mutants were resistant to osmotic , heat and endoplasmic reticulum ( ER ) stress ( the latter as measured by resistance to tunicamycin , an inhibitor of N-linked glycosylation ) ( Figure 1A–C , Tables S1A–C ) . In contrast , pgrn-1 mutants had normal responses to oxidative stress ( paraquat ) , genotoxic stress ( UV light ) and pathogen exposure ( P . aeruginosa and S . enterica; Figure S2A–C and data not shown ) . Reintroducing either the C . elegans or human progranulin gene into pgrn-1 ( - ) mutants rescued or partially rescued the mutant stress resistance phenotypes ( Figure 1C–D , Tables S1C–D ) . The partial rescue by human progranulin only at higher doses of tunicamycin could be due to species differences or differences in binding affinities to the progranulin receptor . What do heat , osmotic stress and tunicamycin have in common ? One possibility is that they all beget unfolded proteins and induce the ER unfolded protein response . To address this idea , we tested the ability of each stressor to increase expression of hsp-4 . HSP-4 is the nematode ortholog of mammalian grp78/BiP/HSP70 , and is upregulated by heat and ER stress [48] , [49] . We confirmed that heat stress and tunicamycin increased Phsp-4::gfp reporter levels ( Figure S3A–B ) , and found that paraquat , UV irradiation and exposure to P . aeruginosa did not ( Figure S4A–C ) . However , under our conditions , osmotic stress did not increase Phsp-4::gfp levels ( Figure S3C ) . Thus , induction of the ER stress-resistance marker Phsp-4::gfp is not a feature that unifies heat , tunicamycin and osmotic stress . Because loss of function of pgrn-1 , a regulator of programmed cell death clearance , caused stress resistance , we asked whether other mutations affecting programmed cell death would also affect the stress response of the whole animal . In contrast to progranulin mutants , loss-of-function mutations in the gene encoding the executioner caspase ced-3 prevent apoptosis [15] . In ced-3 loss-of-function mutants , cells that normally die during development instead persist . Surprisingly , we found that ced-3 ( n717 ) mutant animals also exhibited increased resistance to ER stress ( Figure 2A and Tables S2A ) . Moreover , pgrn-1 ( - ) ; ced-3 ( n717 ) double mutants were no more stress-resistant than were either of the single mutants ( Figure 2A and Table S2A ) , suggesting that pgrn-1 and ced-3 mutations may activate the same stress response pathway . ced-3 ( n717 ) mutants could also exhibit osmotic and heat stress resistance , albeit not as consistently as ER stress resistance ( Figure 2B–C and Table S2B ) . Like pgrn-1 mutants , ced-3 ( n717 ) mutants did not exhibit resistance to paraquat or UV light ( Figure S5 ) . Many alleles of ced-3 have been isolated , and they form an allelic series based on their ability to inhibit programmed cell death [50] . We found that ced-3 alleles exhibited graded levels of ER stress resistance that were correlated with their ability to block programmed cell death . The strong ced-3 allele n717 was more resistant to ER stress than were two intermediate strength ( n1949 and n2436 ) alleles , and these , in turn , were more resistant than the weak ( n2438 ) allele ( Figure 2D and Table S2A ) . These results are consistent with a model in which the stress resistance conferred by ced-3 mutations is mechanistically related to the apoptotic killing conferred by ced-3 . In C . elegans , ced-9 and ced-4 regulate the ability of ced-3 to activate programmed cell death [17] . The Bcl-2-like protein CED-9 inhibits CED-4/Apaf1 activity , blocking cell death; whereas activated CED-4 cleaves CED-3 and activates its caspase function , leading to cell death [18] , [51] . Therefore , ced-9 gain-of-function ( gf ) and ced-4 loss-of-function ( lf or - ) mutations are similar to ced-3 ( - ) mutations in the sense that they impair programmed cell death , whereas ced-9 ( lf ) mutations cause excessive cell death ( and animal lethality ) due to uncontrolled activity of CED-4 and CED-3 . We tested ced-9 ( n1950gf ) and ced-4 ( n1162lf ) single mutants , as well as a ced-4 ( n1162lf ) ced-9 ( n2812lf ) double mutant , for their responses to ER stress . We found that ced-9 ( gf ) mutants were resistant to ER stress ( Figure 2E and Table S2C ) . Further , the stress resistance conferred by ced-9 ( gf ) mutations was not additive with that conferred by either pgrn-1 ( - ) or ced-3 ( - ) mutations , once again suggesting that these genes affect the same stress-response pathway ( Figure 2E , Table S2C ) . We also found that ced-4 ( lf ) single mutants were resistant to ER stress at low doses of tunicamycin in one of two experiments ( Table S2D ) , and that these mutants did not require intact ced-9 for this stress resistance ( Figure 2F , Table S2D ) . These findings suggest that ced-4 ( and likely ced-3 ) may be genetically downstream of ced-9 in the stress-resistance pathway , as it is in the cell death pathway . However , since ced-4 ( lf ) animals displayed an incomplete degree of stress resistance compared to ced-3 ( lf ) and ced-9 ( gf ) mutants , it remains possible that ced-4 is dispensable or redundant in this stress response pathway . In addition to mutations that accelerate the clearance of apoptotic corpses , or prevent apoptosis altogether , we also asked whether mutations in apoptotic cell engulfment pathways affected stress resistance . We found that certain engulfment mutant alleles increased ER stress resistance , although the degree of resistance seen in engulfment mutants was generally less than that seen in animals carrying pgrn-1 ( - ) or strong ced-3 ( - ) mutant alleles . The engulfment mutants ced-1 ( e1735 ) , ced-6 ( n1813 ) , ced-7 ( n1892 ) and ced-2 ( e1752 ) , ced-5 ( n1812 ) , ced-10 ( n3246 ) exhibited resistance to ER stress at low doses of tunicamycin ( 1 µg/mL ) . However they exhibited variable responses to tunicamycin at higher doses ( 5 µg/mL ) , with ced-5 , 7 and 10 mutants demonstrating ER stress resistance but not ced-1 , 2 or 6 mutants ( Figure 3A–B and Table S3A–B ) . Once again , double mutants containing pgrn-1 and engulfment mutations were not more resistant than pgrn-1 ( - ) alone , suggesting that these mutations could potentially induce a common ER stress-resistance pathway ( Figure 3A–B and Table S3A–B ) . We also tested the response of engulfment mutants to other stressors . We found that ced-1 ( e1735 ) and ced-2 ( e1752 ) mutants were resistant to osmotic stress ( Figure S6A and Table S3C ) . The ced-2 mutant was also resistant to thermal stress in 1 of 2 trials ( Figure S6 and Table S3C ) . As a group , the engulfment mutants were not as robustly resistant to environmental stressors as pgrn-1 and ced-3 mutants , which may be due to the partial functional redundancy of engulfment pathway genes ( See Figure S1 ) . Like pgrn-1 , the tyrosine kinase ABL-1 is a negative regulator of apoptotic corpse engulfment [45] . However , unlike pgrn-1 , abl-1 does not act through the canonical engulfment pathways . Instead , abl-1 negatively regulates the engulfment gene abi-1 to inhibit cell death clearance ( See Figure S1 ) . We asked whether these two genes might influence ER stress resistance , and found that both abl-1 and abi-1 mutants exhibited resistance to low doses of tunicamycin ( Figure 3C , Table S3D ) . At higher doses of TM , abl-1 ( - ) and one mutant allele of abi-1 , ok171 , were resistant to tunicamycin stress compared to wild type . Curiously , in some situations , abl-1 mutations actually reduced ER stress resistance . For example , abl-1 ( n1963 ) mutations alone have no visible effect on engulfment of apoptotic corpses; however , abl-1 mutations reduce the severity of the engulfment phenotype of ced-1 ( n2091 ) and ced-6 ( n2095 ) mutants [45] . Likewise , we found that abl-1 ( n1963 ) mutations reduced the level of ER stress resistance conferred by ced-1 ( n2091 ) and ced-6 ( n2095 ) mutations ( Figure S7A and Table S3E ) . We do not have a simple unifying explanation for these findings at this time , but they indicate that pgrn-1 is not the only negative regulator of cell engulfment that can affect ER stress resistance . We also tested two additional genes that may modulate but are not directly involved in programmed cell death for stress response phenotypes . A mutation in unc-73 enhances the effect of other engulfment mutants but alone has no engulfment defect [52] . A mutation in unc-53 results in defective migration of cells and neuronal processes and the UNC-53 protein interacts with ABI-1 [53] . However , neither unc-73 ( e936 ) nor unc-53 ( e404 ) mutants exhibited ER stress resistance phenotypes ( Figure S7B and Table S3F ) . Thus , not all genes involved in apoptotic cell engulfment are utilized for stress response . Recently , pqn-41 was identified as a mediator of a type of non-apoptotic cell death [54] . This type of cell death , characterized by crenellation of the nuclear envelope and organelle swelling , occurs independently of the ced-3 caspase and engulfment genes [55] , [56] . To determine if pqn-41 affects ER stress resistance , we tested a deletion mutant , ns924 . We found that pqn-41 ( ns924 ) mutants were resistant to ER stress at a low dose of tunicamycin but not at a high dose , similar to some of the engulfment mutants we tested such as ced-1 , ced-2 and ced-6 ( Figure S7C and Table S3G ) . These data suggest that organismal stress resistance may be linked to both apoptotic and non-apoptotic programmed cell death . Together , these findings indicate that perturbing C . elegans programmed cell death in a variety of ways , either by affecting the initiation of cell death or the engulfment of the dying cell , can confer whole-animal resistance to environmental stress . Because of recent findings connecting the unfolded protein response with neurodegenerative diseases [57] , we decided to investigate the mechanism by which a pgrn-1 mutation affected ER stress resistance . The UPR is the cellular program that responds to ER stress . The UPR is mediated by three ER resident proteins encoded by ire-1 , pek-1 ( mammalian Perk ) and atf-6 [58] . Mutations in these genes impair the response to ER stress in C . elegans [59] , [60] . Part of this stress response includes alternative splicing of xbp-1 mRNA by activated IRE-1 and the consequent upregulation of XBP-1 target genes , such as hsp-4 , the nematode ortholog of mammalian grp78/BiP/HSP70 [48] , [49] . To investigate the role of the UPR in the stress resistance of apoptosis mutants , we first tested whether pgrn-1 mutants required ire-1 for ER stress resistance . We tested a pgrn-1; ire-1 double mutant and found that resistance in pgrn-1 mutants was dependent on ire-1 ( Figure 4A and Table S4A ) suggesting that the mechanism of stress resistance of cell death mutants requires this branch of the UPR pathway . One possibility was that the IRE-1 pathway is constitutively activated in pgrn-1 mutants . To test this , we measured the levels of spliced xbp-1 mRNA . Interestingly , we found no changes in the levels of spliced xbp-1 mRNA in pgrn-1 mutants compared to wild type ( Figure S8 ) . We also investigated whether ced-3 or ced-1 mutants exhibited increased levels of spliced xbp-1 mRNA . Similar to pgrn-1 mutants , they did not ( Figure S8 ) . Since the hsp-4 gene is a target of active XBP-1 , we measured whether pgrn-1 mutants displayed increased Phsp-4::gfp reporter levels . We found that except for one time point at the L4 stage of larval development , Phsp-4::gfp levels in pgrn-1 mutants were largely unchanged compared to controls . Correspondingly , levels of Phsp-4::gfp in ced-3 ( n717 ) mutants were also generally unchanged compared to controls ( Figure S9 ) . These data suggest that although pgrn-1 mutations affect the ER stress response through the ire-1 gene , the downstream splicing of xbp-1 mRNA and expression of Phsp-4::gfp is not affected . pgrn-1 mutations may somehow make the IRE-1 branch of the UPR more effective without dramatically changing its activity . Mammalian progranulin has been demonstrated to activate the insulin/IGF-1 pathway and downstream MAP kinases [37] , [61] , [62] . Animals carrying mutations in the C . elegans insulin/IGF-1 receptor , daf-2 , are long-lived [63] , [64] and resistant to many stressors , including heat , osmotic stress and ER stress [63] , [64] . The longevity and stress resistance of daf-2 mutants require the FOXO transcription factor daf-16 [60] . Upon inactivation of daf-2 , DAF-16 accumulates in the nucleus [65] where it regulates transcription of stress response genes . We found that the degree of ER stress resistance of pgrn-1 ( tm985 ) ; daf-2 ( e1370 ) double mutants was similar to that of daf-2 ( - ) single mutants ( Figure 4B and Table S4B ) . pgrn-1 mutants also required intact daf-16 for stress resistance , as daf-16 ( mu86 ) pgrn-1 ( tm985 ) double mutants were no more stress resistant than were single daf-16 ( mu86 ) mutants ( Figure 4B and Table S4B ) . These findings suggest that pgrn-1 may be part of the daf-2 pathway or act with daf-2 to confer stress resistance . However , unlike mutations in daf-2 , pgrn-1 ( - ) does not affect nuclear localization of DAF-16::GFP protein ( Figure S10 ) . DAF-16::GFP localization is also unaffected in ced-1 and ced-3 mutant animals ( Figure S10 ) . To determine if nuclear localization of DAF-16 would further increase stress resistance in pgrn-1 mutants , we crossed the daf-16aAM transgene ( which causes DAF-16 nuclear accumulation due to mutation of its AKT-phosphorylation sites ) into a pgrn-1 ( - ) background [65] . We found that a pgrn-1 ( - ) ; daf-16aAM strain was no more stress resistant than was the pgrn-1 mutant alone ( Figure 4C and Table S4C ) . Others have shown that adult-only ced-3 RNAi extends lifespan without altering DAF-16::GFP localization [66] . Given the correlation between lifespan extension and some forms of stress resistance [67] , we tested the lifespan of ced-3 , ced-1 and ced-2 mutants . In earlier work , we showed that pgrn-1 ( - ) mutant lifespan is no different than wild type [26] . Whereas a ced-3 ( - ) mutation significantly extended lifespan compared to wild type , ced-1 and ced-2 mutations did not ( Figure S11 ) , indicating that longevity and this type of organismal stress resistance can be dissociated . Several MAP kinases are required for responses to cellular stressors in C . elegans . The PMK-1/p38 MAP kinase encoded by pmk-1 is required for resistance to oxidative stressors [68] , pathogenic bacteria [69] and exogenously induced ER stress [70] . We confirmed that pmk-1 mutations increased sensitivity to ER stress and found that pgrn-1 ( - ) ; pmk-1 ( km25 ) double mutants were no more resistant to ER stress than were pmk-1 single mutants . Thus , pmk-1 is required for the ER stress resistance induced by pgrn-1 mutations ( Figure 4D and Table S4D ) . Progranulin is a secreted protein . In mammals , two progranulin receptors have been identified , the tumor necrosis factor receptor ( TNFR ) and sortilin [71] . Thus , we tested a downstream TNF receptor associated factor ( TRAF ) mutant , trf-1 ( nr2014 ) , for stress resistance and epistasis with pgrn-1 ( - ) . trf-1 mutants were not stress resistant compared to wild type and pgrn-1 ( - ) did not require trf-1 for its stress resistance ( Figure S12A ) . We also tested two mutant alleles of trk-1 , a C . elegans neurotrophin receptor similar to a co-receptor for sortilin , the other mammalian progranulin receptor . Again , pgrn-1 mutants did not require trk-1 for stress resistance ( Figure S12B ) . Thus , an as yet unidentified receptor ( s ) appears to be required for progranulin to influence ER stress resistance in C . elegans . If mutations that perturb cell death in different ways act in the same stress-resistance pathway , then they might share gene expression patterns that differ from wild type . To test this , we performed gene expression profiling by RNA sequencing ( RNA-seq ) , comparing day 1 adult pgrn-1 ( tm985 ) , ced-3 ( n717 ) and ced-1 ( e1735 ) mutants to wild-type animals . This allowed us to assess 1 ) whether these strains have altered gene expression , and 2 ) whether their differentially expressed genes are shared , suggesting the involvement of a common pathway . In spite of different cell death phenotypes of these mutants , RNA-seq revealed that all three mutants down-regulated the same 95 genes and up-regulated the same 9 genes , a highly significant portion of the total transcriptome ( p<10−16 ) ( Figure 5 and Table S5 ) . Of the genes that share differential regulation in our mutants , a significant number are regulated by DAF-16 ( Table S5 ) . These findings are consistent with the possibility that the stress resistance phenotypes of these three mutants may be due to the involvement of shared pathways .
We have shown that mutations that impair C . elegans programmed cell death in any of three ways—by inhibiting apoptosis , by impairing corpse clearance or by accelerating corpse clearance—all enhance resistance to certain environmental stressors . These mutations do not confer resistance to all cellular stressors , as pgrn-1 and ced-3 mutants exhibit normal sensitivity to UV light and oxidative stress . Several questions naturally follow from these findings . First , why are these mutants resistant to specific stressors ? Tunicamycin is an N-linked glycosylation inhibitor that causes retention of translated proteins in the ER and induces the unfolded protein response . Similarly , heat and osmotic stress cause ER and/or cytosolic proteins to unfold and activate signaling programs that induce expression of heat shock proteins and other chaperones . Thus , it seems possible that cell death mutations all trigger a response specific for unfolded proteins , such as the ER unfolded protein response . Consistent with this , we found that the cell-death-related stressors tunicamycin and heat both activated the ER stress-response gene hsp-4/BIP , whereas the cell-death-unrelated stressors Pseudomonas , UV and paraquat did not . However , Phsp-4::gfp was not induced by osmotic stress and pgrn-1 ( - ) mutation did not increase basal Phsp-4::gfp expression levels . Thus , defects in cell death can do more to protect the animal than simply to induce the canonical UPR . They must generate a more multifaceted response that can maintain proteostasis . The mechanism by which cell death mutations induce animal-wide stress resistance is not known . If different cell-death mutations affected different pathways , or the same pathway to different extents , then one would expect double mutants to be even more stress resistant than the individual single mutants . As this was not the case , it is possible that all of these mutations trigger the same stress response , though other interpretations remain possible . We identified some of the genes required for pgrn-1 mutants to resist the ER stressor tunicamycin . We found that this stress resistance requires an intact ire-1 gene , the MAP kinase PMK-1 and the transcription factor DAF-16/FOXO . We also identified a number of genes that are differentially regulated by all three of our mutants ( compared to wild type ) , suggesting that they may achieve stress resistance by recruiting shared genes and/or pathways . It will be very interesting to explore these shared genes in future studies . In contrast to the selective stress resistance of pgrn-1 mutants , daf-2 mutants are resistant to most environmental stressors . Thus , decreased DAF-2/insulin/IGF-1 signaling activates a genetic program that more generally elevates organismal resilience . daf-2 mutants appear to increase their resistance to ER stress by making the ire-1/xbp-1 pathway more efficient , possibly by activating stress-response transcription factors like DAF-16 that collaborate with ire-1/xbp-1 to induce new protective genes [60] . In concordance with this , daf-2 mutants actually have reduced levels of expression of ire-1/xbp-1-regulated genes such as hsp-4/BIP . While this was not the case for pgrn-1 or ced-3 mutants , whose Phsp-4::gfp levels were not decreased , there are some unexpected similarities between the ER stress-resistance phenotypes of cell-death mutants and daf-2 mutants . First , the ER stress resistance phenotypes of pgrn-1 and daf-2 mutants require daf-16 , either fully ( pgrn-1 mutant ) or partially ( daf-2 mutant ) . Second , both ER stress responses are completely dependent on ire-1 , yet in neither case is xbp-1 splicing increased . Additionally , the cell death mutants exhibited either no increase or only slight increases in Phsp-4::gfp expression , rather than the substantial increase one might expect if this branch of the ER pathway were constitutively active . Finally , the daf-2 and pgrn-1 mutant ER stress-resistance phenotypes are not additive . Thus the ER-stress resistance pathways activated by these mutations likely share at least some mechanistic features . It was striking that mutations that perturb programmed cell death in such different ways all had similar effects on environmental stress resistance . Why should mutations that lead to undead cells ( apoptosis mutations ) , lingering corpses ( engulfment mutations ) , and prematurely-engulfed corpses ( progranulin mutations ) all activate what appears to be ( from genetic tests ) the same stress resistance pathway ? From an evolutionary perspective , this linkage may make sense . Presumably apoptosis evolved not only to sculpt tissues during development , but also to remove cells that are damaged and unable to perform their normal functions , or perhaps that are overtly harmful to the animal . Viewed in this way , impairments in programmed cell death could be interpreted by the organism as an inability to respond normally to stress . Perhaps , under these conditions , the animal uses an alternative , back-up system to survive; namely , the system we have described in this study . Specifically , animals could have developed a sensitive surveillance system that can detect abnormalities in cell death , and respond to them by activating another pathway that enhances their overall level of stress resistance . The existence of this type of alternative system could have increased animal fitness during evolution in turbulent or adverse environments . While this model makes sense from an evolutionary perspective , other models are possible as well . For example , perhaps cell-death proteins , which act in a multi-step pathway to remove unwanted cells , also act together in a different pathway that has the effect of sensitizing the animal to various forms of stress . Non-cell death functions have , in fact , been described for cell-death effectors . In C . elegans , ced-10 ( - ) mutations impair not only cell engulfment but also cell migration [72] . Another engulfment gene , ced-1 , has also been implicated in neuronal regulation of innate immunity [73] , [74] . In mammals , a defect in the BCL2-family protein BID impairs cytokine production in response to immune activation independently of its cell death signaling function [75] . Further , certain mammalian executioner caspases can activate microglia in response to inflammogens without causing microglial death [76] . Finally , in a mouse model of Alzheimer Disease , caspase activation may be responsible for tau cleavage and aggregate formation , thereby serving a protective function [77] . Thus , programmed cell death effectors could hypothetically act together to sensitize animals to certain stressors or , alternatively , to inhibit a stress-response pathway . In either case , perturbing programmed cell death would increase organismal stress resistance . A protein in the flowering plant Arabidopsis may support the model that programmed cell death effectors can sensitize an animal to environmental stress . Arabidopsis can express a protein called RD21 that , like CED-3 , is a cysteine protease , and , like progranulin , contains a granulin domain . Osmotic stress induces RD21 and , possibly as a consequence , leaf senescence . Interestingly , in response to stress RD21 undergoes a process of maturation in which its caspase domain cleaves and releases the granulin domain [78]–[80] . In RD21 , the caspase and granulin domains are contained within the same molecule . However , perhaps in C . elegans , the two domains reside in different proteins but nevertheless act together to influence organismal stress resistance . In summary , our findings indicate that programmed cell death effectors not only kill and remove individual cells , but also influence environmental stress resistance at the level of the whole animal . To our knowledge this is the first time that cell-death effectors like ced-3 , pgrn-1 and ced-1 have been implicated in organismal stress resistance , and these findings raise many interesting new questions about both mechanism and evolution .
Unless otherwise indicated , C . elegans were cultured at 20°C using standard procedures [81] . Strains were kindly provided by the Mitani Laboratory ( National Bioresource Project ) at the Tokyo Women's Medical University and the Caenorhabditis Genetics Center ( CGC ) at the University of Minnesota . Strains were outcrossed four times to the laboratory N2 control strain ( N2 Bristol ) . Descriptions of strains can be found at www . wormbase . org . The following strains were used: AWK2 ced-9 ( n1950gf ) III; ced-3 ( n717 ) IV AWK74 daf-16 ( mu86 ) I; ced-3 ( n717 ) IV AWK76 daf-16 ( mu86 ) pgrn-1 ( tm985 ) I; muIs109[Pdaf-16::daf-16::gfp+Podr-1::RFP] AWK77 daf-16 ( mu86 ) I; ced-3 ( n717 ) IV; muIs109[Pdaf-16::daf-16::gfp+Podr-1::RFP] AWK78 ced-1 ( e1735 ) daf-16 ( mu86 ) I AWK80 ced-1 ( e1735 ) daf-16 ( mu86 ) I; muIs109[Pdaf-16::daf-16::gfp+Podr-1::RFP] AWK109 pgrn-1 ( tm985 ) I; pqn-41 ( ns294 ) III AWK111 pgrn-1 ( tm985 ) I; muIs113[Pdaf-16::daf-16AM::gfp+rol-6] CB404 unc-53 ( e404 ) II CB936 unc-73 ( e936 ) I CF1037 daf-16 ( mu86 ) I CF1041 daf-2 ( e1370 ) III CF1934 daf-16 ( mu86 ) I; muIs109[Pdaf-16::daf-16::gfp+Podr-1::RFP] CF2260 N2; zcIs4[Phsp-4::gfp] V CF2473 ire-1 ( ok799 ) II CF3050 pgrn-1 ( tm985 ) I CF3165 pgrn-1 ( tm985 ) I; zcIs4[Phsp-4-4::gfp] V CF3170 pgrn-1 ( tm985 ) I; ire-1 ( ok799 ) II CF3196 daf-16 ( mu86 ) pgrn-1 ( tm985 ) I CF3206 pgrn-1 ( tm985 ) I; daf-2 ( e1370 ) III CF3324 ced-3 ( n717 ) IV CF3419 pgrn-1 ( tm985 ) I; ced-3 ( n717 ) IV CF3447 pgrn-1 ( tm985 ) I; muIs189[Ppgrn-1:: pgrn-1::polycistronic mCherry+Podr-1::CFP] CF3762 ced-3 ( n2436 ) IV CF3656 ced-2 ( e1752 ) IV CF3660 ced-10 ( n3246 ) IV CF3662 pgrn-1 ( tm985 ) I; ced-2 ( e1752 ) IV CF3667 ced-1 ( e1735 ) I CF3672 pgrn-1 ( tm985 ) ced-1 ( e1735 ) I CF3675 pgrn-1 ( tm985 ) I; ced-10 ( n3246 ) IV CF3680 ced-5 ( n1812 ) IV CF3683 pgrn-1 ( tm985 ) I; ced-7 ( n1892 ) III CF3684 pgrn-1 ( tm985 ) I; ced-5 ( n1812 ) IV CF3685 pgrn-1 ( tm985 ) I; ced-6 ( n1813 ) III CF3687 pgrn-1 ( tm986 ) I; muIs211[Pegl-3::huPGRN::polycistronic mCherry+Podr-1::CFP] Line 1 CF3688 pgrn-1 ( tm986 ) I; muIs211[Pegl-3::huPGRN::polycistronic mCherry+Podr-1::CFP] Line 2 CF3762 ced-3 ( n1949 ) IV CF3802 ced-3 ( n717 ) IV; zcIs4[Phsp-4::gfp] V CF3808 trk-1 ( tm3985 ) X CF3809 trk-1 ( tm4054 ) X CF3817 pgrn-1 ( tm985 ) I; trk-1 ( tm3985 ) X CF3818 pgrn-1 ( tm985 ) I; trk-1 ( tm4054 ) X CF3821 trf-1 ( nr2014 ) III CF3833 pgrn-1 ( tm985 ) I; trf-1 ( nr2014 ) III CF3879 pgrn-1 ( tm985 ) I; pmk-1 ( km25 ) IV CF3881 pgrn-1 ( tm985 ) I; ced-9 ( n1950gf ) III FX494 abi-1 ( tm494 ) III KU25 pmk-1 ( km25 ) IV MT2547 ced-4 ( n1162 ) III MT4433 ced-6 ( n1813 ) III MT4982 ced-7 ( n1892 ) III MT4770 ced-9 ( n1950gf ) III MT7384 ced-4 ( n1162 ) ced-9 ( n2812lf ) III MT16077 ced-1 ( n2091 ) I; abl-1 ( n1963 ) X MT19956 ced-6 ( n2095 ) III; abl-1 ( ok171 ) X OS4023 pqn-41 ( ns294 ) III RB829 abi-1 ( ok640 ) III XR1 abl-1 ( ok171 ) X To generate a C . elegans pgrn-1 rescue construct , full-length pgrn-1a+TAA stop codon and its endogenous 0 . 5 kB upstream promoter were cloned into a Gateway polycistronic mCherry vector ( courtesy K . Ashrafi lab , UCSF ) . The resulting plasmid ( Ppgrn-1::pgrn-1+TAA::polycistronic mCherry ) expresses both progranulin and mCherry and functions as a full length rescuing construct when expressed in the pgrn-1 mutant . To generate a human progranulin rescue strain , the pan-neuronal egl-3 promoter ( courtesy of K . Ashrafi lab ) and the human progranulin cDNA sequence were cloned into a Gateway polycistronic mCherry vector ( Pegl-1::human PGRN::polycistronic mCherry ) . The constructs were microinjected separately into the gonads of day 1 adult C . elegans . Stable monogenic lines were isolated and analyzed using Leica fluorescent , Zeiss Axioplan 2 or Nikon Spectral Confocal microscopes . Extrachromosomal arrays were integrated by UV irradiation by the method of C . Frank et al . [82] and outcrossed at least 5 times to our lab's wild-type N2 control strain . For thermal and osmotic stress assays on Day 1 worms , L4-stage animals were picked and grown at 20°C overnight . For thermal stress assays , worms were moved to a 35°C incubator for 12 hours and then scored for survival . Osmotic stress assays were performed by the method of Lamitina et al . [83] with the following modifications: worms were fed OP50 bacteria , worms were cultured at 20°C prior to the assay , and assays were performed on NG-based plates at 20°C with increasing amounts of NaCl added as indicated . For paraquat stress assays , individual animals were placed in 96-well plates with 100 µL of 250 µM methyl viologen ( paraquat , Sigma-Aldrich ) dissolved in M9 and scored for movement every 1 hour at 25°C . For genotoxic stress assays , day 1 adult animals were transferred to unseeded plates and treated with 1200 J/m2 UV light in a Stratalinker 1800 ( Stratagene ) . Animals were then scored for survival every 24 hours . Pathogen stress was performed by transferring worms to plates seeded with P . aeruginosa or S . enterica starting at day 1 and scoring each subsequent day for survival . In all assays , animals that failed to move in response to a gentle touch with a metal pick were scored as dead . For ER stress assays , synchronized eggs were transferred to plates containing 0 , 1 , 2 or 5 µg/mL of tunicamycin ( EMD Chemicals ) . After 3 days , animals that developed to the L4 stage were quantified . Figures show fraction of animals that develop to L4 stage normalized to percent hatching on 0 µg/mL tunicamycin for each strain . Statistical analyses were performed in GraphPad Prism statistical package with tests as indicated in figure legends . Wild-type , ced-1 ( e1735 ) , ced-2 ( e1752 ) and ced-3 ( n717 ) strains were grown at 20 degrees Celsius ( C ) , then picked to fresh OP50 at the L4 stage and shifted to 25 degrees C . Subsequent lifespan analysis was done at this temperature . Animals were transferred every day to fresh plates until progeny production ceased . Animals that crawled off the plate , exploded , bagged , or became contaminated were censored . GraphPad Prism was used to calculate mean life spans and perform statistical analyses . P values were determined using log-rank ( Mantel-Cox ) statistics . C . elegans eggs were obtained by bleaching , then plated onto E . coli OP50 and allowed to develop at 20°C to day 1 of adulthood . At this point , positive controls were exposed to 5 mg/ml tunicamycin for 5 hours while all other worms were left untreated . After washing animals off plates , Trizol was added and samples were frozen in liquid nitrogen . Animals were lysed in a Mini-Beadbeater ( Biospec products ) for 10 minutes at the maximal setting . Total RNA was isolated using a phenol/chloroform extraction and DNA contamination was removed with DNA-free treatment ( Ambion ) . cDNA was synthesized ( iScript ) using oligo ( dT ) primers and RT-PCR was performed using primers that amplify an ∼200 bp unspliced transcript and an ∼180 bp spliced transcript . Forward primer sequence: 5′ ctacgaagaagaagtcgtcgg 3′ and reverse primer sequence: 5′ ttcttgttgcgatccatgtg 3′ . RT-PCR products were analyzed by running them out on a 3% agarose gel stained with ethidium bromide . Animals expressing the Phsp-4::gfp transgene were anaesthetized on agarose pads containing 2 . 5 mM levamisole . Whole worm images were taken using a Retiga EXi Fast1394 CCD digital camera ( QImaging , Burnaby , BC , Canada ) using the 5× objective on a Zeiss Axioplan 2 compound microscope ( Zeiss Corporation , Germany ) . Each image was taken so that the intestine was in focus and exposure time was calibrated to minimize number of saturated pixels for the set of animals . Images within each experiment were acquired using identical settings and exposure times to allow direct comparisons . Fluorescence intensity was measured by outlining the entire worm . Openlab 4 . 0 . 2 software ( Improvision , Coventry , UK ) was then used to quantify total intensity of each pixel in the selected area . Measurements were obtained by subtracting the minimum intensity from the mean intensity and taking the average of these calculations for 8–10 animals per time-point . Total RNA was isolated from each of the strains pgrn-1 ( tm985 ) , ced-3 ( n717 ) , ced-1 ( e1735 ) and wild-type ( N2E ) using a phenol/chloroform extraction , and DNA contamination was removed with DNA-free treatment ( Ambion ) . Samples were extracted in quadruplicates ( four biological replicates for each strain ) , for a total of 16 samples . Total RNA was quantified using the RiboGreen assay and RNA quality was checked using an Agilent Bioanalyzer ( Agilent ) . RNA Integrity Numbers ( RINs ) were >8 in all the samples . Libraries for RNA-seq were prepared using the Illumina TruSeq library preparation protocol ( Illumina Inc ) , multiplexed into a single pool and sequenced using an Illumina HiSeq 2500 sequencer across 4 lanes of 2 Rapid Run SR 1×50 flow cells . After demultiplexing , we obtained between 13 and 32 million reads per sample , each one 50 bases long . Quality control was performed on base qualities and nucleotide composition of sequences . Alignment to the C . elegans genome ( ce10 ) was performed using the STAR spliced read aligner ( PMID 23104886 ) with default parameters . Additional QC was performed after the alignment to examine the following: level of mismatch rate , mapping rate to the whole genome , repeats , chromosomes , and key transcriptomic regions ( exons , introns , UTRs , genes ) . Between 92 and 93% of the reads mapped uniquely to the worm genome . Total counts of read-fragments aligned to candidate gene regions within the C . elegans reference gene annotation were derived using HTS-seq program and used as a basis for the quantification of gene expression . Only uniquely mapped reads were used for subsequent analyses . Following alignment and read quantification , we performed quality control using a variety of indices , including sample clustering , consistency of replicates , and average gene coverage . Differential expression analysis was performed using the EdgeR Bioconductor package ( 19910308 ) , and differentially expressed genes were selected based on False Discovery Rate ( FDR Benjamini Hochberg adjusted p- values ) estimated at ≤5% . Clustering and overlap analyses were performed using Bioconductor packages within the statistical environment R ( www . r-project . org/ ) . Gene Ontology annotation was performed using DAVID ( david . abcc . ncifcrf . gov/ ) . DAF-16 dependent genes were curated from published reports [84] , [85] and Wormmart annotation ( http://caprica . caltech . edu:9002/biomart/martview ) . | As an animal interacts with its environment , it invariably encounters stressful conditions such as extreme temperatures , drought , UV exposure and harmful xenobiotics . Since the ability to respond appropriately to stressful stimuli is paramount to survival , organisms have developed sophisticated stress response programs . Some stressful conditions cause damaged cells to commit suicide ( undergo apoptosis ) , whereas others cause the entire organism to develop mechanisms to resist environmental stress . Studying the small roundworm C . elegans , we find that these two responses are somehow linked: perturbing the mechanisms that allow cells to undergo apoptosis changes the whole animal's response to environmental stress . In fact , perturbing the apoptosis machinery in any way—through mutations that prevent apoptosis altogether , or through mutations that either slow or accelerate the clearance of dying cells—causes the animal to become more stress resistant . Together our findings raise the possibility that the animal may have a way of detecting defects in the normal programmed cell death pathway , and that in response it induces a new program that protects itself from a harsh environment . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [] | 2013 | A Shift to Organismal Stress Resistance in Programmed Cell Death Mutants |
Notch signalling is a fundamental pathway that shapes the developing embryo and sustains adult tissues by direct communication between ligand and receptor molecules on adjacent cells . Among the ligands are two Delta paralogues , DLL1 and DLL4 , that are conserved in mammals and share a similar structure and sequence . They activate the Notch receptor partly in overlapping expression domains where they fulfil redundant functions in some processes ( e . g . maintenance of the crypt cell progenitor pool ) . In other processes , however , they appear to act differently ( e . g . maintenance of foetal arterial identity ) raising the questions of how similar DLL1 and DLL4 really are and which mechanism causes the apparent context-dependent divergence . By analysing mice that conditionally overexpress DLL1 or DLL4 from the same genomic locus ( Hprt ) and mice that express DLL4 instead of DLL1 from the endogenous Dll1 locus ( Dll1Dll4ki ) , we found functional differences that are tissue-specific: while DLL1 and DLL4 act redundantly during the maintenance of retinal progenitors , their function varies in the presomitic mesoderm ( PSM ) where somites form in a Notch-dependent process . In the anterior PSM , every cell expresses both Notch receptors and ligands , and DLL1 is the only activator of Notch while DLL4 is not endogenously expressed . Transgenic DLL4 cannot replace DLL1 during somitogenesis and in heterozygous Dll1Dll4ki/+ mice , the Dll1Dll4ki allele causes a dominant segmentation phenotype . Testing several aspects of the complex Notch signalling system in vitro , we found that both ligands have a similar trans-activation potential but that only DLL4 is an efficient cis-inhibitor of Notch signalling , causing a reduced net activation of Notch . These differential cis-inhibitory properties are likely to contribute to the functional divergence of DLL1 and DLL4 .
The Notch signalling pathway mediates local interactions between adjacent cells and thereby regulates numerous developmental processes in a wide variety of different tissues throughout the animal kingdom [reviewed in 1–7] . The Notch gene of Drosophila and its vertebrate homologues encode large transmembrane proteins that act as receptors at the surface of the cell . They interact with transmembrane ligand proteins on the surface of neighbouring , signal-sending cells ( i . e . in trans ) encoded by the Delta and Serrate ( called Jagged in vertebrates ) genes . Upon ligand binding , the intracellular domain of Notch ( NICD ) is proteolytically released , translocates to the nucleus , interacts with the transcriptional regulator Suppressor of Hairless ( [Su ( H ) ]; CSL proteins in vertebrates ) and activates the transcription of downstream target genes [8–14] . Ligands coexpressed with the Notch receptor in signal-receiving cells ( i . e . in cis ) are capable of interacting with Notch and attenuate the signal strength [15–17 , reviewed in 18] . Vertebrates possess several Notch receptors and ligands . The mouse genome encodes four Notch ( NOTCH1–4 ) , three Delta ( DLL1 , DLL3 and DLL4 ) and two Jagged ( JAG1 and JAG2 ) proteins . Among the DLL proteins , only DLL1 and DLL4 function as Notch-activating ligands [19–21] . As paralogues , DLL1 and DLL4 are similar in sequence ( 47% identical plus 14% similar amino acids ) , size and domain structure [22] . Both contain a DSL domain , which is essential for the interaction with Notch [23 , 24] , as well as eight EGF-like repeats in their extracellular domain and have a short intracellular domain with a C-terminal PDZ binding motif . Dll1 and Dll4 are expressed both in discrete and overlapping patterns during embryonic development and in adult tissues of the mouse . In shared expression domains , the two ligands have redundant or different functions depending on the developmental context . An example for full redundancy is the maintenance of the crypt progenitor pool in the adult small intestine . Dll1 and Dll4 are coexpressed in crypt cells [25 , 26] and individual inactivation of either ligand has no effect on the crypt progenitor cell pool . However , simultaneous deletion of Dll1 and Dll4 leads to a complete loss of the proliferative crypt compartment and intestinal stem cells [27] . Conversely , in foetal arteries where both ligands are expressed in the vascular endothelium [26 , 28 , 29] inactivation of Dll1 causes loss of NOTCH1 activation despite the presence of DLL4 [29] suggesting that DLL4 cannot compensate for the loss of DLL1 in fetal endothelial cells . In the adult thymus , Dll1 and Dll4 are both expressed in thymic epithelial cells [26 , 30] . Here , DLL4 is the essential Notch ligand required for T-lymphopoiesis [31] and T cell development is unaltered in mice lacking DLL1 in the thymic epithelium [32] suggesting that in this context DLL1 and DLL4 are functionally distinct . This conclusion is supported by in vitro studies showing that DLL1 and DLL4 differ with respect to their binding avidity to Notch receptors on thymocytes and to the steady-state cell surface levels required to induce T cell development , DLL4 being the more effective ligand [33 , 34] as well as by biochemical studies indicating a 10-fold higher Notch binding affinity of DLL4 than DLL1 [19] . Furthermore , DLL4 but not DLL1 can induce a fate switch in skeletal myoblasts and induce pericyte markers [35] . Collectively , these individual reports of context-dependent redundant and distinct functions of coexpressed DLL1 and DLL4 raise the questions of why DLL1 and DLL4 act equally in some processes but differently in others , which mechanism or factor causes their function to vary and whether they are similar enough to replace each other in domains where only one of both DLL ligands is endogenously expressed . In early mouse embryos , expression of Dll1 and Dll4 is largely non-overlapping . Dll1 is expressed in the paraxial mesoderm beginning at E7 . 5 , in the central nervous system from E9 onwards and later on , at E13 . 5 , in arterial endothelial cells [29 , 36] . Deletion of Dll1 disrupts somite patterning and causes premature myogenic differentiation , severe haemorrhages and embryonic death after E11 [37 , 38] . Dll4 is expressed in the vascular endothelium of arteries beginning at E8 [39] but not in the somite-generating presomitic mesoderm , somites or differentiating myoblasts . Inactivation of DLL4 results in severe vascular defects leading to embryonic death prior to E10 . 5 [39 , 40] . Here , we address the functional equivalence of DLL1 and DLL4 in vivo and in vitro . We analyse Notch signalling in mice that conditionally overexpress DLL1 or DLL4 on a Dll1 null genetic background and in mice in which Dll1 is replaced by Dll4 , focussing on young embryos in which both Notch ligands have discrete endogenous expression domains . We show that DLL4 cannot replace DLL1 during somite segmentation but can partially replace DLL1 during myogenesis and fully replace DLL1 during maintenance of retinal progenitors . Cell culture assays that measure Notch activation by DLL1 or DLL4 demonstrate that DLL4 trans-activates Notch signalling similarly to DLL1 but cis-inhibits Notch signalling much more efficiently than DLL1 , partly overruling the activation by interactions in trans . Consistent with these in vitro data , we observe dominant effects on segmentation by DLL4 ectopically expressed in the presomitic mesoderm ( PSM ) . We propose that differential Notch cis-inhibition by DLL1 and DLL4 contributes to the observed tissue-dependent functional divergence of both paralogues , perhaps in combination with other factors not tested in this study .
In order to directly compare the activities of DLL1 and DLL4 in vivo , we generated mice that conditionally express either Dll1 or Dll4 under the CAG promoter from a single-copy transgene insertion in the same genomic locus . We employed an established system for integration of Cre-inducible expression constructs into the Hprt locus , the pMP8 . CAG-Stop vector ( Fig 1A; [41 , 42] ) . The unrecombined pMP8 . CAG-Stop construct expresses neomycin phosphotransferase ( neor ) from the CAG promoter . Cre-mediated recombination of two loxP sites and two mutant loxP2272 ( loxM ) sites [43] flips the gene of interest and excises neor so that the recombined construct expresses the gene of interest from the CAG promoter . 5’ and 3’ homology regions from the Hprt gene enable homologous recombination of pMP8 constructs into the Hprt locus [44] . We cloned the Dll1 and Dll4 open reading frames into the pMP8 . CAG-Stop vector , introduced both unrecombined ( i . e . neor expressing ) constructs into Hprt-deficient E14TG2a ES cells and used homologous recombinant clones to produce transgenic mice with Cre-inducible Dll1 or Dll4 ( alleles termed CAG:Dll1 and CAG:Dll4 ) . To check activation of DLL1 and DLL4 in embryos , we induced ubiquitous expression of the CAG:Dll1 and CAG:Dll4 transgene by mating our mice with mice carrying a ZP3:Cre transgene that causes site-specific recombination during oogenesis [45] . We crossed CAG:Dll1;ZP3:Cre and CAG:Dll4;ZP3:Cre females with wildtype males to obtain embryos that overexpress Dll1 or Dll4 from the zygote stage on . The transgenes are transcribed bicistronically with an IRES-Venus ( Fig 1A ) whose expression marks cells in which Cre-recombination activated the transgene . As Hprt is located on the X chromosome , hemizygous male embryos expressed Venus ubiquitously whereas heterozygous female embryos showed mosaic expression due to random X-inactivation ( Fig 1B ) . Analysis of embryo lysates on a Western blot with anti-GFP antibodies demonstrated CAG:Dll1 and -4 transgene activation at similar levels ( Fig 1C; S1A Fig; S1 Table ) . To directly compare DLL1 and DLL4 protein levels , we generated embryonic stem cells expressing HA-tagged Dll1 or Dll4 from single copy insertions of in the Hprt locus . Western blot analysis of three independent clones for each ligand confirmed similar protein levels in all clones ( average DLL1-HA , 1 . 23±0 . 33; average DLL4-HA , 0 . 88±0 . 53; Fig 1D; S1B Fig; S2 Table ) . In order to test whether DLL4 can compensate for the loss of DLL1 in mesodermal tissues of early embryos , we mated mice to combine three different transgenes: a ) CAG:Dll1 or -4 inducible transgenes; b ) two ( i . e . homozygous ) floxed alleles of endogenous Dll1 [Dll1loxP/loxP; 32] that get inactive upon recombination; and c ) a Cre transgene expressed in the primitive streak driven by a promoter derived from brachyury [T ( s ) :Cre; 46] . In offspring bearing the complete set of transgenes ( identified by genotyping PCR; see Material and Methods ) , recombination by T ( s ) :Cre simultaneously inactivates endogenous Dll1 and activates CAG:DLL1 or CAG:DLL4 expression in all mesoderm-derived tissues . As expected , inactivation of Dll1 throughout the mesoderm resulted in severe somite patterning defects characterised by loss of Uncx4 . 1 expression ( Fig 1Ed ) , a marker for caudal somite compartments [47 , 48] whose expression depends on Notch activation [46] . Expression of CAG:DLL1 in such Dll1-deficient embryos restored robust , largely regularly striped expression of Uncx4 . 1 , which expanded into cranial somite compartments in most axial regions and particularly in hemizygous male embryos ( Fig 1Eb and 1Ec ) . This rescue of somitogenesis demonstrates that expression of CAG:DLL1 ( from the Hprt locus ) is sufficient to substitute for the loss of endogenous DLL1; cranial expansion of Uncx4 . 1 is reminiscent of ectopic Notch activity [46] . In contrast , expression of CAG:DLL4 in Dll1-deficient embryos restored only very weak and irregular expression of Uncx4 . 1 and resembles Dll1loxP/loxP;T ( s ) :Cre embryos without CAG:DLL1 overexpression ( Fig 1Ee and 1Ef ) . Only two out of 16 embryos of this genotype displayed regular Uncx4 . 1 expression in the cranial most somites , which might reflect residual DLL1 activity perhaps due to delayed excision of endogenous Dll1 ( Fig 1Eg ) . The extensively defective segmentation in Dll1loxP/loxP;T ( s ) :Cre embryos with CAG:DLL4 overexpression directly shows a functional difference between DLL1 and DLL4 during early embryogenesis: DLL4 is not able to take over DLL1 function in the paraxial mesoderm during somite formation . Weak and irregular Uncx4 . 1 expression in some of these embryos suggest Notch activation at low levels . To further investigate to which degree DLL4 can compensate for the loss of DLL1 during somite patterning and in other developmental contexts , we generated mice that express DLL4 from the Dll1 locus instead of endogenous DLL1 . To replace endogenous Dll1 with Dll4 , we applied a knock-in strategy inserting a Dll4 mini gene into the first and second exons of Dll1 ( Fig 2A ) . Production of DLL4 protein of the correct size from the Dll4 mini gene was confirmed by Western blot analysis of lysates of CHO cells transiently expressing the Dll4 mini gene ( S2 Fig ) . We generated mice carrying the Dll4 mini gene in the Dll1 locus , referred to as Dll1Dll4ki . As a control , we used the analogous knock-in of a Dll1 mini gene into the Dll1 locus ( Fig 2A bottom; Dll1tm2Gos , here referred to as Dll1Dll1ki ) , which was identical to the Dll4 mini gene with regard to its exon/intron structure , intron sequences and the 5' and 3' UTRs but encoded DLL1 . Homozygous Dll1Dll1ki mice were viable and fertile and appeared phenotypically normal indicating that the Dll1 mini gene can functionally substitute the endogenous Dll1 gene [37] . Heterozygous Dll1Dll4ki/+ mice ( containing one endogenous copy of Dll1 and one copy of Dll4ki ) were viable and fertile and showed no obvious phenotype except for kinky tails ( Fig 2Ba–2Bc , arrow; penetrance 89%; n = 48 ) , a phenotype indicative of irregular somitogenesis rarely observed in Dll1Dll1ki homozygotes or Dll1 null ( Dll1lacZ ) heterozygotes ( penetrance 15%; n = 23 and 53 , respectively; Fig 2Ba’–2Bc’ ) . In contrast to homozygous Dll1Dll1ki , no homozygous Dll1Dll4ki mice were obtained after birth . At E15 . 5 , Dll1Dll4ki homozygotes exhibited short body axes , truncated tails and were oedematic ( Fig 2C; arrow points at tip of tail ) resembling foetuses with severely reduced DLL1 function [37] . Correct expression of Dll4 in the presomitic mesoderm ( PSM ) of Dll1Dll4ki embryos was confirmed by in situ hybridisation using probes specific for the Dll4 ORF or the 3‘UTR ( Dll1 exon 11 ) common to Dll1Dll1ki and Dll1Dll4ki alleles ( Fig 2Db , 2Dc and 2Df , black arrowheads ) . In situ hybridisation with a specific Dll1 ORF probe confirmed the absence of Dll1 transcripts in Dll1Dll4ki homozygotes ( Fig 2Dj , red arrowhead ) . Homozygous Dll1Dll4ki embryos showed strong expression of Dll4 in the neural tube ( Fig 2Dc , white arrow ) , reflecting activation of the Dll1 promoter in this region [50 , 51] . Northern blot analysis of Dll1Dll1ki and Dll1Dll4ki homozygous embryos indicated equal levels of transcription of the transgenes ( Fig 2E ) . In Dll1Dll4ki/Dll4ki embryos , ectopic DLL4 protein was detected at the plasma membrane of PSM cells ( Fig 2Fa–2Fc ) . Likewise , DLL1 protein was detected at the surface of PSM cells in homozygous Dll1Dll1ki embryos ( Fig 2Fj–2Fl ) , confirming that DLL4 and DLL1 protein is generated from their mini genes and targeted to the plasma membrane in vivo . Taken together , these data show that Dll1Dll4ki mice indeed express Dll4 instead of Dll1 from the Dll1 locus at comparable levels and confirm our previous observation that DLL4 is unable to support proper mouse development in the absence of endogenous DLL1 . Cranial-caudal somite patterning critically depends on DLL1-mediated Notch signalling [38 , 46 , 52] . We analysed if DLL4 can functionally replace DLL1 in this process in homozygous Dll1Dll4ki embryos . Unlike embryos that contained at least one wildtype or Dll1Dll1ki allele , homozygous Dll1Dll4ki embryos displayed severely reduced and irregular Uncx4 . 1 expression ( Fig 3A ) , which indicates disrupted somite patterning and reduced Notch activity in the PSM due to the inability of DLL4 to replace DLL1 . Consistent with defective somite formation and the shortened body axis observed in E15 . 5 foetuses , Dll1Dll4ki/Dll4ki axial skeletons were severely disorganised ( Fig 3B ) . Therefore , expression of DLL4 from the Dll1 locus does not cause a significant rescue of the Dll1 somitogenesis phenotype . Remarkably , as anticipated by the kinky tail phenotype of heterozygous Dll1Dll4ki/+ adults ( Fig 2B ) , E18 . 5 Dll1Dll4ki/+ skeletons reveal fusions of dorsal ribs and malformations of individual vertebrae in various regions of the vertebral column ( e . g . Fig 3Bd , red arrowheads; 8/12 E18 . 5 Dll1Dll4ki/+ skeletons displayed apparently irregular vertebrae ) . Additional examination of seven Dll1Dll4ki/+ adult skeletons uncovered fused ribs and/or irregular segments in the tail of all preparations ( S3 Fig ) . A single Dll1 allele is sufficient to support regular segmentation and Dll1lacZ/+ mice form essentially normal skeletons [53] . Our consistent finding of skeletal irregularities in Dll1Dll4ki/+ mice indicates subtle disturbances during segmentation and suggests that the Dll1Dll4ki allele has a dominant effect on segmentation . Processes other than somitogenesis in the developing embryo that depend on DLL1–Notch signalling include myogenesis [37] and retinal development [51] . Embryos lacking DLL1 display excessive differentiation of myoblasts , which exhausts the progenitor pool and leads to severely reduced or absent skeletal muscles [37] . Homozygous E9 . 5 Dll1Dll4ki embryos showed transient upregulation of the myocyte marker Myogenin [54] as also observed in homozygous Dll1lacZ embryos ( Fig 3C , arrowheads; [37] ) . At E15 . 5 , they had significantly less skeletal muscle tissue than wildtype or homozygous Dll1Dll1ki foetuses but clearly more skeletal muscle tissue than Dll1 null mutants ( Dll1lacZ ) as shown for the intercostal muscles , the diaphragm and forelimbs by anti-MHC antibody staining of sectioned foetuses ( Fig 3D–3F , arrowheads ) . These results indicate that DLL4 can partially substitute DLL1 during muscle cell differentiation and Dll1Dll4ki behaves like a hypomorphic Dll1 allele . In the embryonic neural retina , Dll1 and Dll4 are sequentially expressed and can both function to maintain proliferating progenitors , while they have different functions in retinal fate diversification [51 , 55] . In contrast to myogenesis , DLL4 can fully replace DLL1 function in maintaining neuronal progenitors in the embryonic retina . Whereas Dll1 mutants show a striking disruption of the retinal neuroepithelium with formation of rosettes ( Fig 4A ) , due to premature differentiation of retinal progenitors [51] , both Dll1Dll1ki/Dll1ki and Dll1Dll4ki/Dll4ki retinas have a normal neuroepithelial organisation with a clear stratification of Chx10+ progenitors and p27+ differentiating neurons ( Fig 4B ) . Moreover , we find that similar numbers of early born retinal neurons [retinal ganglion cells ( RGCs ) and amacrine cells] are present in Dll1Dll1ki and Dll1Dll4ki retinas ( Fig 4C and 4D; n≥4 retinal sections ) , confirming that DLL1 and DLL4 functions are interchangeable in regulating early retinal neurogenesis . We have further analysed DLL4 expression in Dll1Dll4ki/Dll4ki retinas and found it recapitulates the broader Dll1 expression pattern , with the transgenic protein expressed at similar levels as endogenous DLL4 in the retinal neuroepithelium ( compare Fig 4Ea–4Ec with 4Eb–4Ed ) . Together , these results offer further evidence that the Dll4 transgene is fully functional in Dll1Dll4ki/Dll4ki embryos . The extent of the functional equivalence of DLL1 and DLL4 depends on the developmental context . To investigate the functional difference between DLL1 and DLL4 in vitro , we performed co-culture experiments by mixing cells expressing NOTCH1 receptor or DLL ligands and measured Notch activation with a reporter in the receptor-expressing cells . Specifically , we used HeLa cells that express both the NOTCH1 receptor ( stable HeLa-N1 cells; [10] ) and a transient Notch activity reporter based on an RBP-Jk promoter-driven Luciferase [56] with CHO cells stably expressing Flag-tagged DLL1 or DLL4 ligands . To ensure comparability of results , we integrated single copies of Dll1 or Dll4 ORFs under the control of the CMV promoter into the identical genomic locus of CHO cells by adopting a site-directed attP/attB recombination system ( Fig 5A top; S4 Fig; [57] ) . We established CHO cells with a pre-inserted , randomly integrated single attP site ( termed CHOattP; uniqueness of this attP site was confirmed by Southern blot analysis; S4A and S4B Fig ) and recombined Dll1 or Dll4 ORFs into this site ( cell lines termed CHOattP-DLL1 and CHOattP-DLL4; Fig 5A bottom left ) . Consistent with the expression from the same genomic locus , independent CHOattP-DLL1 ( B5 , C6 ) and CHOattP-DLL4 ( B5 , D3 ) clones expressed DLL1 and DLL4 protein at similar levels ( Fig 5B; n = 4 lysates of each clone; S5A Fig , S3 Table , S5B Fig , S4 Table ) and cell surface representation of DLL1 and DLL4 was similar in all lines ( ~40%; Fig 5C; n≥3 biotinylation assays; S5C and S5D Fig , S5 Table ) . Likewise , half-lives of DLL1 and DLL4 proteins were similar , DLL4 being slightly more stable ( S5E and S5F Fig ) . Co-culture of HeLa-N1 with either CHOattP-DLL1 or CHOattP-DLL4 ( schematically shown in Fig 5A bottom ) led to a >10-fold increase of Notch activity as compared to co-cultures of HeLa-N1 with CHOattP cells that did not express transgenic DLL1 or DLL4 ( Fig 5D; n = 3 ) confirming that all transgenes were functional . DLL4 trended to activate Notch more strongly than DLL1 ( including clone CHOattP-DLL4 B5 whose protein level was slightly reduced in Fig 5B ) ; the difference between individual clones was not statistically significant in these experiments and partly significant in similar experiments with other clones ( S6A and S6G Fig ) . Next , we tested whether coexpression of further factors ( LFNG , JAG1 ) in our cell culture system differently alters Notch activation by DLL1 or DLL4 and thereby provides a plausible explanation for the distinct phenotypes . The glycosyltransferase LUNATIC FRINGE ( LFNG ) , which is expressed in the PSM , is able to modify NOTCH in the trans-Golgi [58 , 59] and thereby modulates receptor activation . The Notch ligand JAG1 is expressed in forming somites [60 , 61] and can act as a competitive inhibitor of DLL ligands [62 , 63] . We performed co-culture assays with HeLa-N1 cells transiently expressing LFNG-HA ( S6A–S6C Fig ) or with CHOattP-DLL1 and CHOattP-DLL4 cells coexpressing JAG1 ( S6D–S6F Fig ) and found no statistically significant changes in Notch activation . Also , different glycosylation patterns of the ligands’ extracellular domain could contribute to differences in their activity . To test this possible influence , we treated co-cultures with tunicamycin to prevent N-glycosylation . Blocking N-glycosylation reduced the activity of DLL4 in cultured cells significantly , but not below DLL1 activity ( S6G and S6H Fig ) , suggesting that distinct N-glycosylation is an unlikely cause for the observed differences between both ligands . Collectively , our results do not reveal a difference in the trans-activation potential of DLL1 and DLL4 that could explain the different segmentation phenotypes of our transgenic DLL1- or DLL4-expressing mice . We modified the co-culture assay by ( transiently ) expressing the ligands in the HeLa-N1 cells instead of in the CHO cells ( Fig 5E , S7 Fig ) . In this setting , DLL ligands ( expressed in HeLa-N1 cells ) can trans-activate Notch in neighboring HeLa-N1+DLL cells ( schematically shown in Fig 5Ec or in detail in S7A Fig ) ; in addition , they can interact with Notch expressed in the same cell , i . e . in cis . When co-culturing HeLa-N1 cells expressing DLL1 with empty CHO cells ( Fig 5Ea , S7 Fig ) , activation of Notch signalling was significantly increased as compared to a co-culture of HeLa-N1 cells expressing no transgenic DLL ligand with empty CHO cells ( Fig 5Ea’; compare light grey bar “+DLL1” and white bar; n = 6; numbers are normalised to white bar ) . Intriguingly , in co-cultures of HeLa-N1 cells expressing DLL4 ( with empty CHO cells ) , Notch activation was significantly lower than with DLL1 ( Fig 5Ea’; compare dark grey bar “+DLL4” and light grey bar “+DLL1”; n = 6 ) . Given the similar trans-activation potential of DLL1 and DLL4 ( Fig 5D; S6A and S6G Fig ) , a likely explanation for the different levels of Notch activation is a higher cis-inhibitory potential of DLL4 than of DLL1 . In order to facilitate the analysis of cis-inhibition , we repeated the experiments shown in Fig 5Ea and 5Ea’ with the only modification of expressing a DLL ligand in the CHO cells ( CHOattP-DLL1 ) so as to enhance the level of Notch activation ( Fig 5Eb and 5Eb’ ) . In the experiments both without any transgenic DLL ligand in HeLa-N1 and with DLL1 in HeLa-N1 , co-culture with CHOattP-DLL1 cells caused a >10-fold increase of Notch activation . In the HeLa-N1 cells with DLL4 , Notch activation was significantly less , i . e . about 5-fold increased ( Fig 5Eb; n = 6; all numbers in Fig 5Ea’ and 5Eb’ are normalised to the left bar in a’ , i . e . HeLa-N1 without transgenic DLL co-cultured with empty CHO cells , set to 1 ) . These results support cis-inhibition of Notch by DLL4 resulting in a strong reduction of net Notch activation . In contrast , DLL1 does not cis-inhibit Notch in this assay ( no significant difference between white and light grey bar in Fig 5Eb’ ) . To approximate the setting of the embryonic cranial PSM in which every cell expresses both DLL1 and NOTCH1 [64] , we also analysed pure cultures of HeLa-N1 cells expressing either no transgenic DLL or ( transient ) DLL1 or DLL4 ( Fig 5Ec ) . Expression of DLL1 enhanced Notch activation ~15-fold whereas expression of DLL4 increased Notch activation only <5-fold which was not significantly different from HeLa-N1 cells without transgenic ligand ( Fig 5Ec’; n = 6; numbers in Fig 5Ec’ are normalised to the left bar in 5Ec , i . e . culture of HeLa-N1 without transgenic DLL , set to 1 ) . These data show that cis-inhibition by DLL4 partially overrides trans-activation , and reduces Notch activation to <30% in an in vitro setting modeling the arrangement of ligand and receptor molecules in the PSM . A conceivable alternative explanation of our in vitro results , which show attenuated Notch signalling when NOTCH and DLL4 are coexpressed , could be a reciprocal mechanism , i . e . cis-inhibition of DLL4 by NOTCH1 [18] . To test this possibility , we modified our first Notch activation assay ( Fig 5A bottom , 5D ) by transiently coexpressing NOTCH1 ( NOTCH1deltaC , see Methods ) in CHOattP-DLL1 and CHOattP-DLL4 cells . In co-cultures with HeLa-N1 cells containing the reporter ( Fig 5Fa and 5Fb , S8 Fig ) , both ligands activated NOTCH in HeLa-N1 >10-fold irrespective of the presence of NOTCH1 in the CHO cells ( Fig 5Fa’ and 5Fb’; n = 3 ) and we measured no significant difference between trans-activation by DLL1 or DLL4 as before ( Fig 5D ) . These data indicate that NOTCH1 does not cis-inhibit its ligands . In summary , our cis-inhibition assays ( Fig 5E ) reveal a functional difference between DLL1 and DLL4 that was not evident in the trans-activation assays ( Fig 5D and 5F ) : DLL4 , but not DLL1 , is a potent cis-inhibitor of NOTCH1 and cis-inhibition by DLL4 can significantly reduce Notch activation . Our in vitro results are consistent with our in vivo data: they can explain both why DLL4 appears to be a weaker activator of Notch signalling than DLL1 during somitogenesis in our transgenic mice and why transgenic DLL4 has a dominant effect on segmentation in Dll1Dll4ki/+ mice ( see Discussion and Fig 6 ) . We propose that in the PSM , DLL1 is a more efficient net activator of Notch than ( ectopic ) DLL4 because it does not efficiently cis-inhibit Notch . In order to identify the protein domain that mediates cis-inhibition by DLL4 , we cloned chimeric Dll1 and Dll4 ORFs by swapping extracellular domains ( resulting in DLL4-ICD+TM/DLL1-ECD , termed DLL4-DLL1ECD , or DLL1-ICD+TM/DLL4-ECD , termed DLL1-DLL4ECD; ICD , intracellular domain; TM , transmembrane domain; ECD , extracellular domain; Fig 5G top ) . We introduced the chimeric Dll1-4 ORFs transiently into HeLa-N1 cells and performed co-culture assays analogous to the cis-inhibition experiments with non-chimeric DLL1 and DLL4 shown in Fig 5E ( Fig 5G; S9 Fig ) . Measurement of Notch activity showed similarity between DLL1 and DLL4-DLL1ECD as well as between DLL4 and DLL1-DLL4ECD ( Fig 5Ga’ , 5Gb’ and 5Gc’; n = 6; compare with Fig 5Ea’ , 5Eb’ and 5Ec’ ) . Particularly the statistically significant differences between bars in Fig 5Gb’ clearly indicate that DLL4-DLL1ECD enhances , but DLL1-DLL4ECD reduces Notch activation by DLL1 . As both chimeric ligands localise to the cell surface ( S9B and S9C Fig; S6 Table ) and are able to trans-activate Notch in a range similar to DLL1 and DLL4 ( S9D Fig ) these results show that cis-inhibition is mediated by the extracellular domain of DLL4 . This observation is consistent with studies that showed that the DSL domain as well as EGF repeats 4–6 of Serrate are essential for cis-inhibition in Drosophila although these EGF repeats are not well conserved between Serrate and Delta ligands [24 , 65 , 66] . Analysis of Notch activation by chimeric proteins in which smaller domains of the extracellular regions are swapped will help to precisely map the cis-inhibitory domain in DLL4 .
Mesodermal expression of DLL1 and DLL4 from the Hprt locus on a Dll1 mutant background caused different phenotypes providing first hints that DLL1 and DLL4 are functionally different during early embryogenesis: CAG:DLL1 largely rescued the somitogenesis defects ( Fig 1Eb and 1Ec ) indicating that expression from this heterologous locus is strong enough to rescue the Dll1 null segmentation phenotype . In contrast , CAG:DLL4 expressed from the same locus failed to sufficiently activate Notch ( Fig 1Ee–1Eg ) . Our Dll1Dll4ki knock-in data independently confirm and extend the CAG:Dll4 expression data , corroborating the inability of DLL4 to substitute for DLL1 function in the PSM ( compare Figs 1Ee , 1Ef and 3Ae ) . We also show that the level of redundancy depends on the developmental process: there is essentially no redundancy during segmentation ( Fig 3Ac , 3Ae , 3Bc and 3Be ) , partial redundancy in myoblast differentiation ( Fig 3C–3F ) and full redundancy in retinal progenitor maintenance ( Fig 4 ) . The effects on myogenesis and retinal development confirm that functional DLL4 is expressed from the Dll1Dll4ki allele ( Figs 3De , 3Ee , 3Fe and 4B–4D ) . Different protein levels of DLL1 and DLL4 are unlikely to account for the different phenotypes observed . Both proteins are expressed from identical genomic sites and the comparison of levels of bicistronic GFP ( Fig 1C ) , transcripts ( Fig 2E ) and HA-tagged proteins ( Fig 1D ) confirm similar expression levels . Consistently , we find similar steady state levels , surface representation and half-lives of both ligands in CHO cells ( Fig 5B and 5C; S5D–S5F Fig ) . Furthermore , immunohistochemistry using anti-DLL4 antibodies show similar levels of endogenous DLL4 and Dll1-driven DLL4ki expression in the retina ( Fig 4E ) as well as similar localisation of ectopic DLL4 and endogenous DLL1 at the cell surface within the PSM ( Fig 2F ) . In our mouse models , we expressed untagged Dll1 and Dll4 transgenes to avoid alteration of protein function by the tag . As a consequence , we were unable to directly compare DLL1 and DLL4 levels in vivo and therefore cannot exclude small differences that may have contributed in part to the observed phenotype; strong differences are not indicated in the controls mentioned above . Also , it is very unlikely that DLL1 or DLL4 have functions other than interacting with and activating Notch receptors . Although it has been previously suggested that the intracellular domain of DLL1 may influence gene transcription in the signal sending cell [67 , 68] , we were unable to reproduce these in vitro results and showed that overexpression of the intracellular domain of DLL1 does not cause a phenotype in mice [42] . Collectively , the distinct ability to cis-inhibit Notch is a plausible explanation for the context-dependent DLL1-DLL4-divergence . The ability of vertebrate DLL homologues to cis-inhibit Notch has been suggested before: overexpression of truncated DLL1 proteins lacking the intracellular domain in Xenopus , chicken and mouse embryos show dominant-negative effects on Notch signalling that are likely to be caused by cis-inhibition of Notch [53 , 69 , 70] . In primary human keratinocyte cultures , expression of DLL1 ( and truncated DLL1T ) renders cells unresponsive to Delta signals from neighbouring cells and controls differentiation of stem cells [71] . Our data show for the first time that DLL4 is a strong cis-inhibitor of Notch signalling , far stronger than DLL1 . We have examined cis-inhibition in various types of cultures , in NOTCH- and DLL-expressing HeLa cells with and without co-culture of empty or DLL-expressing CHO cells and with chimeric DLL1-4 proteins ( Fig 5E and 5G ) . Furthermore , we have tested cis-inhibition of DLL1 and DLL4 ligands by NOTCH1 ( Fig 5F ) . All those assays consistently show a strong reduction of Notch signalling by DLL4 when coexpressed with NOTCH1 . In our assays , DLL1 had no obvious cis-inhibitory effect ( Fig 5Eb’; n = 6 ) , which differs from earlier reports showing that vertebrate DLL1 proteins can cis-inhibit NOTCH1 [20 , 72–74] . This is likely due to different assay conditions: in these previous studies , DLL1 was derived from different vertebrate species or differently tagged , or different cell systems or higher ligand concentrations were used . In studies in which cis-inhibition of Notch by Delta and Serrate was compared , Delta displayed a relatively weaker cis-inhibitory potential [75 , 76] . The ability for strong cis-inhibition resides in the extracellular domain of DLL4 ( Fig 5G ) that physically interacts with the Notch extracellular domain . Possible causes for the higher cis-inhibitory potency of DLL4 as compared to DLL1 include a potentially higher Notch cis-binding affinity of DLL4 as determined for the trans-interaction in vitro [34] or different glycosylation patterns in the extracellular domains of DLL1 and DLL4 ( DLL4 contains an additional O-fucosylation site in EGF5 and four additional N-glycosylation sites , three of which reside in the N-terminal domain , which is essential for Notch activation; e . g . [19]; sites predicted by www . cbs . dtu . dk/services/NetOGlyc ) . Our in vitro findings provide a possible explanation why DLL1 supports regular somite formation whereas DLL4 with its reduced net Notch activation potential is unable to do so . Heterozygous Dll1Dll4ki/+ mice consistently exhibit kinky tails and irregular vertebrae ( Figs 2B and 3Bd; S3 Fig ) despite the presence of one wildtype Dll1 allele , which should be able to support regular somitogenesis [53] . This finding strongly supports an in vivo inhibitory effect of DLL4 in the PSM , in which Dll4 is ectopically expressed at physiological levels ( similar to the endogenous Dll1 levels; Fig 2E ) . Skeletal malformations observed in Dll1Dll4ki/+ mice are distinct from phenotypes observed upon mild overexpression of Dll1 in the paraxial mesoderm that include fused or split vertebral bodies and reduction of costal heads of ribs [77] . This supports the view that cis-inhibitory DLL4 acts in a dominant-negative manner partially overruling Notch activation by wildtype DLL1 causing axial skeleton defects in Dll1Dll4ki/+ mice , similar to the effect of a truncated dominant-negative form of DLL1 expressed in the paraxial mesoderm [53] . An alternative explanation for the dominant segmentation effect in heterozygous Dll1Dll4ki/+ mice could be a competition between DLL4 and DLL1 for NOTCH binding sites with DLL4 binding NOTCH more efficiently but activating it less efficiently than DLL1; although DLL1 has not been shown to be a more potent activator of NOTCH in vitro ( Fig 5D; S6A , S6D and S6G Fig; [33 , 34] ) we cannot exclude that this is the case in certain cellular contexts . cis-Inhibition has been demonstrated to play a physiological role during fly development at the dorso-ventral border of the wing imaginal disc [15 , 16] and in photoreceptor precursors of the eye [17] . In vertebrates , the occurrence of cis-inhibition under physiological conditions is less clear but probable ( see previous section ) . We did not observe apparent phenotypes in Dll1Dll4ki/+ mice that indicate dominant-negative effects of DLL4 outside the PSM . However , we hypothesise that cis-inhibition may occur in the foetal arterial endothelium , where DLL1 , DLL4 and NOTCH1 are coexpressed and where loss of DLL1 abolishes NOTCH1 activation [29] , possibly due to cis-inhibition by DLL4 . The PSM is particularly well suited to test the functionality of Notch ligands in vivo because DLL1 is the only activating ligand endogenously expressed in this tissue [78] and Dll4 mutants have no somitogenesis phenotype [39 , 40] , so the analysis of Notch signalling is not complicated by the presence of several activators or confounded by composite phenotypes . However , two receptors , NOTCH1 and NOTCH2 , are expressed in the PSM and may differ in their response to DLL1 or DLL4 binding . The situation in myoblasts and other tissues is less clear . Outside the PSM , receptor and ligand expression typically exclude each other so that cis-inhibition can occur only during the short process in which the fate as receptor- or ligand-expressing cells is established [76 , 79 , 80] . That way , cis-inhibition may also be responsible for differences between DLL1 and DLL4 observed during myogenesis ( Fig 3D–3F ) . Other reasons may contribute to or cause these differences: Firstly , further Notch receptors and ligands are expressed during myogenesis [81] . The contribution of individual Notch receptors to myogenesis is unknown but their function could vary [82] . Also , different ligands , including DLL1 and DLL4 , have been shown to activate different Notch targets depending on the cell type in vitro [83 , 84] . A future thorough analysis of the functional divergence between DLL1 and DLL4 in myoblasts should aim at identifying the involved receptors and modulators ( perhaps by in vitro analyses of myogenic or mesodermal progenitor cells including knock-down of individual factors ) in order to understand the mechanisms underlying the observed phenotype . Secondly , other processes may cause the divergence , e . g . modification of the ligands or receptors by glycosylation ( may also play a role in the PSM ) . Activation of Notch by its ligands can be modulated by Fringe proteins . While glycosylation of Notch by LFNG enhances interaction with DLL1 in C2C12 cells [85] and with DLL4 in T cells in vitro [86] , it appears to attenuate Notch signalling in the PSM [64 , 87] . However , we did not observe any shortcomings of DLL4 in the ability to trans-activate NOTCH1 compared to DLL1 when LFNG was present in the receptor-presenting cell ( S6A–S6C Fig ) . The trans-activation potential of DLL1 and DLL4 could vary under certain conditions in vivo , perhaps depending on the glycosylation status , although our in vitro assays did not reveal any difference . Finally , the different extent of the functional difference between DLL1 and DLL4 observed in the PSM and during myogenesis may reflect the fact that mild changes of DLL1 activity affect the delicate Notch signalling in the PSM more readily than outside the PSM because somite patterning appears to be particularly sensitive to reduced Notch activity [88] . In conclusion , our genetic studies revealed a context-dependent functional divergence of the NOTCH ligands DLL1 and DLL4 in mice and provide a basis for a more extensive mechanistic analysis of this divergence in future studies . These will identify the relevant protein domain ( s ) and biochemical parameters and contribute to our understanding how different combinations of receptors and ligands determine the outcome of Notch signalling .
Statistical analyses were performed using Prism software ( GraphPad ) . Luciferase measurements were analysed by one-way ANOVA and activities obtained with each protein were compared using Bonferoni’s Multiple Comparison Test with a significance level of 0 . 05 . Means for all three DLL1-and DLL4-HA clones in Fig 1D , cell counts in the retina and cell surface levels of chimeric ligands were analysed using the Student’s t-test . | Notch signalling relies on binding of a ligand to a Notch receptor , both residing on the surfaces of neighbouring cells . This interaction forwards a signal into the receptor-expressing cell , this way coordinating cells in many biological processes such as the segmentation of the axial skeleton . Mammals possess four Notch-activating ligands–including DLL1 and DLL4 -expressed in diverse , partially overlapping regions . Whether the different ligands trigger quantitatively or qualitatively distinct Notch responses is largely unknown . In order to directly compare both ligands we generated transgenic mice that express DLL1 or DLL4 in identical patterns . These mice uncover that only DLL1 but not DLL4 can mediate regular segmentation of the embryo . In experiments with cultured cells expressing either ligand and Notch , we found that the functional difference observed is unlikely to depend on differences in the activation of Notch . Rather , the unsuspected but strong difference between both ligands in cis-inhibition , i . e . repression of Notch by a ligand expressed in the same cell as the receptor , a process described in the fruitfly but not in mammals and not for DLL4 provides a possible explanation for the divergence in tissues that coexpress ligand and receptor . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [] | 2015 | Context-Dependent Functional Divergence of the Notch Ligands DLL1 and DLL4 In Vivo |
Ethylene has been regarded as a stress hormone to regulate myriad stress responses . Salinity stress is one of the most serious abiotic stresses limiting plant growth and development . But how ethylene signaling is involved in plant response to salt stress is poorly understood . Here we showed that Arabidopsis plants pretreated with ethylene exhibited enhanced tolerance to salt stress . Gain- and loss-of-function studies demonstrated that EIN3 ( ETHYLENE INSENSITIVE 3 ) and EIL1 ( EIN3-LIKE 1 ) , two ethylene-activated transcription factors , are necessary and sufficient for the enhanced salt tolerance . High salinity induced the accumulation of EIN3/EIL1 proteins by promoting the proteasomal degradation of two EIN3/EIL1-targeting F-box proteins , EBF1 and EBF2 , in an EIN2-independent manner . Whole-genome transcriptome analysis identified a list of SIED ( Salt-Induced and EIN3/EIL1-Dependent ) genes that participate in salt stress responses , including several genes encoding reactive oxygen species ( ROS ) scavengers . We performed a genetic screen for ein3 eil1-like salt-hypersensitive mutants and identified 5 EIN3 direct target genes including a previously unknown gene , SIED1 ( At5g22270 ) , which encodes a 93-amino acid polypeptide involved in ROS dismissal . We also found that activation of EIN3 increased peroxidase ( POD ) activity through the direct transcriptional regulation of PODs expression . Accordingly , ethylene pretreatment or EIN3 activation was able to preclude excess ROS accumulation and increased tolerance to salt stress . Taken together , our study provides new insights into the molecular action of ethylene signaling to enhance plant salt tolerance , and elucidates the transcriptional network of EIN3 in salt stress response .
Soil salinity is a major abiotic stress that reduces plant growth and limits the productivity of agricultural crops . The detrimental effects of salt on plants are a consequence of both a water deficit resulting in osmotic stress and the effects of excess sodium ions imposed on critical biochemical processes [1] . The sessile nature of plants has favored the evolution of mechanisms to cope with various environmental stresses . One of these mechanisms is the release and utilization of a multitude of phytohormones , including a gaseous molecule ethylene [2] . Ethylene can trigger multiple physiological and morphological responses , including inhibition of cell expansion , induction of fruit ripening and abscission , and adaptation to stress conditions [3] . One of the well documented ethylene responses is the so-called “triple response” of etiolated seedlings , i . e . short , thickened root and hypocotyl , as well as exaggerated curvature of the apical hook [4] . Based on this highly reproducible and specific phenotype , a largely linear ethylene signal transduction pathway has been established [5] . In Arabidopsis , ethylene is perceived by a family of membrane-associated receptors [6] , [7] , [8] , which are negative regulators of the signaling pathway , and ethylene binding leads to functional inactivation of the receptors [9] . In the absence of ethylene , the active receptors recruit CTR1 ( CONSTITUTIVE TRIPLE RESPONSE1 ) to associate with the membrane and thus become activated [10] , which subsequently represses the downstream signaling pathway mediated by ETHYLENE INSENSITIVE2 ( EIN2 ) and EIN3 . EIN2 is a central component of the ethylene signaling transduction pathway , and its null mutant ein2 is completely insensitive to ethylene [11] . EIN2 is shown to locate in endoplasmic reticulum membrane [12] , and undergoes a hormone-induced cleavage and translocation event that is controlled by CTR1-directed phosphorylation of its carboxyl-terminus [13] , [14] , [15] . As the requisite component for ethylene signaling , EIN2 positively regulates the functions of EIN3/EIL1 transcription factors , which results in the activation of transcription of ERF1 and other downstream genes [16] , [17] . EIN3/EIL1 are short-lived proteins , which are quickly stabilized and accumulate in the nucleus in the presence of ethylene . Genetic and biochemical studies revealed that EIN3/EIL1 are subject to ubiquitin/proteasome-mediated proteolysis that requires two F-box proteins , EBF1/EBF2 [16] , [18] , [19] , [20] . Recently , our studies have demonstrated that ethylene stabilizes EIN3/EIL1 at least partly by promoting the proteasomal degradation of EBF1/EBF2 , and that EIN2 is indispensable for mediating ethylene-induced EIN3/EIL1 accumulation and EBF1/2 degradation [18] , highlighting the importance of EIN2 in the control of EIN3/EIL1 abundance . In addition to its role in regulating plant growth and development , ethylene also plays a key role in plant responses to biotic and abiotic stresses [21] . Recently , the functions of components of ethylene signaling in salt stress response were investigated . The ctr1-1 mutant exhibited increased salt tolerance and the germination rate and post-germination development of ctr1-1 were more tolerant under salt and osmotic stress treatments , especially under high concentration of salt [22] . Under salt stress , the ein2-1 mutant was severely affected in both seedling growth and seed germination processes , suggesting that EIN2 is required for salt stress tolerance [23] , [24] . The ein3-1eil1-1 double mutant exhibited remarkably reduced tolerance to high concentration of salt [22] , [23] , [24] . Recently , Jiang et al . reported that salinity-induced ethylene promotes Arabidopsis soil-salinity tolerance by enhancing Na/K homeostasis [25] . Despite such clear demonstration of a vital role of ethylene in salt stress response , the molecular mechanisms of how the ethylene signaling is modulated under salt stress condition and how ethylene signaling increases salinity tolerance are poorly understood . In this study , we demonstrated that plants pretreated with ethylene exhibited increased tolerance to salt stress , and that EIN3/EIL1 are both necessary and sufficient for salt tolerance . Interestingly , we found that salt stabilized EIN3/EIL1 protein by promoting EBF1/EBF2 proteasomal degradation in an EIN2 independent manner . Microarray analysis identified a large number of EIN3/EIL1-regulated genes ( SIEDs ) that participate in salt stress response , including many genes encoding reactive oxygen species ( ROS ) scavengers . A novel EIN3 target gene , SIED1 , was functionally studied and defined as an important mediator of ethylene-evoked salt tolerance .
Previous studies investigating the effect of ethylene in salt stress were conducted in conditions where ethylene and salt stress were simultaneously applied [23] . Because several salt-induced seedlings responses , such as leaf epinasty , chlorophyll loss and growth retardation , are also regulated by ethylene [4] , [5] , it is sometimes not clear how the final morphological output is the result of an altered salt or ethylene response . To specifically ascertain the role of ethylene in salt response , we pretreated Arabidopsis seedlings with ethylene or its biosynthesis precursor ACC and then transferred to MS medium supplemented with 200 mM NaCl alone . Upon ACC pretreatment , wild-type Col-0 displayed enhanced tolerance to salt compared with untreated control , with higher survival rate and lower relative electrolyte leakage ( an indicator for the salt stress damage ) [26] ( Figure 1A–C ) . By comparison , ebf1-1 mutant showed slightly lower survival rate , whereas ebf2-1 , an ethylene hypersensitive mutant [16] , was more tolerant to salt than Col-0 upon ACC pretreatment ( Figure 1A–C ) . Consistent with their respective ethylene response phenotype , ctr1-1 , EIN3ox ( a transgenic plant overexpressing EIN3 ) as well as EIL1ox displayed constitutively enhanced salt tolerance , whereas ACC pretreatment had virtually no effect on the salt tolerance of ethylene insensitive mutants , etr1-1 , ein2-5 and ein3-1eil1-1 ( Figure 1A–C; Figure S1 ) . To further examine the effect of ACC pretreatment on salt tolerance , 5-day-old seedlings of wild-type , ein3-1eil1-1 and EIN3ox were also transferred onto MS medium supplemented with serial concentrations of NaCl ( 0 , 50 , 100 , 150 and 200 mM ) . Similarly , ACC pretreatment significantly increased the survival rate , fresh weight as well as root length of wild-type but not ein3-1eil1-1 compared with ACC-untreated plants when 100 mM or higher concentrations of NaCl were applied ( Figure S2E vs S2B , S2F vs S2C and S2G vs S2D ) . Consistently , EIN3ox seedlings showed constitutively increased salt tolerance in terms of survival rate , fresh weight and root length ( Figure S2E vs S2B , S2F vs S2C and S2G vs S2D ) . Together , these results demonstrate that ACC pretreatment or overexpression of EIN3 leads to increased tolerance to salt stress , which depends on the canonical ethylene signaling pathway . To exclude the possibility that the observed effect of ACC pretreatment was due to the residual ACC remained in the pretreated seedlings , the experiment was repeated using ethylene gas , which was quickly diffusing away . After 5 days of 10 ppm ethylene gas treatment , seedlings were transferred onto MS medium supplemented with 200 mM NaCl in the air , and survival rates were calculated after two , three and four days , respectively . We found that ethylene pretreatment effectively increased the tolerance to salt in wild-type , ebf2-1 and ctr1-1 , evidenced by higher survival rates after three or four days of salt treatment , but had little effect on ein2-5 and ein3-1eil1-1 ( Figure 1D ) . Ethylene pretreatments with different lengths of time were also investigated . Seedlings pretreated with ethylene for 2 or 5 days exhibited increasingly enhanced survival rate in Col-0 , ebf2-1 , but not in ein2-5 and ein3-1eil1-1 , while 1 day of ethylene pretreatment had only marginal effect ( Figure 1E ) . Together with the ACC pretreatment experiments , these results support that exogenous ethylene application beforehand effectively increases salt tolerance . To further study the function of EIN3/EIL1 in ethylene-mediated salt response , a transgenic line expressing estradiol-inducible EIN3-FLAG in the ein3 eil1 ebf1 ebf2 quadruple mutant ( iE/qm ) was investigated [18] . Previous studies demonstrated that the iE/qm seedlings were completely insensitive to exogenously applied ethylene , and the accumulation of EIN3-FLAG fusion protein can be induced by estradiol ( but not by ethylene ) in a dose-dependent manner [18] . As reported , the EIN3-FLAG protein was undetected in estradiol-untreated iE/qm , and it was evidently induced in iE/qm upon estradiol treatment , but ACC treatment did not further increase its protein accumulation ( Figure S3A , Lane 1 , 2 , 8 , or Lane 7 , 9 ) . We also found that salt treatment did not affect EIN3 protein level in estradiol-treated iE/qm regardless of treatment time ( Figure S3A , Lane 3–5 , or Lane 2 , 6 ) . These results demonstrated that the estradiol-induced EIN3 protein accumulation in iE/qm seedlings is not altered by ethylene or salt treatment . Next , we investigated whether the estradiol-induced EIN3 protein in iE/qm seedlings effectively increased the tolerance to salt stress . In the absence of estadiol , where EIN3 protein was not detectable , the cotyledons and leaves of iE/qm seedlings treated with 200 mM NaCl were severely bleached after 3 days ( Figure S3B ) , and the survival rate was declined to less than 30% ( Figure S3C ) . In contrast , in the presence of serial concentrations of estradiol ( from 0 . 1 to 20 µM ) , iE/qm seedlings treated with 200 mM NaCl for 3 days appeared largely green and healthy ( Figure S3B ) , and the survival rates remained over 90% in all cases ( Figure S3C ) . Despite all concentrations of estradiol were sufficient for conferring salt tolerance , we noted that iE/qm seedlings supplemented with lower concentrations ( 0 . 1 or 1 µM ) grew better on salt medium ( Figure S3B ) . Therefore , all subsequent physiological experiments were performed with 1 µM estradiol . In line with cotyledon yellowing phenotype and survival rate , the leaf chlorophyll content and fresh shoot weight were significantly higher in estradiol-treated iE/qm seedlings than the untreated control under salt stress condition ( Figure S3D , S3F ) . Conversely , upon salt treatment , the ion leakage was evidently lower in estradiol-treated iE/qm plants than the untreated control ( Figure S3E ) . Taken together , these results indicate that loss of EIN3/EIL1 function leads to hypersensitivity to salt stress whereas accumulation of EIN3 alone results in enhanced salt tolerance , highlighting the requirement and sufficiency of EIN3/EIL1 for salt tolerance in Arabidopsis . Given that EIN3 is a critical regulator of plant salt responses , we next determined whether , and if so , how EIN3 is modulated by salt stress . We first monitored the level of endogenous EIN3 protein using an anti-EIN3 antibody [16] , [18] in response to salt treatment . We found that the levels of EIN3 protein started to increase after 3 h of salt treatment and dramatically accumulated after 6 h of treatment in wild-type Col-0 ( Figure 2A ) . We also checked the levels of EIN3 mRNA and found no obvious change after 6 h of salt treatment ( Figure S4 ) , suggesting that salt regulates EIN3 accumulation at the protein level . Previous study indicated that EIN2 is absolutely required for ethylene-induced EIN3/EIL1 accumulation , as no EIN3 or EIL1 protein can be detected in ein2 mutant [16] , [18] . To determine whether EIN2 is required for salt-induced EIN3/EIL1 protein accumulation , we detected the EIN3/EIL1 protein level in ein2-5 mutant under salt treatment . Surprisingly , we found that salt treatment ( but not mock treatment , Figure S5A ) promoted EIN3 and EIL1 proteins accumulation in the ein2-5 mutant background , although not as dramatic as in wild-type ( Figure 2B and Figure S6 ) . These results indicate that the salt stress signal is able to promote EIN3/EIL1 proteins accumulation in an EIN2-independent manner . In addition , we also analyzed the EIN3 protein levels in an ethylene receptor mutant etr1-1 [9] upon treatment with 200 mM NaCl for 3 h and 6 h using anti-EIN3 antibody . We found that EIN3 protein was evidently induced in etr1-1 mutants upon salt treatment for 6 h ( Figure S5B ) , suggesting that salt induced EIN3 protein accumulation does not require the canonical ethylene perception . We next investigated whether salt-induced EIN3 protein is transcriptionally functioning . A transgenic reporter line that harbors the GUS report gene driven by five tandem repeats of the EIN3 binding site ( EBS ) followed by the minimal 35S promoter , 5xEBS:GUS , has been previously used to monitor the transcriptional activity of EIN3 [27] , [28] . Upon salt treatment , GUS staining became overly intensified in the cotyledons and hypocotyls of 5xEBS:GUS/Col-0 plants ( Figure 2C ) , indicative of elevated levels of EIN3 activity under this condition , which was further supported by a quantification assay ( Figure 2D ) . Compared to that in Col-0 background , GUS activity was also evidently up-regulated in 5xEBS:GUS/ein2-5 plants upon salt treatment , although to a lesser extent ( Figure 2C and 2D ) . By contrast , GUS activity in 5xEBS:GUS/ein3 eil1 plants did not increase upon salt stress ( Figure 2D ) , suggesting that salt-increased 5xEBS:GUS activity is EIN3/EIL1-dependent . We also observed that the expression levels of several ethylene responsive genes , including ERF1 , ERF2 and PDF1 . 2 , were up-regulated by salt in both wild type and ein2-5 mutants ( Figure 2E–G ) . In keeping with the results of EIN3 accumulation and GUS expression ( Figure 2B–D ) , the expression level of ERF1 , a direct target gene of EIN3 [17] , was also lower in salt-treated ein2-5 mutant compared with that in wild-type ( Figure 2E ) . Thus , although salt treatment did promote EIN3 protein accumulation in ein2-5 , the relative lower level and activity of EIN3 might be inadequate to compensate for the loss of EIN2 that could elicit additional pathways contributing to salt tolerance . Taken together , our results suggest that , in addition to the canonical EIN2-dependent pathway , there exists a new pathway independent of EIN2 to mediate the salt stress signal to promote EIN3/EIL1 protein accumulation . It has been established that the stability of EIN3 is controlled by two F-box proteins , EBF1/EBF2 , and that ethylene-induced EIN3 stabilization is at least partly mediated by the destabilization of EBF1/EBF2 proteins in an EIN2-dependent manner [16] , [18] , [19] , [20] . To further characterize how EIN3 accumulation is enhanced by salt , we examined the levels of EBF1/EBF2 protein after salt treatment . Our initial effort to produce polyclonal antibodies recognizing endogenous EBF1 or EBF2 protein in plant tissues was unsuccessful . Therefore , two transgenic lines , 35S:EBF1-MYC/Col-0 and 35S:EBF2-MYC/Col-0 [18] , were used to detect the EBF1 and EBF2 protein levels . Immunoblot analysis showed that the protein levels of EBF1-MYC and EBF2-MYC markedly decreased upon salt treatment ( Figure 3A ) . We also noted that ACC pretreatment seemed to reinforce the destruction of EBF1/EBF2 proteins , as seedlings pretreated with ACC accumulated less EBF1/EBF2 proteins after salt application ( Figure 3A ) . Together with the finding that EIN3 protein level is not altered by salt in iE/qm seedlings ( Figure S3A ) , these results suggest that the salt-induced accumulation of EIN3 protein is due to reduced levels of EIN3-targeting F-box proteins . Our above data showed that salt induced EIN3 protein accumulation in ein2-5 mutant , so we asked whether salt-induced destruction of EBF1/EBF2 proteins also take place in the absence of EIN2 function . To address this question , a previously generated transgenic line , 35S:EBF2-GFP/ein2-5 , which showed high level of EBF2-GFP accumulation [18] , was used . Immunoblot analysis showed that the protein levels of EBF2-GFP markedly decreased upon salt treatment , regardless of ACC pretreatment ( Figure 3B ) . By contrast , treatment with MG132 , a 26S proteasome inhibitor , promoted a dramatic accumulation of EBF2-GFP and reversed the salt-induced EBF2-GFP degradation ( Figure 3C ) . Similarly , GFP fluorescence was dramatically reduced in the 35S:EBF2-GFP/ein2-5 after salt treatment , but MG132 treatment effectively reversed the salt effect and stabilized EBF2-GFP protein ( Figure 3D ) . We further examined the effect of other salt ions on EBF2 stability , and found that , as NaCl , treatments of KCl , NaNO3 and KNO3 all similarly led to the destruction of EBF2-GFP protein in ein2-5 mutant background ( Figure S7A ) . We also excluded the involvement of osmotic stress in the control of salt-induced EBF protein degradation , as high dose of mannitol treatment ( 200 mM ) had no effect on EBF2-GFP stability ( Figure S7B ) . Collectively , our data clearly demonstrated that salt stress leads to the proteasome-mediated degradation of EBF1/EBF2 proteins independent of the upstream ethylene signaling components , such as EIN2 . Our above data indicated that EIN3/EIL1 are both necessary and sufficient for conferring enhanced salt tolerance . To elucidate the molecular network underlying EIN3/EIL1-induced salt tolerance , we performed transcriptome profiling of EIN3ox , ein3eil1 and wild type Col-0 . For this analysis , 5-day-old light-grown seedlings treated with or without 200 mM NaCl for 6 h were used . This design enabled us to compare the transcriptional profiles among plants with different levels of EIN3 activity , as well as to identify salt-regulated and EIN3/EIL1-dependent genes . Treatment with high salt for 6 h resulted in the induction of 1482 transcripts while the repression of 1745 transcripts ( using both q value<0 . 05 and 2-fold as a cutoff ) in wild-type Col-0 ( Figure 4E , 4F ) . Applying a q value<0 . 05 and 5-fold as a cutoff , 509 transcripts were induced while 209 transcripts were repressed in wild type by salt treatment , which were arbitrarily defined as salt-regulated genes in this study ( Figure 4A and 4B ) . Using the same cutoff , 365 and 74 transcripts in EIN3ox while 281 and 98 transcripts in ein3eil1 were induced and repressed by salt treatment , respectively ( Figure 4A , 4B ) . Of the 509 salt-induced genes , 162 were also elevated in salt-treated EIN3ox and ein3eil1 mutant ( Figure 4A ) . Conversely , 36 out of 209 salt-repressed genes were also down-regulated in salt-treated EIN3ox and ein3eil1 mutant ( Figure 4B ) . To investigate how EIN3 activation leads to increased salt tolerance , we were particularly interested in two classes of genes: salt-induced EIN3/EIL1-dependent ( SIED ) genes and salt-repressed EIN3/EIL1-dependent ( SRED ) genes ( Figure 4C , 4D ) . The former class includes those genes whose levels are induced by salt at least 5-fold in Col-0 ( P value 0 . 0041 ) , plus that salt induction is more pronounced in EIN3ox ( P value 0 . 00099 ) but less in ein3 eil1 ( P value 0 . 0094 ) ( i . e . salt-induced gene expression is at least partly dependent on EIN3 activity ) ( Table S1 ) . The latter class includes those genes whose levels are repressed by salt at least 5-fold in Col-0 ( P value 0 . 0016 ) , plus that salt repression is more pronounced in EIN3ox ( P value 0 . 00093 ) but less in ein3 eil1 ( P value 0 . 0081 ) ( Table S2 ) . Based on these criteria , 114 SIED genes and 14 SRED genes were identified ( Table S1 and S2 ) . The drastic difference on the number of SIED and SRED genes suggested that EIN3/EIL1 might enhance salt tolerance mainly through inducing genes or pathways that participate in plant survival , rather than repressing genes or pathways that lead to plant death under salinity stress . In support of this speculation , 18 out of 114 SIED genes ( ∼16% ) are defense-related genes that function to enhance plant tolerance or resistance to abiotic or biotic stresses . Several ERF ( 11 ) and JAZ ( 3 ) genes were found to be SIED genes , implying that the signaling pathways of ethylene and jasmonic acid ( JA ) , two stress hormones , have been preferentially activated by salt stress , which is consistent with previous studies [23] , [29] , [30] . The considerable enrichment of ERF genes , many of which are direct target genes of EIN3 [17] , [31] , suggests that the identification of SIED genes is biologically relevant . Furthermore , when compared the SIEDs with ethylene-regulated EIN3-target genes identified by ChIP-Sequencing [32] , we found that 15 out of 114 SIEDs ( Highlighted in Table S1 ) , such as At5g22270 and At5g59820 ( ZAT12 ) , are the direct targets of EIN3 . However , most of SIEDs are not the target genes of EIN3 identified by Chang et al . [32] . This could be due to that EIN3 preferentially binds to specific subsets of target promoters dependent on the initial treatment/stimulus . Alternatively , it is also possible that the majority of SIED genes are indirectly induced by EIN3 , for instance , via the ERF transcription factors . The SIED genes were further analyzed using the gene ontology ( GO ) enrichment tool Gorilla [33] . We found that , in terms of molecular function category , there were notable enrichments for metabolic processes , as well as transcription , DNA binding , and oxidoreductase activity ( Figure S8 ) . For instance , of 114 SIED genes , we found 9 genes encoding oxidoreductases and 4 genes involved in electron transport or energy pathways , suggesting that modulation of oxidative/reductive status under salt stress might be an important mechanism of EIN3/EIL1 action to enhance plant survival . In this study , we demonstrated that pretreatment with ethylene conferred increased salt tolerance , which depends on the action of EIN3/EIL1 ( Figure 1 ) . One explanation for this priming effect of ethylene is that EIN3/EIL1 activation in advance alters the expression of genes that ultimately leads to salt tolerance . To test this possibility , we compared the salt-regulated transcriptome ( salt-treated versus untreated Col-0 ) and EIN3-regulated transcriptome ( EIN3ox versus Col-0 without salt treatment ) . By a 2-fold cutoff , 366 genes ( P value 0 . 0096 ) were identified as EIN3-induced while 360 genes ( P value 0 . 0099 ) were EIN3-repressed based on transcriptome profiling ( Figure 4E , 4F ) . We found that 92 out of 366 EIN3-induced genes ( ∼25% ) and 121 out of 360 EIN3-repressed genes ( ∼34% ) were also induced and repressed by salt stress , respectively ( Figure 4E , 4F ) . By a 5-fold cutoff , 17 out of 62 genes ( ∼27% ) vastly up-regulated by EIN3 ( EIN3ox versus Col-0 ) ( P value 0 . 0095 ) , were also highly induced by salt , including several ERFs , defense genes , and biosynthetic process and metabolism genes ( Table S3 ) . Six genes were selected to further verify the microarray data using qRT-PCR , which showed largely similar expression patterns ( Figure S9 ) . These results indicated that overexpression of EIN3 activated the expression of a number of stress-responsive defense and metabolism genes even under unstressed conditions . Therefore , the priming effect of ACC/ethylene pretreatment could be attributed to altered expression of numerous salt-responsive genes , which subsequently increases tolerance when salt stress is encountered . Further investigation on the functionality of these stress-responsive defense and metabolism genes in salt tolerance is needed to test this hypothesis . To further investigate the roles of the SIED genes in salt tolerance , Salk T-DNA insertion lines of SIED genes were ordered from ABRC , and the homozygous lines of 47 insertion mutants were obtained and verified by genotyping ( Table S4 ) . Characterization of salt stress phenotype showed that , while 41 mutants were indistinguishable from wild type , 6 mutants , namely zat12 , azf2 , cni1 , szf2 , phil and SALK_067396 ( hereafter designated as sied1 , salt-induced and EIN3/EIL1-dependent gene 1 ) , exhibited a salt-hypersensitivity phenotype similar to ein3 eil1 , which showed low survival rate under salt stress ( Figure 5A , 5B ) . PCR genotyping assays showed that the six mutant lines were knockout alleles in their corresponding genes ( Figure S10 ) , suggesting that these genes are positive regulators of salt tolerance . Interestingly , five of these SIEDs , ZAT12 [34] , [35] , [36] , AZF2 [37] , SZF2 [38] , CNI1 [39] and PHI1 [40] have been previously demonstrated to modulate various abiotic stresses . For instance , transgenic plants overexpressing ZAT12 were more tolerant to osmotic stress , while zat12 knockout mutants were more sensitive to osmotic and salt stress [34] . Overexpression of CNI1 ( Carbon/Nitrogen Insensitive 1 ) , a RING-type ubiquitin ligase , caused a hyposensitivity to C/N stress , and cni1 knockout mutants resulted in hypersensitivity to C/N stress and salt treatment [39] . Of the five genes , three genes encode zinc-finger transcription factors ( ZAT12 , AZF2 , SZF2 ) . In fact , we have identified at least 9 genes encoding zinc-finger transcriptional regulators as SIED ( Table S1 ) . It thus remains interesting to determine whether all other identified zinc-finger proteins are involved in salt tolerance . Together , our results supported the idea that many EIN3/EIL1 target genes identified in this analysis are involved in various stress responses , including salt stress . The sixth salt-hypersensitivity mutant , sied1 , corresponds to At5g22270 , a functionally unknown gene encoding a 93-amino acid polypeptide . Microarray data and qRT-PCR analysis showed that SIED1 had evidently higher expression level in EIN3ox and lower level in ein3 eil1 compared with that of wild type ( Figure 5C ) . To confirm its EIN3-induced expression pattern , we generated a transgenic reporter line that harbors the β-glucuronidase ( GUS ) gene driven by the promoter of SIED1 . Consistent with the gene expression data , GUS staining was weaker in ein3 eil1 but stronger in EIN3ox than that in wild type ( Figure 5D , 5E ) . Moreover , we found that EIN3 directly binds to the promoter of SIED1 , as well as other SIED genes including ZAT12 , SZF2 , PHI and AZF2 , but does not bind to the promoter region of CNI1 ( Figure S11E–F ) , indicating that EIN3 selectively binds to the promoters of many SIED genes in vivo . Since loss-of-function SIED1 mutation led to salt hypersensitivity , we next generated transgenic plants constitutively overexpressing SIED1 in wild type to further investigate its role in salt tolerance . qRT-PCR analysis revealed that higher levels of SIED1 mRNA were detected in two independent overexpression lines 5# and 9# compared with Col-0 ( Figure 5C ) . Phenotypic analysis showed that overexpression of SIED1 effectively enhanced salt tolerance and greatly increased survival rate upon salt treatment for 4 days ( Figure 5F , 5G ) . Since SIED1 is a direct target of EIN3 , we then determined whether overexpression of SIED1 could repress the salt-hypersensitivity phenotype of ein3eil1 . Toward this end , we generated the 35S:SIED1/ein3eil1 transgenic plants . Compared with ein3eil1 , the seedlings of 35S:SIED1/ein3eil1 showed significantly increased survival rate and fresh weight upon salt treatment ( Figure 5H , 5I ) , indicating that SIED1 acts genetically downstream of EIN3 and overexpression of SIED1 is sufficient to suppress the salt-hypersensitivity phenotype of ein3eil1 . As expected , overexpression of SIED1 also repressed the salt-hypersensitivity phenotype of sied1 mutant ( Figure 5H , 5I ) . Taken together , these results identify SIED1 , acting downstream of EIN3 , as a novel component that plays a positive role in salt tolerance . The transcriptome profiling analysis revealed that genes encoding oxidoreductase activity are enriched in SIED , suggesting that EIN3 might modulate the oxidative/reductive status under salt stress condition . Interestingly , we found that the expression of many genes encoding peroxidases ( PODs ) was induced in wild type by salt treatment , whose expression was also elevated in EIN3ox but reduced in ein3eil1 under salt stress ( Table S1 and Figure S12 ) . These observations were further confirmed by qRT-PCR results with six selected POD genes ( Figure 6A ) . In the meanwhile , we did not find evident differences in the expression levels of genes encoding superoxide dismutases ( SOD ) , catalases ( CAT1 , CAT2 and CAT3 ) and NADPH oxidases ( AtRobhA-F ) among the three genotypes upon salt treatment ( Figure S13A–C ) . These results suggest that EIN3 selectively induces the expression of genes encoding peroxidases . Peroxidase activity assay also showed that POD activity was significantly higher in EIN3ox seedlings than that in wild type ( P<0 . 05 ) or ein3eil1 ( P<0 . 01 ) under salt stress condition ( Figure 6B ) . We next analyzed the promoter regions of two selected POD genes ( At5g42180 and At2g18980 ) and found three EBSs in each promoter ( Figure 6C ) . Chromatin immunoprecipitation ( ChIP ) assay using wild-type seedlings showed that the anti-EIN3 antibody bound strongly to the P2 fragment of At5g42180 and the P1 fragment of At2g18980 ( Figure 6D ) respectively , suggesting that EIN3 binds directly to the promoter regions of these genes in vivo . Furthermore , the promoter sequences of At5g42180 ( P2 ) and At2g18980 ( P1 ) were used for electrophoresis mobility shift assay ( EMSA ) , showing that EIN3 can bind to these promoter sequences in vitro ( Figure 6E ) . These results indicate that EIN3 increases peroxidase activity through the direct transcriptional regulation of PODs expression . Peroxidases have been shown to participate in plant response against abiotic stresses as key scavengers of ROS [41] . The induction of several POD genes by EIN3 suggested that activation of EIN3 might facilitate the scavenging of ROS when plants are stressed with high salinity , thus leading to enhanced salt tolerance . To test this possibility , we determined the endogenous ROS accumulation and H2O2 content of wild-type Col-0 , ein3eil1 and EIN3ox upon salt treatment . Salt treatment evidently increased ROS accumulation and H2O2 content in the cotyledons of Col-0 , indicated by H2DCFA fluorescence [42] , [43] and DAB ( 3 , 3-diaminobenzidine ) staining [44] , respectively ( Figure 7A–C ) . We further found higher level of ROS and H2O2 accumulation in ein3eil1 while lower level of ROS and H2O2 production in EIN3ox than that of Col-0 upon salt treatment ( Figure 7A–C ) , in accordance with the salt tolerance phenotypes of these genotypes ( Figure 1 ) . Additionally , we found that upon salt treatment , a higher ROS accumulation ( indicated by DAB staining and H2O2 contents ) was found in sied1 mutant plants , while a significant decrease was observed in SIED1ox seedlings ( Figure S14 ) , suggesting that SIED1 acts to enhance salt tolerance also via reducing ROS accumulation . Since ACC-pretreated plants exhibited tolerance to salt stress , we next examined the effect of ACC pretreatment on ROS accumulation and H2O2 production in salt-treated plants . Compared with non-pretreated seedlings , ACC-pretreated Col-0 showed evident lower levels of ROS accumulation and H2O2 generation when stressed with high salinity ( Figure 7A–C ) . Meanwhile , ROS level and H2O2 content remained constantly high in ein3eil1 and low in EIN3ox , and no obvious changes were found in these two genotypes upon ACC pretreatment ( Figure 7A–C ) . These results are also in good correlation with their respective salt stress phenotypes with or without ACC pretreatment ( Figure 1 ) . We next examined the expression levels of a well-established ROS marker gene , DEFL ( defensin-like ) , which was shown to be induced by various ROS [45] . In agreement with the histochemical observations of ROS accumulation , ACC-pretreated Col-0 seedlings accumulated less DEFL mRNA than non-pretreated Col-0 upon 3 h and 6 h of salt applications ( Figure 7D ) . By contrast , DEFL expression was highly induced by salt stress in ein3 eil1 regardless of ACC pretreatment , whereas its expression levels remained constantly lower in salt-stressed EIN3ox and ebf2-1 than that in Col-0 ( Figure 7D ) , further supporting the importance of EIN3 action in the modulation of salt-evoked ROS accumulation . Taken together , our results indicate that activation of EIN3 , either by EIN3 overexpression , ACC pretreatment , or EBF2 mutation , induces the expression of numerous POD and SIED genes , which arguably contributes to the decreased accumulation of ROS , and consequently the detoxification of salt stress-elicited damages .
Genetic and biochemical studies revealed that EIN3/EIL1 proteolysis is mediated by two F-box proteins , EBF1/EBF2 . Upon ethylene treatment , the levels of EBF1/EBF2 proteins are down-regulated through 26S proteasome pathway , which leads to the accumulation of EIN3 and EIL1 proteins [16] , [18] , [19] , [20] . Moreover , our previous work indicated that EIN2 is indispensable for ethylene-induced EIN3 and EIL1 stabilization and degradation of EBF1/EBF2 proteins , because no EIN3 or EIL1 protein can be detected in ein2 but EBF1/EBF2 proteins are constitutively accumulated in the ein2 background [18] . In this study , we found that the level of EIN3 and EIL1 proteins was remarkably up-regulated by salt treatment in wild type , and also elevated in ein2-5 background , although to a lesser extent . Conversely , we further found that the levels of EBF1/EBF2 proteins were evidently down-regulated by salt treatment in both Col-0 and ein2-5 background , which is the result of 26S proteasome-executed EBF1/EBF2 degradation . These findings indicate that EIN2 is dispensable for salt-induced EIN3/EIL1 accumulation and EBF1/EBF2 degradation , which is distinct from the regulatory mechanism in ethylene signaling that fully depends on EIN2 . Thus , we propose that salt treatment promotes EBF1/EBF2 protein degradation , which consequently induces EIN3 protein accumulation in both EIN2-dependent and EIN2-independent pathways . These findings for the first time report the existence of an alternative pathway that is distinct from the canonical ethylene signaling pathway to modulate the protein stability of EBF1/EBF2 and EIN3/EIL1 . In addition to NaCl , we found that treatments with equal concentration of KCl , NaNO3 or KNO3 salt also effectively induced EBF2-GFP protein degradation in ein2-5 mutant ( Figure S7A ) , suggesting that this regulation is a general salt stress response rather than the specific effect caused by sodium chloride . High salt condition often affects osmotic homeostasis and causes osmotic stresses . However , treatment with 200 mM mannitol did not affect EBF2-GFP protein level , suggesting that the ionic stress but not osmotic stress regulates EBF1/EBF2 protein turnover ( Figure S7B ) . Further investigation is needed to elucidate the regulatory mechanism behind the salt-induced proteasomal degradation of EBF1/EBF2 . In this study , we demonstrate that the protein stability of EBF1/EBF2 and EIN3/EIL1 is regulated by ethylene and salt stress in different ways . Recently , increasing body of evidence indicates that EIN3/EIL1 might act as a signaling hub that integrates multiple hormone and stress signals [18] , [31] , [43] , [46] , [47] , [48] , [49] , [50] . Our previous works and other studies reported that EIN3 protein stability is also controlled by light irradiation [43] , auxin and its biosynthesis inhibitor , Kyn [28] , and glucose [51] . These studies collectively indicate that EIN3/EIL1 are not limited to the ethylene signaling , but rather participating in the regulation of myriad processes , whose function and/or abundance are also modulated by various signals besides the ethylene gas . Given that only a subset of abiotic stresses ( e . g . high salt , high sucrose , freezing , but not osmosis ) alter the stability of EIN3 or EBF proteins , it is thus interesting to ascertain whether , and if so , how other types of abiotic stresses , such as drought , chilling , heat , and heavy metals , affect the stabilization of EBF1/EBF2 and EIN3/EIL1 proteins . The elucidation of differential regulation of EIN3 and/or EBF protein stability by environmental and stress signals would provide insights into how EIN3 exerts its unique effect in plant adaptation to various growth conditions . Previous studies revealed a number of salt-tolerance genes that might be EIN3 target genes , such as ESE1 ( an AP2/EREBP transcription factor ) [52] and JERF3 [53] . However , a systematic analysis of EIN3-regulated genes in salt tolerance is lacking . In this study , we conducted a genome-wide transcriptome profiling in combination with genetic approach to dissect and identify numerous EIN3-regulated genes and pathways that might contribute to ethylene-directed salt tolerance ( Table S1 ) . One of the pathways is the scavengers of ROS , in which EIN3 up-regulates the expression of numerous peroxidases . One important cause of high salinity-imposed damage is ROS generated by salt stress . ROS plays a dual role in plants , as actors or modulators of cellular signaling pathways on one hand , and as oxidative agents or toxic products elicited by cellular stresses on the other hand [54] . ROS is tightly regulated by the equilibrium between production and scavenging . Transgenic plants overexpressing enzymes involved in oxidative protection , such as glutathione peroxidase ( GPX ) [55] , superoxide dismutase ( SOD ) [56] , ascorbate peroxidase ( APX ) [57] , exhibited enhanced salt tolerance . In this study , based on the transcriptome analysis , we found that the expression of several ROS scavenger peroxidases ( PODs ) was notably elevated in EIN3ox under salt condition compared with that in wild type or ein3 eil1 ( Table S1 and Figure 6 ) . Accordingly , the overall activity of POD enzymes was significantly higher in EIN3ox than that in wild type or ein3 eil1 upon salt stress ( Figure 6B ) . The transcript level of a zinc-finger transcription factor , ZAT12 , which has been shown to induce the expression of APX1 [35] , was also up-regulated in EIN3ox ( Table S1 ) . Consistently , we observed lower levels of H2O2 accumulation and ROS marker gene expression in EIN3ox but higher levels in ein3 eil1 compared with wild type upon salt treatment ( Figure 7 ) . Thus , elimination of excessive ROS accumulation under salt stress through inducing the expression of peroxidases is one contributing mechanism behind EIN3-mediated salt tolerance . In addition , we found that EIN3/EIL1 enhanced salt tolerance through regulating a myriad of SIED genes . We provided genetic evidence to indicate that a portion of these SIED genes participate in salt tolerance , including 5 genes previously known to be induced by and/or involved in various abiotic stresses , and a novel gene ( SIED1 ) whose function is previously unknown . Further biochemical studies uncovered that 5 of these 6 SIED genes , including SIED1 , are direct target genes of EIN3 . We found that overexpression of SIED1 also decreased ROS accumulation upon salt treatment ( Figure S14 ) . Nevertheless , the lack of salt stress phenotype for other SIED knockout mutants does not necessarily mean that these genes are not involved in salt tolerance . For instance , it is well known that ERF family transcription factors and JAZ family transcriptional regulators possess tremendous functional redundancy within the family members [31] , [58] , [59] . In addition , we cannot completely rule out the possibility that the lack of salt stress phenotype is simply because the T-DNA insertions in many SIED genes may have a marginal effect on gene expression/function ( Table S4 ) . Additional mutant alleles , gain-of-function studies or multigenic mutants analysis would help clarify whether those SIED genes play a role in EIN3-induced salt tolerance . Based on our genomic and genetic studies , the identification of numerous SIED and SRED genes would thus serve as a proper starting point to further dissect the complicated signaling network that is directed by EIN3/EIL1 in plant adaptation to salt stress . Intracellular K+/Na+ homeostasis is crucial for cell metabolism and is considered to be a key component of salinity tolerance in plants . Jiang et al . recently reported that salinity-induced ethylene is a potent promoter of salt tolerance through enhancing Na+/K+ homeostasis in Arabidopsis [25] . In fact , we also found that salt-stressed EIN3ox seedlings had accumulated more K+ and less Na+ , whereas ein3-1 eil1-1 seedlings had higher Na+ but lower K+ content compared with wild type ( data not shown ) . These results indicate that the modulation of intracellular K+/Na+ homeostasis serves as another contributing mechanism for ethylene-induced tolerance to high salt stress . In summary , our study provides new insights into how ethylene enhances plants' tolerance to high salinity . We propose that EIN3 and EIL1 play a central role in conferring salt tolerance via at least two mechanisms , one is to modulate intracellular K+/Na+ homeostasis , the other is to deter ROS accumulation by inducing SIEDs and PODs gene expression . Our study also provides a possible explanation for the priming effect of ethylene pretreatment , as ethylene treatment decreases EBF1/2 stability and increases EIN3/EIL1 abundance , which enhances the sensitivity of this pathway to salt stress signal . Once the EIN3 pathway is activated in advance ( by ethylene/ACC ) , it will alter the expression of downstream SIED genes , many of which are direct target genes of EIN3 , and induce a myriad of defense pathways , and switch plants to a more resistant state .
Arabidopsis thaliana ecotype Col-0 was the parent strain for all mutants and transgenic lines used in this study . Surface-sterilized seeds were plated on MS medium supplemented with or without 10 µm ACC and imbibed for 4 d in 4°C to improve germination uniformity . For phenotypic analysis under salt stress , seedlings were transferred onto MS agar plates containing 200 mM NaCl and their subsequent appearance was recorded photographically 3 days after transfer . The salt stress of seedlings was indicated by visibly bleached leaves . Reactive oxygen species ( ROS ) accumulation in seedlings was detected using the cell-permeable fluorescent probe 2′ , 7′-dichlorodihydrofluorescein diacetate ( DCFH2-DA; Molecular Probes ) according to Schopher et al . [42] . 5-day-old seedlings grown on MS with or without 10 µM ACC were treated with 200 mM NaCl for 6 h . Then , seedling were incubated in 100 µM DCFH2-DA in 1% ethanol for 20 min , and washed with distilled H2O to remove the dye before the observation of ROS accumulation under the confocal microscope . Confocal images were obtained after excitation at 488 nm and emission at 522 nm . H2O2 production was detected in seedlings using 3 , 3-diaminobenzidine ( DAB ) as substrate . Relative electrolyte leakage and chlorophyll content were measured as described previously [60] . Histochemical and Fluorimetric GUS assays were performed using the method described by Jefferson [61] . A Leica TCS SP2 inverted confocal laser microscope with ×40 objectives was used to detect GFP fluorescence . The excitation wavelength was 488 nm , and a bandpath filter of 510 to 525 nm was used for emission . Plant samples were ground in liquid N2 and soluble protein extracts were made by homogenization in 50 mM Tris–HCl ( pH 8 . 0 ) , 10 mM NaCl , 0 . 1 M PMSF , and 0 . 1 M DTT , with subsequent centrifugation at 13 . 000× g for 30 min at 4°C . The protein in the supernatants was quantified by Bradford's assay ( Bradford , 1976 ) . Western blot analysis was performed as described previously [18] , [60] with anti-GFP ( Invitrogen ) , anti-FLAG ( Sigma-Aldrich ) , anti-MYC ( AbChem ) , or anti-EIN3 antibodies [16] . Total RNA was extracted from seedlings and analyzed as described previously [62] . First strand cDNA samples were generated from total RNA samples by reverse transcription using an AMV reverse transcriptase 1st strand cDNA synthesis kit ( Life Sciences , Promega ) and were used as templates for qPCR-based gene expression analysis as described previously [60] . The oligonucleotide primer sequences used to amplify specific cDNAs were described in Table S5 . Microarray experiments were performed using Arabidopsis Affymetrix chips ( Santa Clara , CA ) . Total RNA was extracted from 5-d-old post germinated seedlings grown on MS medium , which were subsequently subjected to 200 mM NaCl for 6 h . Each experiment used two biological replicates , and each represented a pool of around 200 seedlings from two individual plates . The expression data were analyzed using Gene Spring version 4 . 2 . 1 ( Silicon Genetics Inc . , Red- wood City , CA , USA ) . A q value<0 . 05 and fold change >2 between control and treatment samples were considered as a cutoff . GO enrichment analysis on SIED genes was performed using the software GOrilla ( Gene Ontology enRIchmentanaLysis and visuaLizAtion tool ) as described previously [33] . 10 g of 5-d-old Col-0 seedlings were prepared for ChIP assays using anti-EIN3 antibody [16] , and the enriched DNA fragments were measured by qPCR as previously described [60] . All assays were performed with two biological replicates and three technical replicates . EMSA assay was performed as described previously [60] . The N-terminus DNA binding domain of EIN3 protein ( amino acids 141 to 352 ) that is sufficient for DNA binding [31] was expressed as a glutathione S-transferase ( GST ) fusion protein in Escherichia coli and purified , and used for EMSA experiments . Plants overexpressing SIED1 were generated by Agrobacterium tumefaciens strain GV3101 -mediated transformation into Arabidopsis Col-0 by floral dip [63] , using a construct that contained the full-length coding region of SIED1 ( At5g22270 ) in the PBI121vector . To generate pSIED1:GUS construct , a 2 . 2-kb SIED1 promoter region was amplified from genomic DNA and inserted into PBI101 vector , introduced into GV3101 , and transformed into Col-0 , ein3 eil1 and EIN3ox plants [63] . To generate EIL1pro:EIL1-GFP construct , the promoter region and the full-length coding region of EIL1 was amplified from genomic DNA and inserted into pCHF3 vector , introduced into GV3101 , and transformed into wild-type Col-0 and ein2-5 plants . The values we obtained in the figures were expressed as the means ( SD ) . Two-tailed Student's t tests were used . The microarray data reported in this paper have been deposited at NASC ( The Nottingham Arabidopsis Stock Centre ) database under accession number NASCARRAYS-659 . | High salinity , as a world-wide abiotic stress , restricts root water uptake , damages cell physiology , and limits the productivity of agricultural crops . Ethylene is a major phytohormone that regulates plant development in response to adverse environments , including high salt stress . However , the molecular mechanisms of how ethylene signal exerts its effect and how ethylene signaling is modulated upon salt stress remain to be explored . Here , we report that high salinity induces EIN3/EIL1 protein accumulation and EBF1/2 protein degradation in an EIN2-independent manner . Moreover , the activated EIN3 deters excess ROS accumulation and increases salt tolerance . Transcriptome analysis and functional studies reveal an EIN3-directed gene network in salt stress response . Functional studies of 114 SIED ( Salt-Induced and EIN3/EIL1-Dependent ) genes identify a novel regulator of ROS dismissal and salt tolerance . This new understanding of ethylene/salt mutual regulation would allow a better manipulation and engineering of EIN3 and its downstream SIED genes to enhance plant tolerance and adaption to salt stress , particularly in those economically important crops in the future . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"biochemistry",
"plant",
"biochemistry",
"plant",
"defenses",
"plant",
"science",
"plant",
"hormones",
"ethylene",
"biology",
"and",
"life",
"sciences",
"plant",
"physiology",
"hormones"
] | 2014 | Salt-Induced Stabilization of EIN3/EIL1 Confers Salinity Tolerance by Deterring ROS Accumulation in Arabidopsis |
Chagas disease is one of the most important neglected tropical diseases in the Americas . Vectorborne transmission of Chagas disease has been historically rare in urban settings . However , in marginal communities near the city of Arequipa , Peru , urban transmission cycles have become established . We examined the history of migration and settlement patterns in these communities , and their connections to Chagas disease transmission . This was a qualitative study that employed focus group discussions and in-depth interviews . Five focus groups and 50 in-depth interviews were carried out with 94 community members from three shantytowns and two traditional towns near Arequipa , Peru . Focus groups utilized participatory methodologies to explore the community's mobility patterns and the historical and current presence of triatomine vectors . In-depth interviews based on event history calendars explored participants' migration patterns and experience with Chagas disease and vectors . Focus group data were analyzed using participatory analysis methodologies , and interview data were coded and analyzed using a grounded theory approach . Entomologic data were provided by an ongoing vector control campaign . We found that migrants to shantytowns in Arequipa were unlikely to have brought triatomines to the city upon arrival . Frequent seasonal moves , however , took shantytown residents to valleys surrounding Arequipa where vectors are prevalent . In addition , the pattern of settlement of shantytowns and the practice of raising domestic animals by residents creates a favorable environment for vector proliferation and dispersal . Finally , we uncovered a phenomenon of population loss and replacement by low-income migrants in one traditional town , which created the human settlement pattern of a new shantytown within this traditional community . The pattern of human migration is therefore an important underlying determinant of Chagas disease risk in and around Arequipa . Frequent seasonal migration by residents of peri-urban shantytowns provides a path of entry of vectors into these communities . Changing demographic dynamics of traditional towns are also leading to favorable conditions for Chagas disease transmission . Control programs must include surveillance for infestation in communities assumed to be free of vectors .
Chagas disease , caused by infection with protozoan parasite Trypanosoma cruzi , causes more morbidity and mortality than any other parasitic disease in the Western Hemisphere [1] . T . cruzi is carried by numerous species of triatomine insects . Humans and other mammals usually become infected when the triatomine vector defecates during its blood meal , and fecal material containing the parasite is inoculated through the bite wound or mucous membranes [2] . Vector-borne transmission only occurs in the Americas , where 8–10 million people , including an estimated 192 , 000 Peruvians , are currently infected with T . cruzi [3] , [4] . The member countries of the Southern Cone Initiative ( INCOSUR ) have worked since 1991 to eliminate household infestation with Triatoma infestans , the most important Chagas disease vector in the southern half of South America , through large-scale residual application of pyrethroid insecticides [5] , [6] . Despite remarkable successes , major challenges remain to vector control , among them the increasing urbanization of the disease [7] . Chagas disease is traditionally associated with rural villages with adobe houses hospitable to T . infestans and other domestic vectors [8] and vector-borne transmission appears to be rare in urban settings [9]–[11] . However , in marginal communities of the city of Arequipa ( pop . 750 , 000 ) in southern Peru , urban T . cruzi transmission cycles have become established [12] , [13] , and a vector control campaign has been in place in the city of Arequipa since 2002 [13] . The settlement and migration patterns in and around cities therefore may be important to understanding the dynamics that make certain communities more susceptible to Chagas disease vectors [14] . Latin America has experienced an overwhelming phenomenon of urbanization due in most part to in-migration , and Peru is no exception [15] , [16] . Few studies have directly examined migration and settlement patterns , and their connections to Chagas disease transmission . Here we use qualitative methods to explore the migration and settlement patterns , and their links with vector infestation , in different communities around the city of Arequipa .
The research protocol was approved by the ethical review committees of the Asociación Benéfica PRISMA and the Johns Hopkins Bloomberg School of Public Health . All participants provided written informed consent prior to data collection , including consent for audio-recording . Arequipa is the second largest city in Peru , located in an arid zone 2 , 300 m above sea level [15] . The outskirts of the city contain hundreds of peri-urban pueblos jóvenes ( young towns or shantytowns ) and pueblos tradicionales ( traditional towns ) . Pueblos jóvenes are low-income hillside squatter settlements founded over the past 60 years [17] , [18] . Pueblos tradicionales tend to be in lower-lying flat areas , are inhabited by higher-income landowners , and date back to the late 19th or early 20th century . ( See Figure 1 for photos of the two types of communities . ) Because preliminary data from our research group indicated that T . infestans prevalence differed between these two types of towns [12] , we compared migration and settlement patterns in 3 pueblos jóvenes and 2 pueblos tradicionales . The research team worked with 2–3 community leader “gatekeepers” in each community to ensure acceptance and to recruit people who could provide detailed information about their personal history of migration and settlement ( for interviews ) or community history ( for focus groups ) . A total of 94 female and male participants were enrolled in the study . This was a qualitative study that employed focus group discussions and in-depth interviews . Focus group sessions were carried out with 8–10 participants in each community at central , well-known locations ( health establishments , community centers ) selected by the gatekeepers . We used participatory methodologies to explore the community's demographic characteristics and mobility patterns , and historical and current presence of triatomine vectors . Participants created community maps [19] which formed the basis for discussions of the history and characteristics of communities . Participants then created a timeline [19] of important community events dating back approximately 40 years , first exploring general events and then focusing on events related to Chagas disease and vector infestation . All sessions were audio-recorded; participatory activities were recorded on large sheets of paper that were hung on the wall and visible to all participants . In-depth interviews utilized an event history calendar ( EHC ) , a highly structured but flexible interview style that facilitates recall by using the individual's own experiences as cues [20] . Interviews were carried out with 10 participants per community , and explored migration history , experience with Chagas disease , the presence of its vectors , and customs of raising animals in each place of residence , starting from birth . Each interview , held at a location selected by the participant ( their home , the focus group location ) , was audio-recorded and lasted 45 to 60 minutes . All interviews and focus groups were conducted by two of the authors ( AB and GH ) . Presence of vectors was examined by asking participants when they had seen triatomine bugs . Because T . infestans is the sole insect vector for Chagas disease in southern Peru , we only asked participants to recall the presence of this species . All of the study communities had been involved in insecticide application programs run by the Ministry of Health ( MOH ) and as a result were broadly familiar with T . infestans , which they refer to as a “chirimacha . ” In this context , there is no other bug that goes by the same name . We also showed images of T . infestans to participants to aid their recall . During interviews , participants were asked if they recalled seeing triatomine bugs in their house or in their community for the specific years they lived in each place of residence . During focus groups , participants were asked to reach a consensus regarding the year that triatomine bugs first appeared and the areas most infested . Data on the number of households , estimated population , and year of insecticide application were collected by our research group in collaboration with the Arequipa Regional Office of the Ministry of Health ( MOH ) . The domiciliary infestation index ( DII ) is a community level variable equivalent to the number of infested houses divided by the total number of houses surveyed . We used ArcView 9 . 1 ( ESRI ) and existing maps to estimate the distance of each community from Arequipa and its surface area ( to calculate population density ) . Detailed notes were taken from focus group audio-recordings and the large sheets of paper , and synthesized into matrices in Excel by theme and by community to carry out further analysis . Audio-recordings of interviews were transcribed verbatim from digital audio recorders to word processing programs and analyzed using the grounded-theory approach . Grounded theory refers to theory that is developed inductively from a body of data , in contrast to theory that is derived deductively from grand theory and not necessarily based on data . The grounded theory approach is applied by reading textual data , in this case transcripts and field notes , in order to discover the main concepts or themes that were mentioned during data collection , allowing the data to reveal its message ( or theory ) instead of looking for confirmation of a previously-developed hypothesis [21] , [22] , [23] , [24] . Further information about grounded theory can be accessed via the web links listed in the references [25] , [26] . Two authors ( AB and GH ) created a code set based on the main themes that emerged in the interviews after an initial reading of the transcripts . The Atlas-ti software ( Scientific Software Development GmbH , 2005 ) was used to apply the codes to each interview transcript . The event history calendars themselves were also analyzed by entering the information into a single timeline in Excel , such that each decade from 1900 to the present year detailed the places of residence of participants and the presence of triatomine vectors in those places . The superimposition of the EHCs on a timeline allowed us to visualize patterns in migration and vector presence over time . Each coded interview was analyzed together with the interviewee's EHC and all focus group and interview data were analyzed by community and then across communities . During the EHC interviews , all moves were recorded , including a change of abode within the same community . For the purposes of this analysis , only a change of resident community for one month or more , whether temporary or permanent , was considered as a movement .
Fifty people provided in-depth interviews and 44 participated in focus groups . Interview participants had a median age of 52 and 47 years for males and females , respectively , with a range of 20 to 80 years . Focus group participants had a median age of 67 for males and 49 for females , with a range of 20 to 79 years ( Table 2 ) . Older community members were purposely recruited for focus groups , to ensure knowledge of historical events . Males moved more frequently than females , with a median of 3 . 0 ( Interquartile range ( IQR ) 1 . 0–4 . 0 ) lifetime moves for females and 4 . 0 ( IQR 2 . 5–5 . 5 ) for males . There was also more migration among residents of pueblos jóvenes: 80% of participants from pueblos jóvenes were in-migrants , while 60% of participants from pueblos tradicionales were non-migrants ( Table 3 ) . Typical patterns of migration are shown in Table 4 . Residents of pueblos jóvenes typically moved from a rural birth community to the Arequipa area early in life due to economic stress . Later moves took them to a pueblo jóven near Arequipa in search of cheap housing and then multiple , short-term moves continued throughout life in search of work . Some residents of pueblos tradicionales also moved from a rural area to Arequipa during childhood or adolescence , usually for schooling . Later moves were few since these residents tended to settle in the pueblo tradicional to focus on building agricultural enterprise and raising a family . The majority of participants who migrated from rural areas recalled raising farm animals in their birth places , and many continued to raise animals , especially guinea pigs , rabbits and poultry , in peri-urban Arequipa . Study participants indicated that the pueblos jóvenes of Arequipa's urban periphery were mostly settled by people from greater Arequipa and the northern Andean regions of Cusco and Puno , and in lesser numbers by people from the southern coastal/Andean region of Moquegua ( see map insert in Figure 2 ) . The formal founding dates of the pueblos jóvenes in this study ranged from 1970 to 1981 , although some residents had lived in these settlements since the 1960s . Migration from rural birthplaces to urban Arequipa was motivated by acute economic stress , with a few families sending their children away to work as early as age six . This early move was followed by migration later in life to a pueblo joven to acquire land and housing . Each quote presented in the Results section is followed by the participant's sex , age , type of current community , and the years corresponding to the movement . Two examples of interviewees who moved alone as children follow: Other participants moved with their entire family for work: Migration to pueblos jóvenes enabled early settlers to acquire land at little cost as squatters invading land . The pueblo tradicional of Tío Chico is located just below several hillside pueblos jóvenes , including Guadalupe: By contrast , residents who arrived after the original invasion bought an existing house or land parcel , rented a house or room , or lived with family members . Pueblos jóvenes are still expanding in geographic area and population , and currently are composed of multiple generations , including the original migrants , their children and grandchildren , and newly-arrived migrants . Pueblos tradicionales are made up primarily of people who have lived in the community since birth and whose families have lived there for several generations . As in pueblos jóvenes , the founding migrants arrived in search of land and housing . However they usually purchased this higher-cost land: Later migrants tended to move to pueblos tradicionales motivated by a return to family roots or the search for a calmer environment: Current migration dynamics are causing important changes to the population of pueblos tradicionales . Many people , especially young people , are moving out of pueblos tradicionales due to a lack of opportunities . In the two pueblos tradicionales in this study , out-migration has resulted in diminished populations and a shortage of agricultural workers . To compensate , land owners hire temporary workers who are often migrants to the peri-urban areas of Arequipa . In Tío Chico , these workers do not live in the town itself since there are few available houses and they can live more affordably in the surrounding pueblos jóvenes . In contrast , Quequeña , being further from the city , lacks housing options other than the town itself . Rental is therefore common since the property owners have moved and need people to watch over their properties and work their land . As a result , Quequeña is experiencing relatively recent in-migration , while Tío Chico is not . A constant across most participants' adult lives was migration to live with a partner or spouse and children: Across communities , the search for work was a constant that often spanned generations , with migration in early life by parents and in later life by participants . These participants described the almost constant movement of their fathers in search of work: Male participants described their own search for work opportunities and female participants discussed similar searches by their partners: Forty-six ( 92% ) of 50 interviewees had seen triatomine vectors ( locally known as chirimachas ) during their lifetimes in some place of residence; the four who did not report seeing chirimachas lived in the pueblo tradicional of Tío Chico ( Figure 3 ) . Participants reported seeing no triatomines in the highland Andes regions of Cusco and Puno , where many migrants were born . In addition , the urban center of Arequipa ( often the first place of residence for low-income migrants to Arequipa ) was described as vector-free . The earliest sighting of triatomines occurred in Moquegua in 1942 , a sending area for some migrants to urban Arequipa that is located south of the study communities ( Figure 2 ) . Other early memories came from the rural areas of Valle de Vitor and La Joya , both valleys west of Arequipa . The study community of Villa La Joya is a pueblo joven founded near the more rural and longer established community of La Joya . Villa La Joya had its first residents in the early 1960s , but chirimachas were not memorable until the town was much more populated in the late 1970s and 1980s . Closer to the urban center , reported vector presence showed a similar pattern of infestation following settlement , but much more recently . The first memories of urban vector presence from study participants were from the pueblo joven of Jacobo D . Hunter in 1968 , a peri-urban district settled in early squatter invasions . Years later , in 2002 , Hunter was the setting of a highly publicized child death due to acute Chagas disease [27] . Peri-urban pueblos jóvenes increased in number and size during the 1960s and 1970s , and vector presence was reported in the urban pueblos jóvenes of our study roughly 20 years following original settlement . Participants from Guadalupe ( founded in 1970 ) and Nueva Alborada ( founded in 1981 ) noted the first widespread appearance of triatomines in 1988 and 2002 , respectively . In pueblos tradicionales , vector reports followed a different pattern than in the pueblos jóvenes . Although their settlement dates back to the mid-1800s , focus group participants recalled first seeing vectors in Quequeña around 1978 . They noted that chirimachas were especially concentrated in the area of town around a communal stable and zones of relatively newer settlement by migrants . Tío Chico , in contrast , had very few reports of infestation , and much later ( from 2000 ) . The following quotes contrast the degree of infestation between Tío Chico and Guadalupe , the hillside pueblo joven located above Tío Chico . Participants specifically associated infestation with the presence of domestic animals . Residents reported continuous presence of chirimachas in all study communities except Tío Chico until insecticide spraying occurred ( Figure 3 ) . According to MOH vector control data , the pueblos tradicionales in our study had considerably lower infestation rates than pueblos jóvenes ( Table 1 ) . The high infestation rates in pueblos jóvenes parallel higher population densities of 10 , 300–15 , 000 inhabitants/km2 , compared to 1 , 700–3 , 300 inhabitants/km2 in pueblos tradicionales ( Table 1 ) .
The prevalence of Chagas disease varies widely among communities around Arequipa . High T . cruzi infection rates in children , reflecting recent transmission , were documented in several recently-formed pueblos jóvenes , while long-established pueblos tradicionales have almost no infection in children [12] . The proximate cause of infection heterogeneity is the difference in triatomine infestation rates [12] , [28] . We show here how migration and associated activities contribute to conditions promoting infestation in pueblos jóvenes . Human migration patterns thus constitute an important underlying determinant of Chagas disease risk in and around Arequipa . Most migrants to pueblos jóvenes came from areas without infestation and are unlikely to have brought the vector with them . However , community members later made multiple short- to medium-term moves , often to the valleys west of Arequipa for seasonal agricultural labor . These valleys are also the best-known historical foci of T . cruzi transmission in the region [29] . Migrants may have become infected while living temporarily in the valleys , or may have carried vectors back to their long-term communities in their belongings . An alternative , not mutually exclusive hypothesis is that vectors were always present in pueblos tradicionales prior to the construction of pueblos jóvenes , but not in large enough numbers to be a memorable event for study participants . When pueblos jóvenes quickly took over peri-urban hillsides , existing vector populations may have exploded . Rural migrants from highland areas brought their domestic animal husbandry practices to the city with them , raising small animals such as guinea pigs and rabbits for sale or for personal consumption , and keeping them in small yards close to human dwellings . At the time of vector control spraying in 2004 , the pueblo joven of Guadalupe contained a total of 5 , 006 domestic animals ( predominantly guinea pigs , rabbits , poultry and dogs , but also sheep and cows ) ; the presence of guinea pigs , in particular , was associated with an increased risk of infestation both in the animal enclosure and in the adjacent human house [13] . The high density of animals provides an abundance of blood meal sources to support large vector populations , and potentially contributes to Chagas disease transmission , since all except poultry are susceptible to T . cruzi infection . Interestingly , one pueblo joven , Nueva Alborada , was highly infested with vectors , but none of the 1 , 460 insects examined during the Ministry insecticide application campaign were carrying T . cruzi [30] . Focus group participants in Nueva Alborada reported chirimachas in the community only back to 2000 , while focus groups in the other two pueblos jóvenes recalled insects in their communities for a much longer period . It is possible that , given the short history of vector infestation in Nueva Alborada , the parasite has yet to be successfully introduced into this community . In contrast , in Guadalupe and Villa La Joya , the longer history of infestation may have led to single or multiple introductions of the parasite in these communities . The relationship between time since introduction of the vector and presence of T . cruzi merits further epidemiological investigation . Two final considerations include the possible transformation of long-standing pueblos tradicionales by an influx of low-income migrants and the urbanization of rural areas . Although many of the original residents had emigrated from both traditional towns , we observed increased population density and levels of in-migration that were similar to those in pueblos jóvenes only in Quequeña , where low-income migrants are filling the population void . In contrast , no new migrants had moved to Tío Chico , because low-income farm workers who cultivate fields surrounding Tío Chico can live more cheaply in a nearby pueblo joven . The case of Quequeña draws attention to a previously undescribed phenomenon of population loss and replacement by low-income migrants , creating the human settlement pattern of a pueblo jóven within a pueblo tradicional , and providing insight into why T . infestans and T . cruzi are prevalent in Quequeña . These complex patterns of human migration and triatomine infestation demonstrate the conditions underlying the urbanization of Chagas disease in Arequipa . Factors common to the infested pueblos jóvenes include recent , rapid settlement , high human and animal density , and frequent temporary migration to Chagas disease-endemic areas . However , some pueblos tradicionales are also undergoing similar processes . As peri-urban areas in South America continue to grow , vector control programs must remain vigilant against reinfestation , as well as infestation of communities not previously recognized to be at risk , such as the pueblos tradicionales of Arequipa . In addition , our data point to at least three potential interventions for improving vector control in Arequipa: 1 ) Intensifying vector surveillance efforts in areas with highly mobile populations; 2 ) Creating educational campaigns for migrant workers to Chagas-endemic areas; and 3 ) Fomenting collaboration between the Arequipa Region's Ministry of Health and Ministry of Housing to monitor the emergence of new pueblos jóvenes for their inclusion in the vector surveillance system . | Chagas disease affects 8–10 million people in the Americas . Although transmission was previously limited to the rural poor , Chagas increasingly affects urban populations , especially near the city of Arequipa , Peru . We interviewed residents of five communities to learn about why and when they migrated to the city and how their movements may link to Chagas vectors and to explore the settlement patterns of shantytowns and traditional towns . We found that migrants to shantytowns were unlikely to introduce Chagas vectors to the city upon first arrival . Frequent seasonal moves , however , took shantytown residents to valleys surrounding Arequipa where vectors are prevalent . In addition , the settlement pattern of shantytowns and the practice of raising domestic animals create a favorable environment for vectors . Finally , population loss and replacement by low-income migrants in one traditional town has created the human settlement pattern of a shantytown . This study exposes potential links between population dynamics and Chagas vector infestation . Suggested methods for improving vector control include focusing future vector surveillance in areas with mobile populations , creating educational campaigns for migrant workers to Chagas-endemic areas , and fomenting collaboration between the Arequipa Ministries of Health and Housing to ensure the inclusion of new shantytowns in vector surveillance . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"infectious",
"diseases/protozoal",
"infections",
"public",
"health",
"and",
"epidemiology/social",
"and",
"behavioral",
"determinants",
"of",
"health",
"public",
"health",
"and",
"epidemiology/infectious",
"diseases"
] | 2009 | Chagas Disease, Migration and Community Settlement Patterns in Arequipa, Peru |
Virus infections can result in a range of cellular injuries and commonly this involves both the plasma and intracellular membranes , resulting in enhanced permeability . Viroporins are a group of proteins that interact with plasma membranes modifying permeability and can promote the release of viral particles . While these proteins are not essential for virus replication , their activity certainly promotes virus growth . Progressive multifocal leukoencephalopathy ( PML ) is a fatal demyelinating disease resulting from lytic infection of oligodendrocytes by the polyomavirus JC virus ( JCV ) . The genome of JCV encodes six major proteins including a small auxiliary protein known as agnoprotein . Studies on other polyomavirus agnoproteins have suggested that the protein may contribute to viral propagation at various stages in the replication cycle , including transcription , translation , processing of late viral proteins , assembly of virions , and viral propagation . Previous studies from our and other laboratories have indicated that JCV agnoprotein plays an important , although as yet incompletely understood role in the propagation of JCV . Here , we demonstrate that agnoprotein possesses properties commonly associated with viroporins . Our findings demonstrate that: ( i ) A deletion mutant of agnoprotein is defective in virion release and viral propagation; ( ii ) Agnoprotein localizes to the ER early in infection , but is also found at the plasma membrane late in infection; ( iii ) Agnoprotein is an integral membrane protein and forms homo-oligomers; ( iv ) Agnoprotein enhances permeability of cells to the translation inhibitor hygromycin B; ( v ) Agnoprotein induces the influx of extracellular Ca2+; ( vi ) The basic residues at amino acid positions 8 and 9 of agnoprotein key are determinants of the viroporin activity . The viroporin-like properties of agnoprotein result in increased membrane permeability and alterations in intracellular Ca2+ homeostasis leading to membrane dysfunction and enhancement of virus release .
Replication of viruses involves an extracellular step in the viral life cycle , involving the release of virus particles from infected cells and subsequent infection of target cells . Most non-enveloped viruses exit their host cells by lytic process , which involves breakdown of the cell membrane and is associated with cell death [1] . JC virus ( JCV ) is the causative agent of progressive multifocal leukoencephalopathy ( PML ) , and belongs to the family of polyomaviruses , which also includes simian virus 40 ( SV40 ) and BK virus ( BKV ) . Polyomaviruses have non-enveloped icosahedral-shaped capsids of about 40 nm in diameter . It has been previously suggested that extracellular exit of the mature progeny virions of SV40 and JCV , which efficiently proliferate in the nuclei , occurs when cells disintegrate or rupture as part of their dying process . However , it remains unclear whether these virions employ specific molecular mechanism ( s ) which may contribute to or regulate cell lysis . Cell lysis is presumably the ultimate result of an increase in plasma membrane permeability [2] . It has long been considered that either the bulk of viral gene expression or the formation and accumulation of progeny virus particles may be responsible for enhancing membrane permeability and lysis of the cell . However , more recent studies have suggested that individual viral proteins may contribute to the enhancement of plasma membrane permeability and release of progeny virions of a number of non-enveloped viruses , including poliovirus , rotavirus , and coxsackievirus [3] [4] [5] [6] [7] . It has been shown that several viral proteins with membrane permeabilizing properties share common characteristics and these proteins have been named “viroporins” [6] . The proteins share a number of features in structure and function . Viroporins are integral membrane proteins which vary in size from about 60–120 amino acids , possessing at least one hydrophobic stretch able to form an amphipatic α-helix . Viroporins interact with membranes to increase permeability to ions and other small molecules [2] [6] . After their insertion into membranes , viroporins tend to oligomerize to create a hydrophilic pore [8] [9] [10] [11] , and these activities result in release of progeny virions . The late coding region of JCV encodes a small and basic regulatory protein , agnoprotein , whose functions in the virus life cycle remains unclear . Studies on the SV40 agnoprotein have suggested that the protein may contribute to viral replication at various stages including transcription , translation , and processing of late viral proteins [12] [13] [14] [15] , assembly of virions [16] [17] [14] , and viral propagation [18] [19] . Agnoprotein has been shown to localize to the cytoplasmic and perinuclear regions in infected cells [20] , whereas the other viral proteins are exclusively expressed in the nuclei [21] [20] . Recent observations also suggest that JCV agnoprotein may be involved in release of progeny virions and promote the propagation of JCV [22] [23] [24] . Furthermore , a small interfering RNA ( siRNA ) specific for agnoprotein mRNA was found to inhibit JCV infection [25] [26] . In the present study , we have investigated the role of the JCV agnoprotein in virion release . Comprising some 71 amino acids , agnoprotein has a hydrophobic transmembrane domain that interacts with and expands the lipid bilayer as homo-oligomers . Furthermore , we have found that agnoprotein acts as a viroporin and its expression induced plasma membrane permeabilization and promoted JCV progeny virion release .
Agnoprotein plays critical roles in viral propagation at multiple steps in JC virus replication [27] [22] [23] [24] . Therefore , we compared the propagation properties of an agnoprotein deletion mutant ( ΔAgno ) virus with those of the wild type ( WT ) virus by using a JCV growth assay , which is an established method for measurement of JCV infectivity [28] [24] . In the ΔAgno mutant , the translation initiation codon ( ATG ) was replaced with the stop codon ( TAA ) , and the mutant does not produce agnoprotein . VP1 protein of cells transfected with ΔAgno mutant was localized in the nuclei similar to that of WT Agno transfected cells ( Figure S1 ) at 4 days after transfection , suggesting that agnoprotein did not affect the intracellular localization of VP1 at this time point . Although the percentage of VP1 of JCV positive cells transfected with either WT or ΔAgno mutant was similar at 4 days after transfection , the percentage of VP1 positive cells transfected with ΔAgno mutant was significantly lower than that of WT at 9 days after transfection ( Figure 1A ) , suggesting that the viral propagation was inhibited in the absence of agnoprotein and which is in accordance with previous reports [27] [25] [24] . We also found that the expression level of VP1 in the culture supernatant from the cells transfected with ΔAgno mutant was substantially decreased compared to that of WT ( SUP in Figure 1B ) . Together , these observations show that agnoprotein plays an important role in the release of virions from infected cells . Previously , we reported that agnoprotein is mainly localized in the cytoplasm and perinuclear regions of JCV infected cells [27] [20] [23] . To delineate in more detail the intracellular localization of agnoprotein , we performed immunofluorescence studies using human fetal glial SVG-A cells infected with JCV at 3 days after infection . Confocal microscopy revealed that agnoprotien immunoreactivity was present in the perinuclear region and extended into the cytoplasm in a mesh-like pattern ( Figure 2A and S1 ) . Although its localization was relatively heterogenous , most of the agnoprotein consistently overlapped with that of calreticulin , an ER marker and specifically in the peripheral regions of the cytoplasm ( Figure 2A ) . We also performed iodixanol density gradient analysis of the membrane fraction of JCV-infected IMR-32 cells . In contrast to other JCV encoded proteins , such as VP1 and Large T antigen , agnoprotein and BiP , which is an ER marker , were detected in the same fractions ( Figure 2B ) , suggesting that agnoprotein predominantly localizes at the ER . We next examined the intracellular-localization of agnoprotein at various time after infection . It could be shown that almost all agnoprotein was distributed at the ER at 2 days after infection , which is designated as an ER pattern ( Figures 2A and C ) . However , at the later stages ( at 5 days after infection ) of infection the localization of agnoprotein in the ER changed to become more diffusely localized in the cytoplasm ( Figures 2C and D ) ( designated as diffuse pattern ) . These observations suggested that while agnoprotein originally localizes at the ER , it's intracellular localization changes to involve other intracellular compartments in time-dependent manner . Next , we attempted to define in more detail the properties of agnoprotein . Analysis of subcellular fractions of 293AG cells which are agnoprotein-inducible cell lines with doxycycline ( DOX ) treatment [22] showed that agnoprotein was mainly detected in the microsomal compared with the cytosol fractions ( Figure 3A ) . We then examined the interaction between agnoprotein and microsomes by treatment of the microsome membrane fractions with various chemical reagents that selectively disrupt interactions on the surface without extracting proteins from phospholipids bilayers ( Figure 3B ) . Treatments with high salt concentration ( 1 M KCl ) and sodium carbonate buffer ( pH 11 ) , as well as incubation with 2 M urea , all abolished the membranous association of BiP which has no hydrophobic transmembrane domain ( Figure 3B , lower panel ) . In contrast , agnoprotein and calnexin , which have a hydrophobic transmembrane domain , were not removed from the microsome membrane even in the presence of these reagents ( Figure 3B , upper and middle panels ) suggesting that agnoprotein is embedded within the lipid bilayer . Consistent with this , agnoprotein was recovered in the detergent phase along with calnexin in the phase separation experiments using Triton X-114 [29] [30] ( Figure 3C ) . As previously shown the intracellular localization of agnoprotein changes in time-dependent manner ( Figures 2C and D ) . The anterograde transit of vesicular proteins within the early secretory pathway is known to connect the ER to plasma membrane trafficking [31] . To determine if agnoprotein is involved in the secretory pathway , we examined the steady-state cell surface expression of agnoprotein . Non-permeabilized HeLa cells transfected with WTAgno were subjected to immunofluorescence staining with antibody which recognizes the agnoprotein C-terminus [20] . Agnoprotein was detected at the plasma membranes of both non-permeabilized HeLa ( Figure 3D ) and SVG-A ( Figure 3E ) cells . The steady-state cell surface expression of agnoprotein was further supported by flow cytometry studies ( Figure 3F ) . These observations suggested that agnoprotein is in part co-translationally inserted into the ER membrane and transported to the plasma membrane at the later stage of infection with the C-terminus of agnoprotein at the extracellular surface . We next sought to identify the motif ( s ) of agnoprotein which are necessary for targeting the ER . A Kyte and Doolittle hydrophobicity plot of agnoprotein revealed that the middle region of agnoprotein contains the only hydrophobic region . This is comprised of 18 amino acids and would serve as a potential transmembrane segment ( Figure 4A ) . To determine the domain of agnoprotein important for its ER distribution , we generated a series of mutants as fusion proteins with the GST-EGFP tag ( ∼50 kDa ) ( Figure 4B ) . Immunoblot analyses of GST-EGFP–fused agnoprotein and its mutants were prepared from the ER-nuclear fraction of transfected HEK293 cells by sucrose density centrifugation . WT and N46 mutant were detected in the ER-rich fractions , but the C6 mutant and GST-EGFP were not ( Figure 4C ) . This result suggested that the N-terminus of agnoprotein was necessary for targeting to the ER . We have demonstrated that agnoprotein has an extracellular carboxyl terminal ( C-terminal ) domain ( Figures 3D , E , and F ) and that the N-terminal region of agnoprotein was important for the ER targeting ( Figure 4C ) . These results suggest that agnoprotein might have a cytoplasmic domain at the amino terminus and should be considered as an integral transmembrane protein type II [32] . In this situation positively charged amino acids at the N-terminal region would be important for determining the orientation of the transmembrane segment [33] . Agnoprotein has some basic residues in the N-terminal 24 amino acids , and to determine the residues responsible for the subcellular localization of agnoprotein , amino acid point mutations of agnoprotein where generated by site-directed mutagenesis ( Figure S3 ) , and the subcellular localizations in 293T cells transiently transfected with these constructs and pDsRed-ER were analyzed by confocal microscopy ( Figure 5 ) . The fluorescent intensities of GST-GFP fused with WT and the N46 , and RK8AA mutants corresponded with those of DsRed-ER , indicating that WT , N46 , and RK8AA mutants localized at the ER . In contrast , the fluorescence intensities of GST-GFP fused with the other mutants did not correlate with those of DsRed-ER . These observations demonstrated that the basic amino acid residues of the N-terminal region of agnoprotein are necessary for its subcellular localization and that agnoprotein is indeed a type II integral transmembrane protein . Viroporins are responsible for membrane leakiness late in infection . Typically , viroporins are comprised of some 60–120 amino acids , contain a highly hydrophobic domain , and may also contain a stretch of basic amino acids . The insertion of viroporins into membranes is followed by their oligomerization and this is thought to be critical for membrane destabilization [2] [6] . Agnoprotein is an integral membrane protein that has some of the expected features of viroporins . To determine whether agnoprotein acts as a viroporin , we examined the formation of agnoprotein homo-oligomers by several methods: 1 ) coimmunoprecipitation of epitope-tagged agnoproteins ( Figure 6A ) , 2 ) chemical cross-linking of SVG-A cells infected with JCV ( Figure 6B ) , and 3 ) intermolecular fluorescence resonance energy transfer ( FRET ) studies using agnoprotein fused with either Venus ( YFP-Agno ) or sECFP ( CFP-Agno ) at the NH2-terminus ( Figures 6C and D ) . Myc-tagged agnoprotein was coimmunoprecipitated with Flag-tagged agnoprotein using anti-Flag antibody ( Figure 6A ) , suggesting that agnoprotein forms homo-oligomers . SVG-A cells infected with JCV were incubated with a crosslinker reagent , disuccinimidyl suberate ( DSS ) or buffer alone , and homogenized with Triton X-100 containing lysis buffer following immunoblotting . The immunoblotting revealed that there were multiple bands , which migrated with the expected positions of monomers ( approximately , 8 kDa ) , dimers ( 16 kDa ) , trimers ( 24 kDa ) , tetramers ( 32 kDa ) , and pentamers ( 40 kDa ) ( Figure 6B ) . Furthermore , FRET assays confirmed that agnoprotein forms homo-oligomers at the cytoplasmic organelles where agnoprotein is known to localize ( Figures 6C and D ) . To determine whether agnoprotein is capable of producing permeabilization of the plasma membrane , we examined plasma membrane structures using Merocyanine 540 ( MC540 ) , which is a lipophilic fluorescence dye that binds to the outer leaflet of plasma membranes [34] . Several studies in model membranes have demonstrated that its fluorescence characteristics are sensitive to subtle differences in lipid packing [35] [36] . We used flow cytometry to examine MC540 binding of HeLa cells transfected with agnoprotein ( Agno ) or mock plasmid ( Mock ) , or without transfection ( NT ) . The MC540 intensity in Agno cells was significantly higher than in Mock or NT cells ( Figure 7A ) , suggesting that agnoprotein disrupts lipid packing and modifies the structure of the plasma membrane . To extend these findings , we performed a Hygromycin B ( HygB ) permeability assay . HygB is a general inhibitor of translation , and intact mammalian cells are impermeable to HygB when exposed for short time periods at low concentrations . However , HygB will enter mammalian cells with permeabilized plasma membranes [37] . HeLa cells were transfected with pCFPNLS-Agno or mock plasmid . The plasmid contains the internal ribosome entry site ( IRES ) of the encephalomyocarditis virus ( ECMV ) between the multiple cloning sites ( MCS ) and the CFPNLS [sECFP with nuclear localization signal ( NLS ) of simian virus 40 large T-antigen fused to its C-terminus] following the immediate early cytomegalovirus promoter which permits both the inserted gene in the MCS and the CFPNLS gene to be translated from a single bicistronic mRNA . The cells transfected with the plasmids were identified by expression of sECFP in the nucleus . Following incubation for 72 h , cells were then radiolabeled with [35S] Met-Cys for 2 h in the absence or presence of HygB . Cell extracts were harvested , and the protein synthesis marker CFPNLS was detected by immunoprecipitation with an anti-GFP antibody that cross-reacts with CFP . In cells transfected with pCFPNLS-Agno , radio-labeled CFPNLS levels were markedly decreased in the presence of HygB; in contrast in mock transfected cells , CFPNLS levels were unaffected by HygB ( Figures 7B and C ) . Thus , agnoprotein expression clearly altered the cell membrane permeability for HygB . This increase in membrane permeability for HygB was also confirmed using SVG-A cells ( Figure S4A and S6A ) . Viral proteins , which can enhance plasma membrane permeabilization to small molecules such as HygB , can also induce an increased permeability to ions , such as Ca2+ [2] [6] . We measured the influx of extracellular calcium using the intramolecular FRET-based Ca2+ indicator , yellow cameleon 3 . 60 ( YC3 . 60 ) [38] . Ca2+ influx into cells was induced by addition of CaCl2 to the extracellular medium in the presence , but not in the absence , of agnoprotein ( Figures 7D , E , and F; the video files can be viewed as supplementary Videos S1 and S2 ) . These studies show an increased permeability for Ca2+ and it can be concluded that agnoprotein enhances membrane permeability to ions as well as to small molecules . We also investigated whether agnoprotein possesses inherent properties that result in cell permeabilization and death independent of the signaling pathways within the host organism . To address this question , the cell permeability of Escherichia coli ( E . coli ) was examined after expression of either agnoprotein containing His tag at the NH2-terminus or VP1 without a His tag driven by a T7 promoter as has been previously described for SV40 VP4 [39] . The integrity of the E . coli double-membrane barrier was analyzed by determining the sensitivity of protein synthesis to HygB . To inhibit the E . coli RNA polymerase and prevent the synthesis of endogenous bacterial proteins , rifampicin was also added during the induction of E . coli growth . The expression of VP1 had no effect on E . coli membrane permeability , as the synthesis of 35S-labeled VP1 was not inhibited by addition of HygB ( Figure 7G , left panel , arrow head ) . In contrast , agnoprotein expression resulted in the permeabilization of E . coli , as the synthesis of 35S-labeled agnoprotein was clearly inhibited in the presence of HygB ( Figure 7G , right panel , arrow head ) . Cellular membrane permeabilization can lead to cell lysis or cell death . Therefore , we investigated the viability of E . coli after the induction of either VP1 or agnoprotein expression to determine whether these viral proteins possess lytic properties . The induction of VP1 expression did not cause obvious lysis of the bacteria , as no decrease in the OD was observed ( Figure S4B ) Instead , the induction of agnoprotein showed a relatively low increase rate of OD compared to that of VP1 ( Figure S4B ) . These observations indicated that agnoprotein expression does not cause lysis but would appear to have some toxic effect on E . coli growth . While the molecular basis of this effect is unclear it may be due to or be related to the permeabilization of the bacterial membrane . To further investigate the viroporin activity , we performed a HygB permeability assay with agnoprotein mutants ( N46 mutant and RK8AA mutant ) , which localize at the ER similar to WT protein ( Figure 5 ) . In N46 mutant- or WT Agno-transfected HeLa cells , the amount of radio-labeled CFPNLS was markedly decreased in the presence of HygB compared to its absence ( Figure 8A ) . In contrast , the levels of CFPNLS were unaltered in mock or RK8AA- transfected cells ( Figure 8A ) , suggesting that WT Agno and N46 , but not RK8AA , enhances membrane permeabilization . To further demonstrate the importance of this viroporin activity , we generated viral genomes containing the RK8AA mutation of agnoprotein ( RK8AAJCV ) and performed JCV growth assays . Viral growth was measured by monitoring the percentage of VP1 positive cells ( Figure 8B ) . At 3 and 5 days post-transfection , the percentage of VP1 positive cells were similar in wtJCV- and RK8AAJCV-transfected cells ( Figure 8B ) . However , by 9 days the percentage of VP1 positive cells of RK8AAJCV-transfected cells had significantly decreased compared to that at day 5 , whereas that continued to increase in wtJCV-transfected cells ( Figure 8B ) , suggesting that RK8AA mutation of agnoprotein was associated with impaired viral propagation . The amount of VP1 in the culture supernatant of RK8AAJCV-transfected cells was , as expected , substantially decreased ( SUP in Figure 8C and S6B ) . The intracellular localization and expression level of agnoprotein of RK8AAJCV-transfected cells was similar to those of wtJCV-transfected cells ( Figures S5A and B ) . These results suggested that RK8AAJCV has a defect in the release of progeny virus . Transmission electron microscopy was carried out to evaluate whether the RK8AA mutant of agnoprotein influences the virion assembly . This revealed the presence of virus particles of 40-nm in diameter in the nuclei of RK8AAJCV-transfected cells similar to that observed in wtJCV-transfected cells at 5 days post-transfection ( Figure 8D , left and middle columns ) . In contrast , particles were not observed in cells without transfection ( Figure 8D , right columns ) . The virions extracted from the cells transfected with RK8AAJCV were able to infect to the SVG-A cells similar to that of wtJCV and thus were not defective ( Figure S5C ) . These findings show that the inability of the RK8AAJCV to propagate viral infection is due to a defect ( s ) in virion release . Taken together , Arg-8 and Lys-9 in the NH2-terminus of agnoprotein are necessary for viroporin activities , which permeabilize the cell membrane and release virions from the JCV infected cells .
The amino acid sequence of JCV agnoprotein shares ∼60% identity with that of the other polyomaviruses SV40 and BKV . In contrast the sequences of other viral proteins , such as VP1 or large T Ag , are highly conserved ( 80–90% identity ) . The C-terminal amino acid sequences of the agnoproteins are unique in the individual viruses , whereas those at the N-terminal 40 amino acids exhibit around 90% identity . As the subcellular localizations of the agnoproteins are almost identical , this suggests the existence of a functional domain in the N-termini . JCV agnoprotein is the only viral transcript expressed in cytoplasm rather than the nucleus , and especially in perinuclear region [20] . In contrast to the other viral late proteins , agnoprotein is not incorporated into in the viral capsid and is undetectable in purified infectious virions . In this report we demonstrate that JCV agnoprotein is an integral transmembrane protein type II targeted to the organelles in an exocytic route , involving the ER and subsequently localizing to the plasma membrane in time-dependent manner . The integral transmembrane protein type II agnoprotein has a cytoplasmic amino terminus and an extracellular carboxyl terminus [32] . Positively charged amino acids at the N-terminal region are important for determining the orientation of the transmembrane segment [33] . In the N-terminus of agnoprotein , there are several positively charged amino acids , and these residues with the exception of Arg-8 and Lys-9 contribute to the subcellular localization of the protein . Viruses produce a number of injuries during infection of susceptible cells . Some of these affect cell membranes , and a typical feature observed during the replication of a number of animal viruses is enhanced membrane permeability . Several viral gene products are considered to be responsible for these changes and these include proteases , glycoproteins , and viroporins [2] [6] . Viroporins are small , highly hydrophobic , virus-encoded proteins that interact with membranes modifying the cell permeability to ions or other small molecules . Typically , viroporins are comprised of some 60–120 amino acids and contain a highly hydrophobic domain . The insertion of viroporins into membranes followed by their oligomerization leads to membrane destabilization , thus enhancing permeability [2] [6] . We have found that agnoprotein has all of the features of viroporins and is capable of enhancing the membrane permeability for Hygromycin-B and Ca2+ . In addition , agnoprotein enhanced the plasma membrane binding of MC540 which is a lipophilic fluorescence dye binding to the outer leaflet of plasma membranes [34] . While the changes of binding of MC540 support alterations of membrane structures , the studies do not allow clear definition of any particular mechanism [40] . Our results also show that agnoprotein can permit small molecules [molecular weight ( MW ) under 1 , 000 Da] to enter the cytoplasm . Because the MW of virions are much larger than 1 , 000 Da , they may be released from the cells not through the direct effect of membrane permeabilization but perhaps indirectly by the alteration of cytoplasmic concentrations of monovalent cations , which can provoke membrane depolarization , leading to cell lysis [7] . In SV40 , VP2 and VP3 , but not VP1 , possess membrane-permeabilizing activity in prokaryotic cells [41] . Furthermore , the ability to induce bacterial lysis following permeabilization was an exclusive property of VP3 , suggesting that membrane permeability changes induced by VP3 , and perhaps VP2 as well , are largely responsible for the necrosis resulting from SV40 infection and that the lytic activity of VP3 may be necessary for the release of progeny virions [41] . In this study we also show that JCV agnoprotein possesses bacterial membrane-permeabilizing activity . However , agnoprotein could not induce bacterial lysis . Thus , is possible that cell-lytic activity may require expression of other viral proteins in addition to agnoprotein during JCV infection . Some viroporins contain a stretch of basic amino acids that act as a detergent [6] and agnoprotein has some clusters of basic residues in the NH2-terminal 24 amino acids . The R4A and KKR22AAG mutants of agnoprotein did not distribute to the ER , suggesting that these residues are important for the correct intracellular localization of agnoprotein . Although the RK8AA mutant of agnoprotein localized at the ER as well as WT , this mutant did not have the viroporin activity . These observations indicate that the “RK” residues at amino acid positions 8 and 9 of agnoprotein are not necessary for the intact intracellular localization but contribute to the plasma membrane permeability . These residues may serve as a detergent , modify the membrane structure , or interact with other molecules in host cells . However , the exact functions of these basic amino acids will further investigation and these studies will certainly help elucidate the molecular mechanisms of the viroporin activity . Furthermore , previous studies with other polyomaviruses have implied a direct interaction of agnoprotein and VP1 [42] , and indeed JCV agnoprotein does interact with GST-fused VP1 synthesized in E . coli [our unpublished observation] . These interactions would appear to be in addition to the role of agnoprotein as a viroporin . Thus agnoprotein can be considered to be a multifunctional auxiliary protein [43] , and studies on other polyomavirus agnoproteins have suggested that the protein may contribute to a variety of stages of the viral life cycle including assembly of virions and viral propagation [14] [16] [17] [18] [19] . Thus , the function of agnoprotein and the interaction with VP1 will require further investigation to clarify the roles of JCV agnoprotein in virion assembly . In summary , our studies demonstrate that agnoprotein acts as a viroporin resulting in plasma membrane permeabilization and virion release . The permeabilization induced by agnoprotein is crucial for viral propagation and could be a potential target for a therapeutic intervention in PML . Furthermore , our study highlights that the mechanism of virion release of a non-enveloped DNA virus is highly regulated by a single viral protein .
For expression of JCV agnoprotein in mammalian cells , the cDNA of JCV agnoprotein was amplified by polymerase chain reaction ( PCR ) using a plasmid encoding the JCV genome , pJC1->4pJCV [VG015 , Health Science Research Resources Bank ( HSRRB ) ] and subcloned into a pGST-EGFP plasmid including GST into the Kpn I and Bam HI sites of pEGFP-N1 ( Clontech Mountain View , CA ) ; pCXSN plasmid which was constructed by removing myc-tag from pCMV-myc ( Clontech ) and adding Xho I , Sal I , and Not I recognition sites; pCXSN-FlagN which was constructed by adding Flag-tag to pCXSN plasmid at the 5′ region of Xho I site; pCXSN-MycN which was constructed by adding Myc-tag to pCXSN plasmid at the 5′ region of Xho I site; pERedNLS ( kindly provided by Dr . M . Matsuda ) ; and pCFPNLS which was constructed by replacing DsRedExpress with sECFP ( kindly provided by Dr . A . Miyawaki ) ; pCXSN-YFP/CFP which constructed by insertion of Venus ( kindly provided by Dr . A . Miyawaki ) or sECFP to the Xho I site of pCXSN and adding linker sequences ( RSTGNSADGGGGSGGSGGSGGGSTQGGSSGTGTAAENSGNSRTK ) at the 3′ region of Venus or sECFP . Deletion mutants of agnoprotein were constructed by PCR with KODplus polymerase ( Toyobo , Tokyo , Japan ) . The substitution mutants of agnoprotein were constructed by using the QuickChange site-directed mutagenesis kit ( Stratagene ) according to the manufacture's procedure . The primers used for generating mutants of agnoprotein were ( 5′ -> 3′ ) : 1 , ctcagatctaatggttcttGCCcagctgtcacg and ctggtaccgtagcttttggttcaggcaaagc ( R4A ) ; 2 , ctcagatctaatggttcttcgccagctgtcaGCTGCTgct and ctggtaccgtagcttttggttcaggcaaagc ( RK8AA ) ; 3 , gtaaaacctggagtggaactGCAGCAGGAgctcaaaggatttta and taaaatcctttgagcTCCTGCTGCagttccactccaggttttac ( KKR22AAG ) . The following plasmids containing wild type ( WT ) agnoprotein were used as PCR templates: pGST-EGFP-Agno , pCXSN-Agno , and pERedNLS-Agno . The plasmid containing the genome of JCV Mad1-SVEΔ ( pUC19-Mad1SVEΔ ) was kindly provided by Dr . W . J . Atwood [28] [44] . Agnoprotein with mutated viral genomes ( ΔAgno and RK8AA ) were generated by site-directed mutagenesis using PCR with pUC19-Mad1SVEΔ , which has WT agnoprotein , and KODplus polymerase . In the ΔAgno mutant , the translation initiation codon ( ATG ) of agnoprotein was changed to the stop codon ( TAA ) by base substitution , and the ΔAgno mutant viral DNA is defective in the production of agnoprotein . Primers used for mutagenesis were ( 5′-> 3′ ) : 1 , TAAgttcttcgccagctgtcacg and ggccagcggtacctgtggaat ( ΔAgno ) . 2 , GCTgctttctgtgaaagttagtaaaacctgg and AGCtgacagctggcgaagaaccatg ( RK8AA ) . Mismatched nucleotides are shown in uppercase letters . Successful mutagenesis was confirmed by sequencing . A DsRed-ER expressing plasmid ( pDsRed-ER ) was obtained from Clontech . All the constructs and materials used in the experiments were described in Table S1 . Human embryonic kidney 293 cells with SV40 T antigen ( HEK293T ) , human cervical carcinoma cells ( HeLa ) , and human neuroblastoma cells IMR-32 cells were obtained from the HSRRB . A JCV-producing cell line ( JCI cells [45] ) and SV40-transformed human glial SVG-A cells ( kindly provided by Dr . W . J . Atwood ) [46] were also used . All cells were maintained under 5% CO2 at 37°C condition in Dulbecco's minimal essential medium ( DMEM ) supplemented with 10% heat-inactivated fetal bovine serum ( FBS ) , 2 mM L-glutamine , penicillin , and streptomycin ( Sigma ) . Establishment and maintenance of HEK293 cells expressing JCV agnoprotein in an inducible manner ( 293AG cells ) was described previously [22] [23] . For virus preparation , JCI cells or JC virus-infected SVG-A cells were harvested and suspended in Tris-HCl ( pH 7 . 5 ) containing 0 . 2% bovine serum albumin ( BSA ) , frozen and thawed three times , and then treated with 0 . 05 U/ml of neuraminidase type V ( Sigma ) at 37°C for 16 h . After incubation at 56°C for 30 min , cell lysates were centrifuged at 1 , 000×g for 10 min . The supernatant was quantified by hemagglutination ( HA ) assays and stored at −80°C until use . Rabbit anti-JCV agnoprotein , anti-JCV VP1 , and anti-JCV Large T polyclonal antibodies were produced as described previously [47] [20] [48] . Alexa 594–labeled anti-agnoprotein antibody was used for double-immunofluorostaining with another rabbit polyclonal antibody . Mouse anti-BiP and anti-Calnexin monoclonal antibodies were purchased from BD Transduction Laboratories . Mouse anti-actin ( MAB1501R ) monoclonal antibodies were purchased from Chemicon International . Mouse anti-Flag ( M2 ) and anti-alpha tubulin monoclonal antibodies were purchased from Sigma . Goat anti-GFP polyclonal antibody was purchased from Rockland . Rabbit anti-calreticulin polyclonal antibody was purchased from Calbiochem . Mutant and WT viral DNA were linearized at the Bam HI site , and equal amounts of viral DNA were transfected into permissive SVG-A cells by Fugene HD ( Roche Diagnostics ) reagents according to the manufacturer's instructions . Viral growth at the indicated time points was monitored by indirect immunofluorescence of VP1 . For quantification of viral particle release , the culture supernatant was collected at each indicated time point and ultracentrifuged in a Beckman TLA-100 . 3 rotor at 80 , 000 rpm for 60 min at 4°C . The pellet fraction and whole cell lysates ( WCL ) were analyzed simultaneously by immunoblotting . Results were confirmed by at least three independent experiments . Cells were fixed with 3% paraformaldehyde ( PFA ) in phosphate-buffered saline ( PBS ) , permeabilized with 0 . 5% Triton X-100 in PBS , and incubated at room temperature with 1% BSA in PBS . The cells were then incubated with primary antibodies , followed by Alexa 488– or Alexa 594–labeled goat antibodies to rabbit IgG or with Alexa 488– or Alexa 594–labeled goat antibodies to mouse IgG ( Molecular Probes ) . The cells were observed with a confocal laser-scanning microscope ( Olympus , Tokyo , Japan ) . For analysis of colocalization between GST-EGFP proteins and DsRed-ER , the 293T cells transfected with pGST-EGFP-Agno or mutant plasmid and pDsRed-ER were fixed with 3% PFA in PBS at 48 h post-transfection . The cells were then observed with a confocal laser-scanning microscope . The fluorescence intensity along a white straight line of the captured images was measured by using FV10-ASW ver . 1 . 6 imaging software ( Olympus ) . Cell surface immunofluorescent staining was previously described [49] . Briefly , cells were incubated in DMEM with 10% fetal calf serum and an appropriate primary antibody for 1 h at 4°C . After washing three times with cold PBS , the cells were fixed with 3% PFA and then stained with Alexa488-conjugated anti-mouse IgG antibody . Microsome fractions were prepared essentially as described previously [30] . Briefly , 293AG cells in five 10 cm-diameter dishes were homogenized with a Dounce homogenizer in 5 ml of homogenizing buffer ( 0 . 25 M sucrose and 20 mM Hepes-NaOH ) supplemented with Complete protease inhibitor cocktail ( Roche ) . The homogenate was centrifuged at 1 , 000×g for 10 min at 4°C . The supernatant was centrifuged at 2 , 000×g for 30 min at 4°C . The supernatant was then ultracentrifuged in a Beckman TLA-100 . 3 rotor at 85 , 000 rpm for 30 min at 4°C . The pellet ( microsomes ) and the supernatant ( cytosol ) fractions were mixed with an equal volume of 2 × SDS-PAGE sample buffer and analyzed by immunoblotting . To analyze the association of proteins with membranes , the pellet fraction ( microsomes ) were further incubated with 100 µl of 1 M KCl , 0 . 2 M sodium carbonate ( pH 11 ) , or 2 M Urea in the buffer for 1 h on ice and then recovered by ultracentrifuging in a Beckman TLA-100 . 3 rotor at 85 , 000 rpm for 30 min at 4°C . The supernatant and pellet fractions were then analyzed by immunoblotting . Triton X-114 phase separation was performed as described previously [30] . Briefly , the microsomes thus prepared were incubated in the Tx114-lysis buffer [1% Triton X-114 , 10 mM Tris-HCl ( pH 7 . 4 ) , 150 mM NaCl , and 1 mM EDTA] with Complete protease inhibitor cocktail for 60 min at 4°C . The lysate was centrifuged at 10 , 000×g for 15 min at 4°C . The supernatant was incubated at 37°C for 3 min and then centrifuged at 10 , 000×g for 1 min at 4°C . The upper ( aqueous ) phase , the lower ( detergent ) phase and the pellet were analyzed by immunoblotting . In order to analyze the membranous fraction from JCI cells , cells in a 10 cm-diameter dish were suspended in 1 ml of the homogenization buffer A [0 . 25 M sucrose and 60 mM HEPES-NaOH ( pH 7 . 4 ) ] containing protease inhibitor ( Complete Mini , EDTA-free , Roche Diagnostics ) . The cells were homogenized using a tight-fitting Dounce homogenizer ( 5 strokes ) , and the homogenates were centrifuged at 400×g for 3 min to remove the nuclei and cell debris . The supernatant was treated with a solution containing 20 µg/ml DNase I , 7 . 5 mM manganese chloride , and 0 . 5 mM DTT for 30 min at room temperature , and centrifuged at 13 , 000×g for 15 min at 4°C . The pellets were resuspended with 2 ml of the homogenization buffer B [0 . 25 M sucrose , 1 mM EDTA , and 10 mM HEPES-NaOH ( pH 7 . 4 ) ] , and immediately used for fractionation . Iodixanol ( OptiPrep , Nycomed ) density gradient analysis was performed according to the manufacturer's instructions . A working solution of 50% iodixanol [5∶1 v/v mixture of OptiPrep and 0 . 25 M sucrose , 6 mM EDTA , 60 mM HEPES–NaOH ( pH 7 . 4 ) ] was diluted with the homogenization buffer to prepare a 25% solution . Iodixanol continuous gradients were formed with 5 . 5 ml each of 0% ( the homogenization buffer containing the protease inhibitor ) and 25% iodixanol solutions in 13-ml open-top centrifuge tubes . Two ml of the cell lysates was overlaid on the top of the gradients , and then ultracentrifuged in a Hitachi P40ST rotor at 40 , 000 rpm for 2 . 5 h at 4°C . The gradients were fractionated into 24 fractions of each 500 µl from the top , and analyzed by immunoblotting . Quantification of the intensities of obtained bands by immunoblotting was performed using the Image Gauge V3 . 2 software ( Fuji Film , Tokyo , Japan ) . 293T cells were detached using an enzyme-free/PBS-based cell dissociation buffer ( Gibco/BRL ) according to the manufacturer's instruction . Aliquots of 106 cells were washed in PBS/2% FBS and suspended in 100 µl of PBS/2%FBS . Cells were then incubated with 200 ng of primary antibody or control IgG ( BD Pharmingen ) as a negative control for 30 min at 4°C . After washing , bound antibodies were visualized by addition of phycoerythrin ( PE ) -conjugated anti-mouse or -rabbit Ig antibody ( Beckman Coulter ) . After washing , cells were suspended in 250 µl of PBS/2% FBS . Cell surface fluorescence was analyzed with a Becton Dickinson FACScalibur ( BD Bioscience , San Jose , California ) . Cell transfection was performed with Lipofectamine 2000 ( Invitrogen ) for 293T cells or Fugene HD for HeLa and SVG-A cells . For immunoblot analysis , cells were harvested at the indicated time points after transfection , lysed in TNE buffer [10 mM Tris-HCl ( pH 7 . 5 ) , 150 mM NaCl , 5 mM EDTA , 10% glycerol , 1% Triton X-100 , and 0 . 5 mM phenylmethylsulfonyl fluoride ( PMSF ) ] , and mixed with Complete protease inhibitor cocktail . The cell lysates were fractionated by SDS-PAGE , and the separated proteins were transferred to a polyvinylidene difluoride filter ( Millipore ) . The filter was incubated with primary antibodies , and immune complexes were then detected with horseradish peroxidase–conjugated secondary antibodies and ECL reagents ( GE Healthcare ) . The Flag epitope was detected directly with horseradish peroxidase–conjugated primary antibodies . For detection of the homo-interaction of agnoproteins , 293T cells transfected with Myc-tagged agnoprotein/Flag-tagged agnoprotein or Myc-tagged agnoprotein alone were incubated for 72 h and then lysed in TNE buffer and subjected to immunoprecipitation . Immunoprecipitation was performed by incubation of cell lysates at 4°C first for 4 h with antibody-coupled protein G–Sepharose FF beads ( GE Healthcare ) . After washing with cell lysis buffer , the bead-bound proteins were subjected to immunoblot analysis . Subcellular fractionation was performed by a procedure that allows separation of nuclei and ER from other membrane fractions . Cells were suspended in ice-cold sucrose buffer I [0 . 32 M sucrose , 3 mM CaCl2 , 2 mM magnesium acetate , 0 . 1 mM EDTA , 10 mM Tris-HCl ( pH 8 . 0 ) , 1 mM dithiothreitol ( DTT ) , and 0 . 5% Nonidet P-40] and then mixed with sucrose buffer II [1 . 8 M sucrose , 5 mM magnesium acetate , 0 . 1 mM EDTA , 10 mM Tris-HCl ( pH 8 . 0 ) , and 1 mM DTT] . The resulting mixture was gently overlaid on 4 . 4 ml of sucrose buffer II in a 13-ml open-top centrifuge tube to form a discontinuous sucrose gradient . The gradient was centrifuged at 30 , 000×g for 45 min at 4°C , and the resulting pellet was washed with PBS containing 0 . 5% Triton X-100 . The pellet was then lysed in nuclear lysis buffer [25 mM Tris-HCl ( pH 7 . 4 ) , 300 mM NaCl , 0 . 5% Nonidet P-40 , and 0 . 5% sodium deoxycholate] and rotated for 1 h at 4°C . The lysate was centrifuged at 20 , 000×g for 30 min at 4°C , and the resulting supernatant was subjected to immunoblot analysis . Crosslinking of agnoprotein in SVG-A cells infected with JCV was performed as described previously with minor modifications [50] . SVG-A cells infected with JCV for 1 week were detached using an enzyme-free/PBS-based cell dissociation buffer ( Gibco/BRL ) according to the manufacturer's instruction . Aliquots of 106 cells were washed in DPBS ( + ) and suspended in 100 µl of PBS/2% FBS . The cells were suspend in 100 µl of DPBS ( + ) and then reacted with 0 . 5 , 1 , 2 or 5 mM disuccinimidyl suberate ( DSS , Pierce ) or DMSO for 2 h at 4°C . The cells were quenched with 50 mM Tris ( pH 7 . 5 ) for 30 min at 4°C and then lysed in 1% Triton X-100 in PBS at 4°C and centrifuged at 15 , 000×g for 15 min at 4°C . Supernatants were subjected to SDS-PAGE and immunoblot analysis using anti-agnoprotein antibody . The method of analyzing the oligomerization of agnoprotein was adapted from a previously published method [51] . SVG-A cells plated on collagen-coated 35-mm diameter glass base dishes ( Asahi Techno Glass , Tokyo , Japan ) were transfected with pCXSN-YFP-Agno/pCXSN-CFP-Agno or pCXSN-YFP/pCXSN-CFP . The cells were then placed in a chamber box on a microscope , in which the temperature was maintained at 37°C , and were imaged with an IX71 inverted microscope ( Olympus ) , as described previously [52] . The analysis of the cell image data , FRET signal correction and normalization was conducted as reported previously [51] . The lipid packing of plasma membrane can be studied by inserting the fluorescence probe , MC540 ( Sigma ) into cell membranes and assessing the degree of insertion by fluorescence intensity measurement using flow cytometry [34] [35] . HeLa cells were transfected with pCXSN-Agno or pCXSN vector , and were incubated for 72 h . The cells were then detached by using an enzyme-free/PBS-based cell dissociation buffer . The cells were suspended in 10 µg/ml MC540/DPBS ( + ) . After 10 min incubation at 37°C , cells were washed in DPBS ( + ) . After washing , cells were suspended in 250 µl of DPBS ( + ) . Cell surface fluorescence was analyzed with a Becton Dickinson FACScalibur . FACScalibur was used at 488 nm excitation and 575 nm emission wavelengths . Permeability of the plasma membranes of agnoprotein-transfected cells to HygB ( Clontech ) was determined as described previously [37] . Briefly , HeLa cells or SVG-A cells were plated on a 35-mm dishes and transfected with pCFPNLS-Agno , pCFPNLS-N46 , pCFPNLS-RK8AA , or pCFPNLS . At 72 h post-transfection , the cells were pretreated with HygB ( 400 µg/ml ) for 15 min , and then 50 µCi of [35S] Met-Cys were added to the culture medium . The cells were then incubated at 37°C for 2 h in the presence or absence of HygB . The cell extracts were lysed in 500 µl of RIPA buffer mixed with protease inhibitor and subjected to immunoprecipitation with goat anti-GFP antibody-coupled protein G-Sepharose FF beads . The bead-bound proteins were analyzed by SDS-PAGE and autoradiography . The method of analyzing the rate of Ca2+ influx was adapted from a previously published method [38] . HeLa cells plated on a collagen-coated 35-mm diameter glass base dish were transfected with pERedNLS-Agno/pERedNLS and a FRET-based fluorescence indicator for Ca2+ , YC3 . 60 ( kindly provided by Dr . A . Miyawaki ) . After 96 h , cells were washed once in the recording medium [10 mM Hepes-NaOH ( pH 7 . 4 ) , 140 mM NaCl , 5 mM KCl , 1 mM MgCl2 , and 0 . 55 mM Glucose] . The cells were incubated with the calcium chelater dimethyl-BAPTA-AM ( Molecular Probes ) for 30 min at 37°C to prolong the linear phase of unidirectional Ca2+ uptake . The cells were then placed in a chamber box on a microscope , in which the temperature was maintained at 37°C , and were imaged with an IX71 inverted microscope , as described previously [52] . Fluorescent images of CFP and FRET were recorded every 10 sec . At 100 sec , 5 mM CaCl2 was added to the cells . MetaMorph software ( Universal Image ) was used for control of the CCD camera and the filter wheels , and also for the analysis of the cell image data . Assays for E . coli membrane permeability and viability have been previously described [41] . The entire coding sequences of VP1 ( pET15b-VP1 ) [53] and agnoprotein ( pET15b-His-Agno ) were amplified by PCR and cloned into the pET15b expression vector ( Novagen , Madison , WI ) . The integrity of vectors were verified by sequencing and transformed into the E . coli BL21 strain ( DE3: pLysS ) for protein expression ( Novagen ) . For permeability assays , cultures were grown in the presence of 100 µg/ml ampicillin at 37°C to an optical density ( OD ) at 590 nm of ∼1 . 0 , and protein expression was induced with 1 mM isopropyl-β-D-thiogalactopyranoside ( IPTG ) for 10 min prior to the addition of rifampicin ( 150 µg/ml ) to inhibit transcription of endogenous genes by the E . coli RNA polymerase . Samples were incubated for an additional 30 min followed by a 10-min pulse with 50 µCi [35S] Met-Cys in the absence or presence of 250 µg/ml hygromycin B . Bacterial cell lysis was monitored by following the OD at 590 nm after protein expression in cultures at an OD at 590 nm of ∼0 . 20 was induced by the addition of 1 mM IPTG . Electron microscopy of JCV-infected cells was performed as follows: Cells were fixed in 2 . 5% glutaraldehyde ( TAAB , Aldermaston , UK ) in 0 . 1 M phosphate buffer ( PB , pH 7 . 4 ) for 48 h at 4°C . Thereafter , cells were washed in the same buffer containing 7% sucrose , post-fixed in 1% osmium tetroxide ( Merck , Darmstadt , Germany ) in 0 . 1 M PB for 1 . 5 h at room temperature , dehydrated in graded acetones , and embedded in Epon ( TAAB ) . Ultrathin sections ( 0 . 1 µm in thickness ) were stained with 1 . 5% uranyl acetate for 20 min and 0 . 2% lead citrate for 15 min , and were examined under an H-7100 electron microscope ( Hitachi , Tokyo , Japan ) . All data were expressed as mean ± S . D . Student's t-test was used to analyze differences between two groups . A value of p<0 . 05 was considered as statistically significant . | Most non-enveloped viruses exit their host cells following cell lysis , which involves breakdown of the cell membrane and death of the host cell , and which is presumably the final result of an increase in plasma membrane permeability . JC virus ( JCV ) is the causative agent of progressive multifocal leukoencephalopathy ( PML ) and belongs to the polyomavirus family , which have non-enveloped virions . The extracellular release of mature progeny polyomavirus virions has been suggested to occur when cells disintegrate or rupture; however , the molecular mechanism ( s ) employed by JCV to induce cell lysis and facilitate virion release remain elusive . Viroporins are a group of proteins that modify the permeability of cellular membranes and promote the release of viral particles from infected cells . These proteins are not essential for the replication of viruses , but their presence often enhances virus growth . Here , we demonstrate that the JCV agnoprotein forms homo-oligomers as an integral membrane protein and acts as a viroporin , and that expression of agnoprotein results in plasma membrane permeabilization and virion release . These observations suggest that the process of virion release of this non-enveloped DNA virus is highly regulated by a single viral protein . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"infectious",
"diseases/infectious",
"diseases",
"of",
"the",
"nervous",
"system",
"infectious",
"diseases",
"infectious",
"diseases/viral",
"infections"
] | 2010 | The Human Polyoma JC Virus Agnoprotein Acts as a Viroporin |
Leishmaniasis is a tropical disease threatening 350 million people from endemic regions . The available drugs for treatment are inadequate , with limitations such as serious side effects , parasite resistance or high cost . Driven by this need for new drugs , we developed a high-content , high-throughput image-based screening assay targeting the intracellular amastigote stage of different species of Leishmania in infected human macrophages . The in vitro infection protocol was adapted to a 384-well-plate format , enabling acquisition of a large amount of readouts by automated confocal microscopy . The reading method was based on DNA staining and required the development of a customized algorithm to analyze the images , which enabled the use of non-modified parasites . The automated analysis generated parameters used to quantify compound activity , including infection ratio as well as the number of intracellular amastigote parasites and yielded cytotoxicity information based on the number of host cells . Comparison of this assay with one that used the promastigote form to screen 26 , 500 compounds showed that 50% of the hits selected against the intracellular amastigote were not selected in the promastigote screening . These data corroborate the idea that the intracellular amastigote form of the parasite is the most appropriate to be used in primary screening assay for Leishmania .
Leishmaniasis is a tropical disease caused by parasites of the genus Leishmania , with clinical manifestations ranging from localized cutaneous ulcers to systemic visceral organ damage . The visceral form of the disease is the most severe and is lethal if not treated . The organs targeted by the parasites are determined mainly by the infecting parasite species and the patient's immune system . In the human host , the parasite is able to infect different cell types , with macrophages as the final host in which the parasites differentiate from promastigotes into amastigotes and multiply [1] . Leishmaniasis is endemic to 88 countries in tropical and sub-tropical areas and threatens 350 million people [2] . There is no available vaccine [3] , and the drugs used for treatment have major drawbacks , including parasite resistance and high toxicity , with strong side effects for the patient [4] , [5] . New drugs or formulations are therefore urgently needed [6] , [7] . High-throughput screening ( HTS ) is an efficient way of identifying active compounds among large numbers of small molecules , thereby feeding drug discovery pipelines with new candidates and optimizing both research costs and time [8] . High-content screening ( HCS ) combines the efficiency of HTS with information-rich assays to provide several measures of a compound's effect in the assay system . The requirements for HCS assays include quantifiable and reproducible measurements of compound activity compared to standard reference drugs in concentrations achievable in serum/tissues , handling of ultra-small ( nanogram ) amounts of compound , and adaptation of the assay to standard microplate formats and laboratory automation platforms . Here we report the development and validation of a protocol for in vitro drug screening and automated image mining for leishmaniasis using the intracellular amastigote form of Leishmania to infect human macrophages . We applied an image-based approach and developed computer-assisted algorithms to interpret the infection and quantify the activities of the anti-parasitic compounds . Other assays in medium- to high-throughput format have already been developed for anti-leishmanial drug screening using promastigotes ( insect forms ) or axenic amastigotes [9] , [10] , [11] , [12] . While these forms of the parasite are easier to adapt to an HTS assay format , promastigotes are not representative of the human disease , and the use of axenic amastigotes as a model of intracellular amastigotes is controversial because amastigotes are exclusively intracellular in vivo [13] . To confirm the importance of the parasite stage used in an HTS assay with a large number of compounds , we compared screening data for a subset of 26 , 500 compounds obtained using either this newly developed assay or a previously developed HTS assay that used the promastigote form of the parasite .
Leishmania donovani MHOM/ET/67/HU3 , Leishmania amazonensis MHOM/BR/73/M2269 , Leishmania braziliensis MHOM/BR/2903 and Leishmania major MHOM/IL/81/FRIEDLIN promastigotes were axenically cultivated at 28°C in 199 Culture Medium ( Sigma M5017 ) with 40 mM Hepes ( Gibco 15630 ) , 0 . 1 mM adenine ( Sigma A5251 ) , 0 . 0001% biotin ( Sigma B329 ) and 4 . 62 mM NaHCO3 ( Sigma S5761 ) , supplemented with 10% ( or 20% for L . braziliensis ) heat-inactivated fetal bovine serum ( FBS , Gibco 16000 ) and 1% streptomycin/penicillin ( Gibco 15140 ) . The cultures were diluted every 3 or 4 days to maintain the parasite density between 106 parasites/ml and 4×107 parasites/ml . To avoid generation of genetic variability , we kept the parasite for a maximum of 20 sub-cultured dilution cycles , thawing new vials from the same frozen stock . THP-1 , the human acute leukemia monocyte cell line ( ATCC TIB-202 ) , was cultivated in RPMI medium ( Gibco 61870-036 ) supplemented with 10% heat-inactivated FBS ( Gibco 16000 ) and 1% streptomycin/penicillin ( Gibco 15140 ) at 37°C and 5% CO2 . The cultures were diluted every 3 or 4 days to maintain the cell density between 105 cells/ml and 8×105 cells/ml . Cells were kept for a maximum of 20 sub-cultured dilution cycles , thawing new vials from the same frozen stock . The Leishmania culture at a density of 106 parasites/ml was incubated for 6 days before infection to enrich the proportion of metacyclic promastigotes . THP-1 cells at 5×105 cells/ml were differentiated with 50 ng/ml of phorbol 12-myristate 13-acetate ( PMA , Sigma P1585 ) for 48 hours at 37°C , 5% CO2 . Differentiated THP-1 cells are adherent and were seeded at a confluence of 1 . 8×105 cells/cm2 . Trypsinized THP-1 cells were mixed with the 6-day-old Leishmania promastigotes at a final density of 4×105 THP-1/ml and 2×107 parasites/ml in RPMI medium supplemented with 10% FBS . This homogeneous mixture of differentiated THP-1 cells and parasites was seeded in 384-well plates at 50 µl/well using a WellMate ( Thermo Scientific ) liquid handler and incubated for 5 days at 37°C , 5% CO2 . To visualize amastigote replication , differentiated THP-1 cells were seeded on coverslip slides in 24-well plates ( 8×104 cells/well ) and infected with 4×106 parasites in a final volume of 200 µl RPMI supplemented with 10% FBS . The cells were washed 3 times with PBS 12 hours after parasite addition . To detect DNA replication , 5-bromo-2-deoxyuridine ( BrdU ) was used . A mixture of 1 mM BrdU and 1 mM deoxycytidine ( dC ) was added to the infected THP-1 culture and incubated for 12 hours . The cultures were then washed with PBS , fixed with cold 100% methanol for 5 minutes and air dried for 5 minutes at room temperature . Next , 1 . 5 M HCl prepared in PBS was added to the samples and incubated for 15 minutes , followed by PBS washing and permeabilization with 0 . 1% Triton X-100 in PBS for 10 minutes . After another PBS wash , the cells were incubated for 1 hour at 4°C with mouse anti-BrdU conjugated monoclonal antibody ( Invitrogen , 1∶400 dilution ) in PBS containing 4% BSA , washed 3 times with PBS and further incubated for 1 hour at 4°C with Alexa Fluor® 488 goat anti-mouse IgG ( Invitrogen , 1∶400 dilution ) , 2 µM DAPI ( Sigma D9564 ) and 2 . 5 µM Syto60 ( Invitrogen MOP-S-11342 ) in PBS containing 4% BSA . The samples were washed 3 times with PBS and mounted over slides with Vecta-shield ( Vector H-1000 ) . Staining was visualized with a Leica confocal TCS SP2 fluorescence microscope , taking pictures in different focal planes ( z-stack ) . Imaris ( Bitplane ) software was used to develop the 3D model after acquisition of a series of images in different focal planes . The anti-leishmanial reference drugs used were amphotericin B ( Sigma A9528 ) , miltefosine ( Merck 475841 ) , paromomycin sulfate ( Sigma P9297 ) and sodium stibogluconate ( Sigma S5319 ) . The reference drugs and tested compounds were added 24 hours after infection and incubated at 37°C and 5% CO2 for 4 days . The cells and parasites were then fixed with 2% paraformaldehyde and stained with 5 µM Draq5 ( Biostatus DR50200 ) in PBS . Fluorescent images ( 4 images per well ) were acquired from each assay well using an Opera confocal microscope ( Perkin Elmer ) with a 635 nm laser at 20× lens magnification . The acquired images were analyzed with an Image Mining ( IM ) platform developed in-house . The IM software directly accessed the databases of image acquisition platforms and created a flow of images , which were sequentially analyzed by dedicated algorithms developed as plugins of the software . The results of all analyses were stored in a centralized database . Using a simple DNA staining technique to permit the use of wild-type parasites without any reporter gene was interesting because its quantification relied solely on the accuracy of the image analysis , and it permitted various parasite species to be used without the need for genetic manipulation . The DNA staining proved to be a simple and stable cell marker , and the image analysis software plugin that was developed took into account both the accuracy and speed constraints of HCS . Several methods were tested , and some steps were necessary to detect cells and parasites to obtain a robust separation between the positive and negative controls . Fig . 1 presents a flow chart of those steps . The quantification of the infection ratio consisted of cell segmentation based on nuclei and parasite detection components . The two components were processed independently from the same raw input image ( Fig . 1A ) and merged to identify individual cells and assign detected parasites to each cell object ( Fig . 1J ) . The cell segmentation was as follows ( Fig . 1B–E ) : the raw input image was smoothed with a Gaussian kernel [14] of standard deviation 5 , which roughly corresponds to the radius of the nucleus ( Fig . 1B ) . Then , the local extreme detection method [15] was applied to the smoothed image to extract local maxima points , which indicated the number and positions of nuclei ( Fig . 1C ) . Subsequently , the inner boundaries of the individual cells were identified by computing the Voronoi diagram [16] on the foreground area using the previously obtained set of local maxima as seed points ( Fig . 1D ) . Finally , a foreground mask ( Fig . 1E ) was applied to the Voronoi diagram image to remove background area . For the parasite detection ( Fig . 1F–I ) , the algorithm calculated the higher 50% cumulative intensity level of the raw input image ( Fig . 1F ) , used a threshold cut-off in the image based on the intensity level to remove background area and the cytoplasm of cells to isolate nuclei and parasites ( Fig . 1G ) and , finally , used the connected component labeling method [17] to filter out nuclei objects and other artifacts by detecting as parasites the connected component objects that were simultaneously larger than 4 pixels and smaller than 15 pixels ( Fig . 1I ) . The final result from a processed image is shown ( Fig . 1J ) , with cell detection and segmentation merged with parasite detection . The IM software interface is shown in Fig . 2 . A color-based graphical representation of the 384-well plate enabled a quick visual analysis of the results . Fig . 2A illustrates one DMSO control plate , highlighting the infection ratio as the scaling factor for the blue color . Algorithm parameters may be modified to optimize the analysis of the images as shown in the window illustrated in Fig . 2B . The software provides 7 output features that can be used for statistical analysis ( Fig . 2C ) . These features include the following: 1 ) number of cells; 2 ) number of infected cells ( cells containing at least one parasite ) ; 3 ) infection ratio ( the number of infected cells divided by the number of cells ) ; 4 ) total number of parasites per cell; 5 ) standard deviation of the number of parasites per cell; 6 ) total area of the cells ( number of pixels occupied by cells ) , and 7 ) average area per cell ( average number of pixels per cell ) . Fig . 2D–E shows images from the wells before ( left ) and after ( right ) IM analysis , highlighting the identification of THP-1 cells and Leishmania parasites . The calculated activity was normalized to percentage infection ( Inf ) based on the amphotericin B EC100 ( effective concentration showing 100% activity , meaning minimum measured infection ratio ) and 1% DMSO ( 0% activity , meaning maximum measured infection ratio ) controls according to the formula: % infection = ( measured IR−μAmpB EC100 ) / ( μ1%DMSO−μAmpB EC100 ) ×100 , in which μ1%DMSO is the average infection ratio of the 1% DMSO controls and μAmpB EC100 is the average infection ratio of the amphotericin B EC100 control . The quality of the assay data was primarily assessed with the Z' factor [18] . In addition to the Z' factor , a number of other parameters were taken into consideration to assess the robustness of the developed assay . These include the following: i ) pharmacological validation as assessed by dose response curves for the reference drug amphotericin B . In the current assay , the EC50 for amphotericin B was determined to be approximately 0 . 3 µM , in agreement with previously published values [19]; ii ) evaluation of any type of plate pattern such as edge effects , particularly important for assays with multi-day incubations , or other patterns introduced , for example , by liquid handling devices; iii ) quantitative assessment of variability in the infection ratio , cell and parasite number on different days and with different batches of cells and parasites .
The screening assay was based on the use of late-stage promastigote Leishmania cultures to infect differentiated THP-1 macrophages and the quantification of the infection ratio 4 days after compound addition . THP-1 human macrophages have been used as a Leishmania host model for in vitro infection for over two decades [20] , [21] , [22] and have been proposed to be suitable for drug screening against the intracellular form of the parasite [12] . We investigated different time points in the growth curve of the THP-1 cells and Leishmania parasites to determine the optimal development stage for each cell line , in order to optimize the infection ratio . Fig . 3A illustrates the growth curves and highlights the optimal culture durations for the host cells and parasites prior to infection . The use of 5- or 6-day-old late-log-phase promastigote cultures resulted in higher infection ratios because of the enrichment of metacyclic promastigotes in the culture media , as previously reported [23] . A high infection ratio obtained with this protocol is illustrated in Fig . 3B , with an average of 88 . 7% ( ±4 . 7% ) . Reference drugs and assay compounds were added to the plate 24 hours after infection . The protocol was intended to select compounds active against intracellular amastigotes . To demonstrate that within this time interval the late stage promastigotes had been phagocytized by macrophages and had differentiated into amastigotes within the macrophages , we visualized the intracellular replication of the parasite using a BrdU incorporation assay . The presence of incorporated BrdU , as detected by immunofluorescence , demonstrated that 24 hours after infection , the intracellular parasites were indeed differentiated into amastigotes and were replicating inside the host cell . This result validated the protocol of adding compounds 24 hours after infection to target intracellular amastigotes . The same replication assay was also performed 4 days after infection . Pictures were taken in different focal planes , and a 3D model was built , as shown in Movie S1 . Fig . 3C shows snapshots of the 3D model . Parasites labeled in green indicate replicating parasites . When high throughput is considered for biological experiments , the assay must be adapted to best fit the requirements of the large scale . However , to compare results from different laboratories , we developed a Leishmania infection standard protocol adapted to an HCS/HTS method , with the potential for implementation on a smaller laboratory scale without the need for automation . To define this standard protocol , several conditions were tested to find the highest and most reproducible infection ratio , using L . donovani parasites to define and validate the assay . The parasite containers were one of the tested variables , comparing a T175 flask with a 1 . 0 L Erlenmeyer bottle under agitation , being the latter the best option ( Figure S1 ) . We also tested the effect of PMA concentration and incubation time on THP-1 cell differentiation , as illustrated in Figure S2 , showing that 50 ng/ml yielded best results . The method for releasing differentiated THP-1 cells was also important for the preservation and integrity of the cells , and different approaches were tested , being trypsinization the chosen method ( Figure S3 ) . To evaluate the pharmacological relevance of the assay , we tested four different anti-leishmanial reference drugs in our system: amphotericin B , miltefosine , paromomycin and sodium stibogluconate . Dose-response curves ( DRC ) were determined in three independent experiments , with four replicates used to define the EC50s . The results for each reference drug are given in Table 1 . The EC50s of all of the reference drugs were comparable to the reported values [19] , [24] , thus validating this new assay system for compound screening . Miltefosine is a known anti-cancer drug [25] , and , at the EC100 dose , cytotoxicity was observed for the THP-1 host cell , an acute monocytic leukemia cell line . Sodium stibogluconate did not eliminate intracellular parasites in the experiment; this compound is known to be poorly active in vitro , and its slow effect was incompatible with the exposure time window for our experiment [23] , [26] . Paromomycin was not efficient either in eliminating intracellular L . donovani from the THP-1 host; its activity was inconsistent , and it was inactive in some experiments , as previously reported [19] . This variability is not acceptable for HTS purposes . We chose amphotericin B as a reference drug in the screening assay because it had an EC50 in the target range for new active compounds ( nanomolar ) , was stable , was not cytotoxic up to 20 µM and showed reproducible results . In addition , amphotericin B is a first-line drug used for the treatment of visceral leishmaniasis in many endemic countries [4] . Because the assay was conducted with live parasites and host cells over multiple days , a fully automated ( i . e . , unattended ) protocol was not feasible . However , several time-consuming and repetitive steps in the assay were automated , enabling a potential throughput of ∼20 , 000 wells per screening day . An infection batch of host cells and parasites was added to the plates with a bulk dispenser ( Wellmate , Thermo Scientific ) . The plates were then loaded onto the automated platform for compound addition , incubation and reading ( all automated steps ) . The first step in automation is the scale-up of the assay . In this case , cells and reagents were prepared at the same scale as the one to be used during an HTS campaign and compared to the results for a small scale preparation ( Figure S4 ) . Each of the liquid handling devices to be used in the automation process was individually validated for accuracy and reproducibility . The assay validation consisted of the simulation of 3 independent screening days using only control plates . Each run was composed of 20 plates , 10 DRC plates containing amphotericin B as the reference drug and 10 DMSO plates as the negative control . The validation was performed on a cell::explorer™ automated platform with an Opera confocal microscope ( Perkin Elmer ) for imaging . The validation run using the semi-automated screening method and the results indicated that the assay was robust , both within a screening day and across screening days . Within a screening day , the number of THP-1 cells in both the DMSO controls and the amphotericin B ( EC100 ) wells was similar , indicating that the cell number was consistent between the control and DMSO wells , within a plate and across multiple plates ( Fig . 4A ) . Based on the infection ratio , there was a clear window between the DMSO controls and amphotericin B ( Fig . 4B ) . This window was consistent throughout the validation day and across multiple validation days ( Fig . 4C ) . The assay also showed a consistent EC50 in the range of 0 . 3 µM for the reference compound , amphotericin B , between screening days ( Fig . 4D ) . The Z' factor within a validation day and across multiple days was 0 . 5 , indicating that the assay was high quality . A flowchart of the entire screening methodology is illustrated in Figure S5 . To determine the specificity of the compounds , we used the same assay principle with species of Leishmania other than L . donovani , which cause different clinical manifestations . We used L . major , L . amazonensis and L . braziliensis as representatives of cutaneous and mucocutaneous leishmaniasis . Images illustrating the infection of THP-1 cells with these 3 species in addition to L . donovani are depicted in Fig . 5A . The DRCs for amphotericin B demonstrated that all of the species had similar in vitro sensitivities , with the following EC50s: 0 . 30 µM for L . amazonensis , 0 . 28 µM for L . braziliensis , 0 . 28 µM for L . donovani and 0 . 31 µM for L . major ( Fig . 5B ) . We previously published an HTS screen with Leishmania promastigote forms [12] , and here we evaluated the correlation between the results from that promastigote-based screen and the amastigote-based screen ( this work ) . We used a library of 26 , 500 compounds and evaluated the outputs from each approach . The screen using L . major promastigotes with compounds at 10 µM generated 124 hits after a 70% activity cut-off ( 3 standard deviations from the negative control average ) and the exclusion of compounds that interfered with the growth of non-differentiated THP-1 ( potentially toxic ) , with only 5 compounds ( 4% ) showing activity against intracellular L . major amastigotes at concentrations up to 20 µM [12] . An independent screen was performed using the same compound library ( 26 , 500 ) at 20 µM against L . donovani using the new infection assay described in this study . The number of hits based on a 55% activity cut-off ( 3 standard deviations from the negative control average ) , no cytotoxicity against differentiated THP-1 ( based on cell counting ) and an EC50<20 µM was 123 compounds ( Figure S6 ) , coincidently almost the same number of hits selected in the promastigote screening . From the 123 hit compounds , 62 showed activity against promastigotes . We also confirmed these data , in which only 51% of the hits obtained from the intracellular amastigotes ( 24 compounds out of 47 hits ) were also active against the promastigote form , by screening a focused library of 4 , 000 kinase and phosphatase inhibitors using L . donovani ( data not shown ) . The results from the 26 , 500 compounds are illustrated in Figure S6 and provide evidence that performing screens with intracellular amastigote forms would increase the probability of finding compounds active against the human form of the parasite ( amastigote ) and would be the most appropriate way to find compounds that exclusively target the intracellular form that infects the host cell .
HTS technology has been primarily associated with target-based assays . This approach requires substantial efforts prior to screening to identify and validate a target involved in the disease process . Once a target is identified and validated , compounds can be screened against it to identify inhibitors in either biochemical ( e . g . , purified enzyme ) assays or cell-based ( e . g . , receptor internalization ) assays . For non-infectious diseases , for which protein targets are known , a target-based screening approach is very relevant . In contrast , targeting the causative infectious agent as a whole is an interesting approach for parasitic or other infectious diseases . Indeed , if one considers the field of antibacterials , target-based screens have been fairly unsuccessful in recent years [27] . We describe here the development of a new screening assay to identify new compounds active against Leishmania . This assay selects for a specific phenotype ( absence or reduction of parasites in macrophage host cells ) , and this parameter is used to measure compound activity; all potential targets will thus be exposed to the tested compounds , thereby increasing the probability of finding active compounds with different modes of action in an optimized fashion . Reporter genes have been successfully developed for anti-leishmanial HTS by various research groups [28] , [29] . However , a major drawback of current reporter assay approaches was the need for continuous drug selection to maintain expression of the reporter gene over time . For Leishmania , this is particularly true for both episomal transient transfection and integrated transgenes , because the parasite has high genomic plasticity and a high recombination ratio [30] . Furthermore , the use of selection drugs can be problematic because they may directly or indirectly interact with the tested compound , interfering with the screening results . Recently , two independent research groups presented stable modified Leishmania parasites expressing GFP that are suitable for compound testing [31] , [32] . Although engineered parasites are an invaluable tool in drug screenings , the use of wild-type organisms should be prioritized whenever possible because of potential for modifications in the general metabolism of the organism , due to expression of the transgene , resulting , for example , in a loss of virulence , as reported by de Toledo and collaborators [33] and as observed during this study with GFP-expressing L . donovani ( data not shown ) . The use of wild-type parasites allows for the anti-leishmanial activity of compounds to be evaluated against different species of parasites , including clinical isolates , without the need for any genetic modification of these parasites . This approach required the development of a new tool for the image analysis of the parasite-infected cells: because DNA staining was the chosen method , the image analysis algorithm used had to identify each host cell as an isolated object , while also detecting intracellular parasites , all from the same image channel . Cell segmentation was the first challenge . By choosing the Voronoi diagram method , we were able to identify the inner boundaries of individual cells . The foreground mask applied to the Voronoi diagram image was then used to remove the background area , excluding all extracellular parasites from the analysis . The foreground mask was defined by pixels with intensity greater than 50 . We observed that , given the stability of the image statistics across the screen and the good contrast between foreground and background , a fixed value was a robust way of separating foreground and background across the entire set of images . We tested several precise segmentation methods , such as region growing or a watershed algorithm , to measure the influence of the methods on the analysis results , and concluded that there was no significant difference in the infection ratio when precise segmentation methods or Voronoi diagrams were used . However , computational cost was much lower when we used the Voronoi diagram . In addition , Leishmania parasites could be detected inside the boundaries of each cell in the same image . Image-based processing is a promising approach for chemical screening [34] . In our study , using a single channel for fluorescence labeling of both the host cells and parasites considerably simplified the technical complexity of the assay , but created a computational challenge for selectively classifying cells and parasites from the same fluorescence signal channel . An algorithm was developed and implemented to automatically process the readout images , allowing the analysis of the huge amount of data generated from the screens ( Figs . 1 and 2 ) . The development of an accurate tool to quantify the infection ratio enabled the measurement and comparison of compound activity against the intracellular stage of the parasite . This result , in addition to validating the biological model by confirming that replicative intracellular L . donovani can infect a human macrophage cell line , establishes this system as the most relevant and promising for the discovery of new compounds for leishmaniasis treatment . In addition to the infection ratio and parasite number , another parameter widely used to assess compound cytotoxicity is the number of host cells . Potentially toxic compounds that affect replication or cell multiplication will not be detected as toxic based on host cell counting because THP-1 cells usually do not replicate after differentiation , resulting in a constant cell number . However , compounds that induce necrosis or apoptosis or interfere with the adherence of the macrophages to the surface will be interpreted as toxic because of the low host cell number . Ultimately , this newly developed standard protocol for the Leishmania infection model allowed the screening of 200 , 000 compounds at 20 µM for anti-parasitic activity in a high-throughput mode using optimal conditions , to be reported elsewhere . A subset of this library ( 26 , 500 compounds ) has also been screened at 10 µM by our group against the promastigote form of Leishmania parasites in a fluorimetric assay [12] . Approximately 50% of the hits found in the intracellular amastigote assay were not found in the hit list obtained against the promastigote form . Conversely , only 4% of the hits from the promastigote screen were active against the intracellular amastigote . Even though the screenings used different species of parasites ( L . major promastigotes and L . donovani intracellular amastigotes ) , we consider this difference relevant because we recently obtained similar data when screening a focused library of 4 , 000 compounds using only the L . donovani strain . These data are in accordance with recently published work from Muylder et al . , which shows that using the promastigote form in a primary screen leads to a great number of hits that are likely to be inactive on the amastigote , and misses active compounds that are only found when using the intracellular amastigote [35] . Taken together , these results illustrate that a substantial number of compounds may be specifically active against intracellular amastigotes and would only be selected as true positive hits if tested in this screening system , supporting the proposal that the intracellular amastigote model is the most appropriate for drug discovery in leishmaniasis . In conclusion , the development and validation of this HTS protocol for Leishmania infection of human macrophages without the need for a reporter gene is a major breakthrough in the field of leishmaniasis drug discovery . It fills a major gap and should allow the screening of diverse and focused compound library sets , opening up a new avenue for the identification of new compound series , which are critically needed to develop new drugs for the treatment of Leishmaniasis . The combination of a infection model with image-based analysis has proven to be a relevant method for screening compounds for their activity against the intracellular amastigote Leishmania . Moreover , this protocol is well established for HTS and may be used on a smaller scale , such as in 96-well plates or even in lower throughput , in any research laboratory to test single compounds and/or natural product extracts . | Leishmaniasis , one of the most neglected tropical diseases , affects over 2 million people each year . Visceral leishmaniasis ( VL ) , also known as Kala-azar , is caused by the protozoan parasites Leishmania donovani and Leishmania infantum and is fatal if left untreated . Because existing treatments are often ineffective due to parasite resistance and/or toxicity new drugs are urgently needed . Leishmaniasis is transmitted to humans by the bite of an infected sandfly . In the insect vector , parasites exist as flagellated forms—promastigotes , which infect macrophage cells of the human host , where they differentiate to round forms known as amastigotes . Amastigotes and promastigotes are substantially different from a molecular perspective . Drug discovery for leishmaniasis has traditionally been complicated by the unavailability of validated drug targets and of relevant drug assays . Whole cell-based assays have been widely used , as they bypass the need for a validated target . However , they use the insect form of the parasite; indeed , the human form , the intracellular amastigote , is difficult to obtain in the laboratory in quantities compatible with drug screening . We describe here the technical advances that made it possible to adapt the intracellular amastigote form of L . donovani to a drug assay compatible with high-throughput screening . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [
"parastic",
"protozoans",
"protozoology",
"biology",
"microbiology"
] | 2012 | An Image-Based High-Content Screening Assay for Compounds Targeting Intracellular Leishmania donovani Amastigotes in Human Macrophages |
During B cell development , the precursor B cell receptor ( pre-BCR ) checkpoint is thought to increase immunoglobulin κ light chain ( Igκ ) locus accessibility to the V ( D ) J recombinase . Accordingly , pre-B cells lacking the pre-BCR signaling molecules Btk or Slp65 showed reduced germline Vκ transcription . To investigate whether pre-BCR signaling modulates Vκ accessibility through enhancer-mediated Igκ locus topology , we performed chromosome conformation capture and sequencing analyses . These revealed that already in pro-B cells the κ enhancers robustly interact with the ∼3 . 2 Mb Vκ region and its flanking sequences . Analyses in wild-type , Btk , and Slp65 single- and double-deficient pre-B cells demonstrated that pre-BCR signaling reduces interactions of both enhancers with Igκ locus flanking sequences and increases interactions of the 3′κ enhancer with Vκ genes . Remarkably , pre-BCR signaling does not significantly affect interactions between the intronic enhancer and Vκ genes , which are already robust in pro-B cells . Both enhancers interact most frequently with highly used Vκ genes , which are often marked by transcription factor E2a . We conclude that the κ enhancers interact with the Vκ region already in pro-B cells and that pre-BCR signaling induces accessibility through a functional redistribution of long-range chromatin interactions within the Vκ region , whereby the two enhancers play distinct roles .
B lymphocyte development is characterized by stepwise recombination of immunoglobulin ( Ig ) , variable ( V ) , diversity ( D ) , and joining ( J ) genes , whereby in pro-B cells the Ig heavy ( H ) chain locus rearranges before the Igκ or Igλ light ( L ) chain loci [1] , [2] . Productive IgH chain rearrangement is monitored by deposition of the IgH μ chain protein on the cell surface , together with the preexisting surrogate light chain ( SLC ) proteins λ5 and VpreB , as the pre-B cell receptor ( pre-BCR ) complex [3] . Pre-BCR expression serves as a checkpoint that monitors for functional IgH chain rearrangement , triggers proliferative expansion , and induces developmental progression of large cycling into small resting Ig μ+ pre-B cells in which the recombination machinery is reactivated for rearrangement of the Igκ or Igλ L chain loci [3] , [4] . During the V ( D ) J recombination process , the spatial organization of large antigen receptor loci is actively remodeled [5] . Overall locus contraction is achieved through long-range chromatin interactions between proximal and distal regions within these loci . This process brings distal V genes in close proximity to ( D ) J regions , to which Rag ( recombination activating gene ) protein binding occurs [6] and the nearby regulatory elements that are required for topological organization and recombination [5] , [7] , [8] . The recombination-associated changes in locus topology thereby provide equal opportunities for individual V genes to be recombined to a ( D ) J segment . Accessibility and recombination of antigen receptor loci are controlled by many DNA-binding factors that interact with local cis-regulatory elements , such as promoters , enhancers , or silencers [7]–[9] . The long-range chromatin interactions involved in this process are thought to be crucial for the regulation of V ( D ) J recombination and orchestrate changes in subnuclear relocation , germline transcription , histone acetylation and/or methylation , DNA demethylation , and compaction of antigen receptor loci [5] , [10] . The mouse Igκ locus harbors 101 functional Vκ genes and four functional Jκ elements and is spread over >3 Mb of genomic DNA [11] . Mechanisms regulating the site-specific DNA recombination reactions that create a diverse Igκ repertoire are complex and involve local differences in the accessibility of the Vκ and Jκ genes to the recombinase proteins [12] . Developmental-stage-specific changes in gene accessibility are reflected by germline transcription , which precedes or accompanies gene recombination [13] . In the Igκ locus , germline transcription is initiated from promoters located upstream of Jκ ( referred to as κ0 transcripts ) and from Vκ promoters [14] . Deletion of the intronic enhancer ( iEκ ) , located between Jκ and Cκ , or the downstream 3′κ enhancer ( 3′Eκ ) , both containing binding sites for the E2a and Irf4/Irf8 transcription factors ( TFs ) , diminishes Igκ locus germline transcription and recombination [15]–[19] . On the other hand , the Sis ( silencer in intervening sequence ) element in the Vκ–Jκ region negatively regulates Igκ rearrangement [20] . This Sis element was shown to target Igκ alleles to centromeric heterochromatin and to associate with the Ikaros repressor protein that also colocalizes with centromeric heterochromatin . Sis contains a strong binding site for the zinc-finger transcription regulator CTCC-binding factor ( Ctcf ) [21] , [22] . Interestingly , deletion of the Sis element or conditional deletion of the Ctcf gene in the B cell lineage both resulted in reduced κ0 germline transcription and enhanced proximal Vκ usage [21] , [23] . Very recently , a novel Ctcf binding element located directly upstream of the Sis region was shown to be essential for locus contraction and recombination to distal Vκ genes [23] . In addition , the Igκ repertoire is controlled by the polycomb group protein YY1 [24] . Induction of Igκ rearrangements requires the expression of the Rag1 and Rag2 proteins , the attenuation of the cell cycle , and transcriptional activation of the Igκ locus , all of which are thought to be crucially dependent on pre-BCR signaling [4] , [25] . At first , pre-BCR signals synergize with interleukin-7 receptor ( IL-7R ) signals to drive proliferative expansion of IgH μ+ large pre-B cells [4] . In these cells , transcription of the Rag genes is low and the Rag2 protein is unstable due to cell-cycle-dependent degradation [26] . Subsequently , signaling through the pre-BCR downstream adapter Slp65 ( SH2-domain-containing leukocyte protein of 65 kDa , also known as Blnk or Bash ) switches cell fate from proliferation to differentiation [4] . Importantly , Slp65 ( i ) induces the TF Aiolos , which down-regulates λ5 expression [27]; ( ii ) binds Jak3 and thereby interferes with IL-7R signaling [28]; and ( iii ) reduces inhibitory phosphorylation of Foxo TFs [29] . All these changes result in attenuation of the cell cycle and thus Rag protein stabilization . Moreover , Rag gene transcription is induced by Foxo proteins [30] . Although rearrangement and expression of the Igκ locus can occur independently of IgH μ chain expression [31] , [32] , several lines of evidence indicate that pre-BCR signaling is actively involved in inducing Igκ and Igλ locus accessibility and gene rearrangement . First , surface IgH μ chain expression correlates with germline transcription in the Igκ locus [33] . Second , in the absence of Slp65 , κ0 germline transcription is reduced [34] . Third , mice deficient for Bruton's tyrosine kinase ( Btk ) , which is a pre-BCR downstream signaling molecule interacting with Slp65 , show reduced Igλ L chain germline transcription and reduced Igλ usage [35] . Fourth , transgenic expression of the constitutively active E41K-Btk mutant in IgH μ chain negative pro-B cells induces premature rearrangement and protein expression of Igκ L chain [34] . Based on fluorescence in situ hybridization ( FISH ) studies , it has been proposed that in pro-B cells distal Vκ and Cκ genes are separated by large distances and that the Igκ locus specifically undergoes contraction in small pre-B and immature B cells actively undergoing Vκ-Jκ recombination [36] . However , it remains unknown how pre-BCR-induced signals affect the accessibility , contraction , and topology of the Vκ region , or how they affect the long-range interactions of the κ regulatory elements involved in organizing these events . In this study , we identified the effects of pre-BCR signaling on germline Vκ transcription and on the expression of TFs implicated in the regulation of Igκ gene rearrangement . We found that the decrease in pre-BCR signaling capacity in wild-type , Btk-deficient , Slp65-deficient , and Btk/Slp65 double-deficient pre-B cells was paralleled by a gradient of decreased expression of many TFs including Ikaros , Aiolos , Irf4 , and ( to a lesser extent ) E2a , as well as by a decreased Igκ locus accessibility for recombination . Several of these factors can mediate long-range chromatin interactions and are known to occupy κ regulatory elements that regulate locus accessibility [37]–[40] . We therefore sought to analyze the effect of pre-BCR signaling on the higher order chromatin structure organized by these regulatory sequences at the Igκ locus . To this end , we performed chromosome conformation capture and sequencing ( 3C-seq ) analyses [41] on pro-B cells and pre-B cells from mice single or double deficient for Btk or Slp65 to evaluate the effects of this pre-BCR signaling gradient on Igκ locus topology . These 3C-seq experiments demonstrated that already in pro-B cells the κ enhancers robustly interact with the ∼3 . 2 Mb Vκ region and its flanking sequences , and that pre-BCR signaling induces accessibility by a functional redistribution of enhancer-mediated chromatin interactions within the Vκ region .
Whereas mice deficient for the pre-BCR signaling molecules Btk and Slp65 have a partial block at the pre-B cell stage [42] , [43] , in Btk/Slp65 double-deficient mice , only very few pre-B cells show progression to the immature B cell stage characterized by functional IgL chain gene recombination [44] . To enable analysis of the effects of pre-BCR signaling on ( i ) the expression of genes involved in Igκ gene rearrangement and on ( ii ) long-distance chromatin interactions in the Igκ locus in pre-B cells in the absence of Igκ gene recombination events , we bred Btk and Slp65 single- and double-deficient mice on the Rag1−/− background . In these mice , progression of B cell progenitors to the pre-B cell stage was conferred by the transgenic , functionally rearranged VH81x IgH μ chain , which ensures pre-BCR expression and cellular proliferation . The absence of functional Rag1 protein precludes IgL chain gene rearrangement and cells are completely arrested at the small pre-B cell stage ( Figure 1A ) . We performed genome-wide expression profiling of FACS-purified B220+CD19+ pre-B cell fractions from wild-type ( WT ) , Btk , and Slp65 single- and double-deficient VH81x transgenic Rag1−/− mice ( Figure 1A ) . In these experiments non-VH81x transgenic Rag1−/− pro-B cells served as controls . One-way ANOVA analysis using MeV software ( p<0 . 01 ) [45] revealed that 266 genes were differentially expressed between the five groups of pro-B/pre-B cells ( Figure 1B ) . When compared with WT VH81x transgenic Rag1−/− pre-B cells , 174 genes were up-regulated , whereby the average values of the fold increase were ∼1 . 70 , ∼3 . 28 , ∼3 . 36 , and ∼3 . 47 for Btk−/− , Slp65−/− , Btk−/−Slp65−/− VH81x transgenic Rag1−/− pre-B cells and non-VH81x transgenic Rag1−/− pro-B cells , respectively ( see Table S1 ) . A similar gradient of gene expression changes was apparent from the average values of the fold change for the 192 significantly down-regulated genes , which were ∼1 . 65 , ∼2 . 29 , ∼3 . 79 , and ∼4 . 15 in the four groups of pre-B/pro-B cells , respectively ( see Table S2 ) . In a hierarchical clustering analysis of the five groups of B cell precursors , the expression profiles of Btk−/−Slp65−/− VH81x transgenic Rag1−/− pre-B cells and non-VH81x transgenic Rag1−/− pro-B cells were very similar ( Figure 1B ) . This implies that expression of the 266 genes is not substantially influenced by pre-BCR-mediated proliferation , which is still induced in pre-B cells lacking both Btk and Slp65 [44] , [46] but not in Rag1−/− pro-B cells . Consistent with these findings , gene distance matrix analysis revealed a clear gene expression gradient among the five groups of pre-B/pro-B cells , in which Btk−/−Slp65−/− pre-B and Rag1−/− pro-B cells again showed highly comparably expression signatures ( Figure S1 ) . In agreement with previous findings [34] , [43] , [46] , pre-BCR signaling-defective pre-B cells manifested increased expression of Dntt , encoding terminal deoxynucleotidyl transferase and the SLC components Vpre ( Vpreb1 ) and λ5 ( Igll1 ) , as well as decreased expression of the cell surface markers Cd2 , Cd22 , Cd25 ( IL-2R ) , and MHC class II ( Table 1 ) . Btk and Slp65 single-deficient and particularly double-deficient pre-B cells failed to up-regulate various genes known to be involved in IgL chain recombination , such as Ikzf3 ( Aiolos ) , Ikzf1 ( Ikaros ) , Irf4 , Spib , Pou2f2 ( Oct2 ) , polymerase-μ [47] , as well as Hivep1 encoding the Mbp-1 protein , which has been shown to bind to the κ enhancers [48] . In addition , pre-BCR signaling influenced the expression levels of many other DNA-binding or modifying factors that were not previously associated with IgL chain recombination , including Lmo4 , Zfp710 , Arid1a/3a/3b , the lysine-specific demethylases Aof1 and Phf2 , Prdm2 ( a H3K9 methyltransferase ) , the sik1 gene encoding a histon deacetylase ( HDAC ) kinase , Hdac5 , Hdac8 , and the DNA repair protein gene Rev1 ( Table S2 ) . We did not find significant differences in the expression of several other TFs implicated in Ig gene recombination—for example , Obf1/Oca-B , Pax5 , E2a , and Irf8 ( Table 1 ) . In addition , in signaling-deficient pre-B cells , we found reduced transcription of genes encoding several signaling molecules ( e . g . , Rasgrp1 , Rapgefl1 , Ralgps2 , Blk , Traf5 , Hck , Nfkbia ( IκBα ) , Syk , Csk ) , cell surface markers ( Cd38 , Cd72 , Cd74 , Cd55 , and Notch2 ) , or genes regulating cell survival ( Bmf and Bcl2l1 encoding BclXL ) ( Table S2 ) . Interestingly , we observed concomitant up-regulation of signaling molecules that are also associated with the T cell receptor ( Lat , Zap70 , and Prkcq ( PKCθ ) ; Table S1 ) . Next , we used quantitative RT-PCR to confirm the observed differential expression of several TFs . Expression levels of these genes were indeed significantly reduced in a pre-BCR signaling-dependent manner , especially for Aiolos , Ikaros , and Irf4 , with residual expression levels in Btk−/−Slp65−/− VH81x transgenic Rag1−/− pre-B cells that were ∼1% , ∼20% , and ∼9% of those observed in WT VH81x Rag1−/− mice , respectively ( Figure 1C ) . In addition , we found moderate effects on Obf1 ( Oca-B ) and E2a with residual expression levels of ∼28% and ∼44% , respectively . In chromatin immunoprecipitation ( ChIP ) assays , we observed in pre-B cells substantial binding of E2a protein to the intronic and 3′ κ enhancer regions and to the three Vκ regions analyzed . Under conditions of reduced pre-BCR signaling activity , E2a binding to the enhancers was essentially maintained ( 3′Eκ ) or reduced ( iEκ ) , but E2a binding to the Vκ regions was lost ( Table S3 ) . Consistent with the significant reduction of Ikaros expression in Slp65−/− pre-B cells , Ikaros binding to both κ enhancers and Vκ regions was undetectable in these cells ( Table S3 ) . Taken together , from these findings we conclude that the five groups of pro-B/pre-B cells , representing a gradient of progressively diminished pre-BCR signaling , show in parallel a gradient of diminished modulation of many genes that signify pre-B cell differentiation , including key genes implicated in Igκ gene recombination . In these expression profiling studies , we only detected limited differences in germline transcription ( GLT ) over unrearranged Jκ and Vκ gene segments , which is thought to reflect locus accessibility [12] . However , we previously showed by serial-dilution RT-PCR that the levels of κ0 0 . 8 and κ0 1 . 1 germline transcripts , which are initiated in different regions 5′of Jκ and spliced to the Cκ region [49] , are apparently normal in Btk−/− pre-B cells , modestly reduced in Slp65−/− pre-B cells , and severely reduced in Btk−/−Slp65−/− pre-B cells [34] . We could confirm these findings for κ0 GLT by quantitative RT-PCR assays on FACS-purified B220+CD19+ pro-B/pre-B cell fractions ( Figure 2A ) . In agreement with our reported findings [34] , we also found that Btk−/−and Slp65−/− pre-B cells have defective λ0 transcription , which is initiated 5′ of the Jλ segments ( Figure 2B ) [49] . GLT across the Vκ region showed a similar pattern of sensitivity to pre-BCR signaling: decreased transcription of six individual Vκ regions tested ( Vκ3–7 , Vκ8–24 , Vκ4–55 , Vκ10–96 , Vκ1–35 , and Vκ2–137 ) correlated with decreased pre-BCR signaling activity ( Figure 2C ) in the pre-B cells of the four groups of mice . GLT over unrearranged Vλ1 and Vλ2 segments was strongly reduced in the absence of Btk or Slp65 , as detected by the expression arrays ( Table 1 ) . These observations indicate that Igκ locus accessibility , a hallmark of recombination-competent antigen receptor loci , is progressively reduced under conditions of diminishing pre-BCR signaling . Accessibility of antigen receptor loci for V ( D ) J recombination is thought to be initiated by enhancers , in part through long-range chromatin interactions with promoters of noncoding transcription , resulting in the activation of germline transcription [8] . Because pre-BCR signaling affects the expression of GLT and various nuclear proteins that mediate long-range chromatin interactions and bind the κ enhancers , it is conceivable that pre-BCR signaling induces changes in the enhancer-mediated higher order chromatin structure of the Igκ locus that facilitates Vκ gene accessibility . We therefore performed 3C-Seq analyses on FACS-purified B220+CD19+ fractions from the same five groups of mice ( WT , Btk−/− , Slp65−/− , and Btk−/−Slp65−/− VH81x transgenic Rag1−/− pre-B cells , as well as Rag1−/− pro-B cells ) . Erythroid progenitors were analyzed in parallel as a nonlymphoid control , in which the Igκ locus was not contracted . Genome-wide chromatin interactions were measured for three regulatory elements involved in the control of Igκ locus accessibility and recombination: the iEκ and 3′Eκ enhancers [50]–[52] and the Sis element [20] , which contain binding sites for Ikaros/Aiolos , E2a , and Irf4 [16] , [17] , [20] , [38] , [53] . In WT pre-B cells , all three regulatory elements showed extensive long-range chromatin interactions within the Vκ region and substantially less interactions with regions up- or downstream of the ∼3 . 2 Mb Igκ domain ( Figure 3A; see Figure S2 , Figure S3 , and Figure S4 for line graphs ) , confirming previous observations [21] . Under conditions of reduced pre-BCR signaling activity , the three Igκ regulatory elements still showed strong interactions with the Vκ region . Surprisingly , even in the complete absence of pre-BCR signaling in Rag1−/− pro-B cells , long-range interactions were still observed at frequencies well above those seen in nonlymphoid cells , suggesting that a contracted Igκ locus topology is not strictly dependent on pre-BCR signaling ( Figure 3A , Figure S2 , Figure S3 , and Figure S4 ) . Next , we used 3D DNA FISH analyses using BAC probes hybridizing to the distal Vκ and Cκ/enhancer regions to confirm that Igκ locus contraction was similar in Rag-1−/− pro-B cells and VH81x transgenic Rag-1−/− pre-B cells ( both showing a contracted topology , compared with noncontracted pre–pro-B cells deficient for the TF E2a; Figure 3B ) . Nevertheless , we did observe that pre-BCR signaling induced clear differences in interaction frequencies . Whereas an increase in pre-BCR signaling was associated with a decrease in the interaction frequencies between the two κ enhancers and regions flanking the Igκ locus ( as also revealed by more detailed images of selected regions upstream and downstream of the Igκ domain; see Figure S5 ) , the overall interaction frequency within the Igκ domain appeared unchanged ( Figure S3 , Figure S4 , and Figure S5 ) . Remarkably , interactions with the Sis element showed quite an opposite pattern: pre-BCR signaling correlated with increased overall interactions within the Igκ domain and did not substantially affect interaction frequencies in the Igκ flanking regions ( Figure S2 and Figure S5 ) . Taken together , these analyses show that ( i ) the Igκ locus is already contracted at the pro-B cell stage and that ( ii ) pre-BCR signaling induces changes in long-range chromatin interactions , both within the Igκ locus and in the flanking regions . The differential effects of pre-BCR signaling on long-range chromatin interactions of the iEκ , 3′Eκ , and Sis elements clearly emerged in a quantitative analysis of the 3C-seq datasets ( Figure 4A; see Materials and Methods for a detailed description of the quantification methods used ) . When pre-BCR signaling was absent ( Rag1−/− pro-B cells ) or very low ( Btk−/−Slp65−/− pre-B cells ) , the average interaction frequencies were similar within the ∼3 . 2 Mb Vκ region and the ∼3 . 2 Mb downstream flanking region , for all three regulatory elements . Interaction frequencies with the upstream flanking region were lower , consistent with the larger chromosomal distance to the three viewpoints . The presence of increasing levels of Btk/Slp65-mediated pre-BCR signaling was associated with reduced interaction of iEκ and 3′Eκ with the Igκ flanking regions and with increased interaction of the Sis element and ( to a lesser extent ) 3′Eκ with the Vκ region ( Figure 4A ) . As a result , for all three regulatory elements pre-BCR signaling resulted in a preference for interaction with fragments inside the Vκ region over fragments outside the Vκ region ( Figure S7 ) . We next focused our analysis on the Vκ region and compared fragments that harbor a functional Vκ gene ( Vκ+ fragment ) and those that do not ( Vκ− fragment ) . When pre-BCR signaling was absent ( Rag1−/− pro-B cells ) or very low ( Btk−/−Slp65−/− pre-B cells ) , the average interaction frequencies of the Sis or iEκ elements with Vκ+ fragments were higher than with Vκ− fragments . The average interaction frequencies of 3′Eκ with Vκ+ and Vκ− fragments , however , were similar ( Figure 4B ) . Upon pre-BCR signaling , the Sis element showed an increase in interaction frequencies with both Vκ+ and Vκ− fragments , with nevertheless an interaction preference for Vκ+ fragments . In contrast , interaction frequencies between the iEκ element and Vκ+ or Vκ− fragments were not modulated by pre-BCR signaling at all ( Figure 4B ) . The 3′Eκ element exhibited yet another profile: pre-BCR signaling induced increased interaction frequencies specifically with Vκ+ fragments , while interactions with Vκ− fragments were not notably modulated by pre-BCR signaling ( Figure 4B ) . When we separately analyzed nonfunctional pseudo-Vκ genes , we found for the Sis and 3′Eκ elements that the interaction patterns with functional and nonfunctional Vκ genes were similar ( Figure S8 ) . In contrast , the iEκ enhancer did show an overall increased interaction frequency with Vκ functional genes , compared with nonfunctional Vκ genes , a phenomenon which was again independent from pre-BCR signaling ( Figure S8 ) . The finding that interactions of Vκ genes with the intronic enhancer are already robust in pro-B cells , while those with the 3′κ enhancer are dependent on pre-BCR signaling , suggested that for individual Vκ genes pre-BCR signaling may result in more similar interaction frequencies with the two enhancers . To investigate this , we examined for all individual Vκ genes the correlation between their 3C-seq interaction frequencies with the iEκ and 3′κ elements and found that these were highly correlated in WT pre-B cells ( R2 = 0 . 68; Figure 4C ) . Correlation was severely reduced when pre-BCR signaling was low in Btk−/−Slp65−/− pre-B cells ( R2 = 0 . 26; Figure 4C ) . Similar pre-BCR signaling-dependent correlations were observed between Vκ-interactions with the Sis element and those with the two enhancers ( Figure S9 ) . As the Sis element particularly suppresses recombination of the proximal Vκ3 family , we investigated interaction correlations specifically for this Vκ family . Similar to our findings for all Vκ genes , a subanalysis showed strong correlations for the interactions of Vκ3 family genes with iEκ , 3′κ , and Sis in WT pre-B cells , which were diminished when pre-BCR signaling was low , except for iEκ–Sis correlations , which were pre-BCR signaling-independent ( Figure S9 ) . In summary , we conclude that pre-BCR signaling induces a redistribution of long-range interactions of the iEκ , 3′Eκ , and Sis elements , thereby restricting interactions towards the Vκ gene region . Moreover , upon pre-BCR signaling the long-range interactions mediated by 3′Eκ and Sis—but not those mediated by iEκ—become enriched for fragments harboring a Vκ gene , demonstrating increased proximity of 3′Eκ and Sis to Vκ genes . Finally , for individual Vκ genes , the interactions with iEκ , 3′Eκ , and Sis become highly correlated upon pre-BCR signaling , indicating that pre-BCR signals result in regulatory coordination between these three elements that govern Igκ locus recombination . In contrast , interactions between genes of the proximal Vκ3 family , Sis and iEκ—but not 3′κ—appear to be coordinated already in the absence of pre-BCR signaling . Next , we investigated the effects of pre-BCR signaling on the interaction frequencies of individual functional Vκ genes with the three κ regulatory elements ( Figure 5A , B ) . The 3C-seq patterns of the majority ( ∼91% ) of the 101 individual Vκ+ fragments showed evidence for interaction with one or more of the κ regulatory elements ( >25 average counts ) . When comparing Btk−/−Slp65−/− with WT pre-B cells , we observed that for a large proportion ( ∼38–52% ) of Vκ+ fragments , interaction frequencies increased upon pre-BCR signaling ( Figure 5B ) . Smaller proportions of Vκ+ fragments showed a decrease ( ∼12–29% ) or were not significantly affected by pre-BCR signaling ( ∼17–25% with <1 . 5-fold change ) . The observed increase or decrease was not related to proximal or distal location of the Vκ genes , nor to their sense or antisense orientation ( not shown ) . Distributions of the three different classes of Vκ+ fragments showed substantial differences between the κ regulatory elements . For the Sis and 3′Eκ elements , more Vκ+ fragments showed increased than decreased interactions ( Figure 5B ) , in agreement with the signaling-dependent increase in average interaction frequencies of all Vκ+ fragments ( Figure 4B ) . In contrast , for the iEκ viewpoint , Vκ+ fragments showing increased and decreased interactions were more equal in number , consistent with the limited effects of pre-BCR signaling on overall iEκ interaction frequencies of all Vκ+ fragments ( Figure 4B ) . Although antigen receptor recombination is in principle regarded as a random process , a significant skewing of the primary Igκ repertoire of C57BL/6 mice was recently reported: one third of the Vκ genes was shown to account for >85% of the Vκ segments used by B cells [54] . To assess whether a correlation exists between usage of Vκ genes and their interaction frequencies with κ regulatory elements , we divided the Vκ genes into four usage categories ( <0 . 1% , 0 . 1–0 . 3% , 0 . 3–0 . 5% , and >0 . 5% ) and calculated their average 3C-Seq interaction frequencies with Sis , iEκ , and 3′κ ( Figure 5C ) . In WT pre-B cells , Vκ usage showed a strong positive correlation with 3C-Seq interaction frequencies for all three regulatory elements ( R2 = ∼0 . 7–0 . 9; Figure 5C ) . These correlations were pre-BCR signaling-dependent , since in Btk−/−Slp65−/− pre-B cells , they were reduced ( for iEκ; R2 = 0 . 33 ) or absent ( for Sis and 3′κ; R2<0 . 10 and R2<0 . 16 , respectively ) ( Figure 5C ) . Collectively , our results indicate that specifically the most frequently used Vκ genes are the main interaction targets of κ regulatory elements , whereby pre-BCR signaling completely underlies this specificity for the Sis and 3′Eκ elements , and to a lesser extent for iEκ . Next , we investigated whether long-range interactions between κ regulatory elements and the Vκ region correlated with the presence of the TFs Ctcf [21] , Ikaros [55] , and E2a [56] , which have been implicated in Igκ locus recombination [21] , [37] , [55] , [57] , [58] . Notably , Ikaros and E2a both strongly bind all three κ regulatory elements , while the Sis element is also occupied by Ctcf ( [21]; unpublished data ) . Remarkably , we found similar striking correlations between the presence of in vivo binding sites for each of these TFs ( as determined by ChIP experiments; see Materials and Methods for the relevant references ) and long-range chromatin interactions with the κ regulatory elements ( Figure 6A–C ) , even though Ctcf sites are mostly located in between Vκ genes [21] and Ikaros/E2a sites were frequently found close to Vκ gene promoter regions ( [2]; Figure 7A ) . Even when pre-BCR signaling was absent ( Rag1−/− pro B cells ) or very low ( Btk−/−Slp65−/− pre-B cells ) , the average interaction frequencies of the κ regulatory elements with fragments containing Ctcf , Ikaros , or E2a bindings sites were higher than those without binding sites . Irrespective of the presence or absence of bindings sites for these TFs , we found that upon pre-BCR signaling interaction frequencies with the Sis element increased and those with the iEκ did not change . In contrast , for the 3′Eκ we found that pre-BCR signaling specifically increased interaction frequencies with fragments occupied by Ctcf , Ikaros , or E2a . Finally , we found that the presence of di- or trimethylation of histone 3 lysine 4 ( H3K4Me2/3 ) , an epigenetic signature associated with locus accessibility [59] and Rag-binding [60] , [61] , also correlated with increased interaction frequencies with κ regulatory elements , revealing a similar pre-BCR signaling dependency as seen for the TFs analyzed ( Figure 6D ) . We conclude that the presence of essential TFs or H3K4Me2/3 in the Vκ region strongly correlates with the formation of long-range chromatin interactions with the κ regulatory elements , and that for the Sis and 3′Eκ elements this interaction preference is further enhanced by pre-BCR signaling . Since the long-range interactions with κ regulatory elements correlated with the presence of TFs implicated in Igκ recombination , we next asked whether the κ regulatory elements preferentially interacted with Vκ genes that are in close proximity to binding sites for Ctcf , Ikaros , or E2a . Strikingly , the majority of functional Vκ genes ( 95/101 ) was found to have an Ikaros binding site in close proximity—that is , located on the same 3C-seq restriction fragment ( average length of ∼3 kb , unpublished data ) ( Figure 7A ) . Proximity of Vκ genes to an E2a binding site ( 37% ) or H3K4Me2/3 positive region ( ∼28% ) is more selective , while only a small fraction of Vκ genes are close to Ctcf binding sites ( ∼12% ) ( [22]; Figure 7A ) . All Vκ genes marked by E2a , Ctcf , H3K4Me2/3 , or a combination of these also contain an Ikaros binding site . Frequently used Vκ genes ( >1 . 0% usage; 33/101 genes ) were located in two separate regions , a proximal and a distal region , which also contained virtually all E2a and H2K4Me2/3-marked Vκ genes ( Figure 7A ) . We found that Vκ genes marked by both Ikaros and E2a were used substantially more often than those only bound by Ikaros ( Figure 7B ) , suggesting that these Vκ genes are preferentially targeted for Vκ-to-Jκ gene rearrangement . Our 3C-seq analyses showed that in WT pre-B cells , interaction frequencies with the three κ regulatory elements were higher for Ikaros/E2a-marked Vκ genes compared to genes marked by Ikaros binding alone ( Figure 7C ) . In fact , Vκ+ restriction fragments containing an Ikaros binding site but not an E2a binding site showed interaction frequencies similar to Vκ− restriction fragments . Under conditions of very low pre-BCR signaling ( in Btk−/−Slp65−/− pre-B cells ) , we observed strongly reduced interaction frequencies of Vκ+ E2a binding restriction fragments with the Sis and 3′Eκ elements . These interaction frequencies were in the same range as those of Vκ− fragments or Vκ+ fragments that harbored an Ikaros site only ( Figure 7C ) . Interaction frequencies with the iEκ enhancer , however , were independent of pre-BCR signaling . As shown in Figure 7D , for the majority of Ikaros/E2a-marked Vκ+ fragments ( 65% ) , pre-BCR signaling was associated with increased interactions with the Sis and 3′Eκ elements ( comparing wild-type and Btk−/−Slp65−/− pre-B cells ) . In these analyses , only ∼13 . 5% and ∼5 . 4% of Ikaros/E2a-marked Vκ+ fragments showed a decreased interaction frequency upon pre-BCR signaling . In contrast , almost equal proportions of Ikaros/E2a-marked Vκ+ fragments showed increased ( ∼37% ) and decreased ( ∼30% ) interactions with iEκ upon pre-BCR signaling . Taken together , these data reveal strong positive correlations between the presence of E2a binding sites , Vκ usage , and long-range chromatin interactions with κ regulatory elements in pre-B cells . Remarkably , for the iEκ element , these correlations are largely independent of Btk/Slp65-mediated pre-BCR signaling , whereas for the 3′Eκ they are completely dependent on signaling .
During B-cell development the pre-BCR checkpoint is known to regulate the expression of many genes , part of which control the increase in Igκ locus accessibility to the V ( D ) J recombinase complex . However , it remained unknown how pre-BCR signaling events affect accessibility in terms of Igκ locus contraction and topology . Here we identified numerous genes involved in IgL chain recombination , chromatin modification , signaling , and cell survival to be aberrantly expressed in pre-B cells lacking the pre-BCR signaling molecules Btk and/or Slp65 . We found that GLT over the Vκ region , reflecting Vκ accessibility , is strongly reduced in these cells . We used 3C-Seq to show that in pro-B cells both the intronic and the 3′ κ enhancers frequently interact with the ∼3 . 2 Mb Vκ region , as well as with Igκ flanking sequences , indicating that the Igκ locus is already contracted at the pro-B cell stage . 3C-Seq analyses in wild-type and Btk/Slp65 single- and double-deficient pre-B cells demonstrated that pre-BCR signaling significantly affects Igκ locus topology . First , pre-BCR signaling reduces the interactions of the intronic and 3′κ enhancers with Igκ flanking regions , effectively focusing enhancer action towards the Vκ region to facilitate Vκ-to-Jκ recombination . Second , pre-BCR signaling strongly increases nuclear proximity of the 3′κ enhancer to Vκ genes , whereby this increase is more substantial for more frequently used Vκ genes and for Vκ genes close to a binding site for the basic helix-loop-helix protein E2a . Third , pre-BCR signaling augments interactions between κ regulatory elements and fragments within the Vκ region bound by the key B-cell TFs Ikaros and E2a and the architectural protein Ctcf . Fourth , pre-BCR signaling has limited effects on interactions of the intronic κ enhancer with fragments within the Igκ locus , as this enhancer already displays interaction specificity for functional Vκ genes and TF-bound regions in pro-B cells . Fifth , pre-BCR signaling has limited effects on the interactions between the intronic or 3′κ enhancers and fragments that do not contain a Vκ gene or an Ikaros , E2a , or Ctcf binding site , emphasizing the specificity of pre-BCR signaling-induced changes in Igκ locus topology . Sixth , pre-BCR signaling appears to induce mutual regulatory coordination between the three regulatory elements , as their interaction profiles with individual Vκ genes become highly correlated upon signaling . Finally , pre-BCR signaling increases interactions of the Sis element with DNA fragments in the Igκ locus , irrespective of the presence of a Vκ gene or TF . Collectively , our findings demonstrate that pre-BCR signals relayed through Btk and Slp65 are required to create a chromatin environment that facilitates proper Igκ locus recombination . This multistep process is initiated by up-regulation of key TFs like Aiolos , Ikaros , Irf4 , and E2a . These proteins are then recruited to or further accumulate at the Igκ locus and its regulatory elements , resulting in a specific fine-tuning of enhancer-mediated locus topology that increases locus accessibility to the Rag recombinase proteins . Importantly , the presence of strong lineage-specific interaction signals between the Cκ/enhancer region and distal Vκ genes in pro-B cells indicates that the Igκ locus is already contracted at this stage . In contrast to a previous microscopy study indicating that Igκ locus contraction did not occur until the small pre-B cell stage [36] , our 3D DNA FISH analysis indeed detected similar nuclear distances between distal Vκ and the Cκ/enhancer region in cultured pro-B and pre-B cells . Recently Hi-C was employed to study global early B cell genomic organization whereby substantial interaction frequencies were found between the intronic κ enhancer and the Vκ region in pro-B cells [40] . E2a-deficient pre–pro-B cells , which are not yet fully committed to the B-cell lineage [62] , showed very few interactions among the iEκ and the distal part of the Vκ region [40] , resembling the interactions we observed in nonlymphoid cells ( Figure 3A ) . Accordingly , 3D-FISH analysis showed that the Igκ locus adopted a noncontracted topology in these pre–pro-B cells ( Figure 3B ) . These data indicate that Igκ locus contraction is already achieved in pro-B cells and depends on the presence of E2a . Supporting this notion , active histone modifications and E2a were already detected at the κ enhancers and Vκ genes at the pro-B cell stage [56] , [63] , whereby E2a was frequently found at the base of long-range chromatin interactions together with Ctcf and Pu . 1 , possibly acting as “anchors” to organize genome topology [40] . The observed correlation between E2a binding , Vκ gene usage and iEκ proximity in pro-B cells ( Figure 5C , Figure 7C ) further strengthens an early critical role for E2a in regulating Igκ locus topology , Vκ gene accessibility , and recombination . Our 3C-seq experiments revealed that pre-BCR signaling is not required to induce long-range interactions between the κ regulatory elements and distal parts of the Vκ locus , indicating that TFs strongly induced by signaling—that is , Aiolos , Ikaros , and Irf4—are not strictly necessary to form a contracted Igκ locus . Prime candidates for achieving Igκ locus contraction at the pro-B cell stage are E2a and Ctcf , as they have been implicated in regulating Ig locus topology [21] , [40] , [64] , [65] and E2a already marks frequently used Vκ genes at the pro-B cell stage ( Figure 7 ) , although we did observe reduced E2a expression and binding to the iEκ enhancer and Vκ genes when pre-B cell signaling was low ( Figure 1 and Table S3 ) , suggesting that pre-BCR signaling is required for high-level E2a occupancy of the Vκ genes . We previously reported that Igκ gene recombination can occur in the absence of Ctcf and that Ctcf mainly functions to limit interactions of the κ enhancers with proximal Vκ regions and to prevent inappropriate interactions between these strong enhancers and elements outside the Igκ locus [21] . Because at the pro-to-pre–B cell transition Aiolos , Ikaros , and Irf4 are recruited to the Igκ locus and histone acetylation and H3K4 methylation increases [17] , [38] , [63] , [66] , we hypothesize that pre-BCR–induced TFs act upon an E2a/Ctcf-mediated topological scaffold to further refine the long-range chromatin interactions of the κ regulatory elements . Hereby , these TFs mainly act to focus and to coordinate the interactions of the two κ enhancers to the Vκ gene segments , in particular to frequently used Vκ genes , thereby increasing their accessibility for recombination ( see Figure 7E for a model of pre-BCR signaling-induced changes in Igκ locus accessibility ) . In this context , our 3C-seq data show that the two κ enhancer elements have distinct roles . Both 3′Eκ and iEκ elements manifest interaction specificity for highly used , E2a-marked , Vκ genes . However , whereas iEκ already shows this specificity in pro-B cells ( although pre-BCR signaling does augment this specificity ) , 3′Eκ only does so in pre-B cells upon pre-BCR signaling . These observations indicate that iEκ is already “prefocused” at the pro-B cell stage and that pre-BCR signals are required to fully activate and focus the 3′Eκ to allow synergistic promotion of Igκ recombination by both enhancers ( see Figure 7E ) [52] . In agreement with such distinct sequential roles , iEκ and not the 3′Eκ was found to be required for the initial increase in Igκ locus accessibility , which occurred upon binding of E2a only [37] , [38] , [67] . The 3′Eκ on the other hand requires binding of pre-BCR signaling-induced Irf4 to promote locus accessibility [19] , [38] , followed by further recruitment of E2a to both κ enhancers and highly used Vκ genes ( Table S3 and [38] , [57] ) . The Sis regulatory element was shown to dampen proximal Vκ–Jκ rearrangements and to specify the targeting of Igκ transgenes to centromeric heterochromatin in pre-B cells [20] . As Sis is extensively occupied by the architectural Ctcf protein and deletion of Sis or Ctcf both resulted in increased proximal Vκ usage [21] , [23] , it was postulated that Sis functions as a barrier element to prevent the κ enhancers from too frequently targeting proximal Vκ genes for recombination . In this context , we now provide evidence that interactions between the proximal Vκ genes , Sis , and iEκ—but not 3′κ—are already coordinated before pre-BCR signaling occurs ( Figure S9 ) . Perhaps not surprisingly , Sis-mediated long-range chromatin interactions displayed a pattern and pre-BCR signaling response that was different from the κ enhancers . Unlike for the enhancers , upon pre-BCR signaling , Sis-mediated interactions with regions outside the Igκ locus were maintained and interaction within the Vκ region increased , irrespective of the presence of Vκ genes or TF binding sites . Because Sis is involved in targeting the nonrecombining Igκ allele to heterochromatin [20] , the observed interaction pattern of the Sis element might reflect its action in pre-B cells to sequester the nonrecombining Igκ locus and target it towards heterochromatin . This might also explain the increased interaction frequencies of Sis with highly used Vκ genes upon pre-BCR signaling ( Figures 5C and 7C ) , as such highly accessible genes likely require an even tighter association with Sis and heterochromatin to prevent undue recombination . Surprisingly , we observed a striking correlation between Ikaros binding and Vκ gene location ( 94% of Vκ genes were in close proximity to an Ikaros binding site; Figure 7A ) . Although Ikaros and Aiolos have a positive role in regulating gene expression during B-cell development [55] , [58] and Ikaros is required for IgH and IgL recombination [39] , [58] , Ikaros has also been reported to silence gene expression through its association with pericentromeric heterochromatin [68] or through recruitment of repressive cofactor complexes [69] , [70] . Recruitment of Ikaros to the Igκ locus was found increased in pre-B cells as compared to pro-B cells [63] , in agreement with its up-regulation in pre-B cells ( Figure 1 ) . Furthermore , Ikaros binds the Sis element , where it was suggested to mediate heterochromatin targeting of Igκ alleles by the Sis region [20] . Aiolos , although not essential for B-cell development like Ikaros [58] , [71] , is strongly induced by pre-B cell signaling and has been reported to cooperate with Ikaros in regulation gene expression [27] . Although their synergistic role during IgL chain recombination has not been extensively studied , the Ikaros/Aiolos ratio changes upon pre-BCR signaling ( Figure 1 ) . Increased recruitment of Ikaros/Aiolos to Vκ genes and the κ enhancers likely increases Igκ locus accessibility and contraction ( see Figure 6 ) , as Ikaros was very recently shown to be essential for IgL recombination [58] . On the other hand , it is conceivable that on the nonrecombining allele , increased recruitment of Ikaros/Aiolos to Vκ genes and the Sis region could facilitate silencing of this allele . Further investigations using allele-specific approaches [72] will be required to clarify the allele-specific action of the Sis element during Igκ recombination . In summary , by investigating the effects of a pre-BCR signaling gradient—rather than deleting individual TFs—we have taken a more integrative approach to study the regulation of Igκ locus topology . Our 3C-Seq analyses in wild-type , Btk , and Slp65 single- and double-deficient pre-B cells show that interaction frequencies between Sis , iEκ , or 3′ Eκ and the Vκ region are already high in pro-B cells and that pre-BCR signaling induces accessibility through a functional redistribution of long-range chromatin interactions within the Vκ region , whereby the iEκ and 3′Eκ enhancer elements play distinct roles .
VH81x transgenic mice [73] on the Rag-1−/− background [74] that were either wild-type , Btk−/− [75] , Slp65−/− [42] , or Btk−/−Slp65−/− have been previously described [34] . Mice were crossed on the C57BL/6 background for >8 generations , bred , and maintained in the Erasmus MC animal care facility under specific pathogen-free conditions and were used at 6–13 wk of age . Experimental procedures were reviewed and approved by the Erasmus University Committee of Animal Experiments . Preparation of single-cell suspensions and incubations with monoclonal antibodies ( mAbs ) were performed using standard procedures . Bone marrow B-lineage cells were purified using fluorescein isothiocyanate ( FITC ) -conjugated anti-B220 ( RA3-6B2 ) and peridinin chlorophyll protein ( PCP ) -conjugated anti-CD19 , together with biotinylated mAbs specific for lineage markers Gr-1 , Ter119 , and CD11b and APC-conjugated streptavidin as a second step to further exclude non-B cells . Cells were sorted with a FACSARia ( BD Biosciences ) . The following mAbs were used for flow cytometry: FITC- , PerCP–anti-B220 ( RA3-6B2 ) , phycoerythrin ( PE ) –anti-CD2 ( LFA-2 ) , PCP- , allophycocyanin ( APC ) - or APC–Cy7–anti-CD19 ( ID3 ) , PE- , or APC anti-CD43 ( S7 ) . All these antibodies were purchased from BD Biosciences or eBiosciences . Samples were acquired on an LSRII flow cytometer ( BD Biosciences ) and analyzed with FlowJo ( Tree Star ) and FACSDiva ( BD Biosciences ) software . Extraction of total RNA , reverse-transcription procedures , design of primers , and cDNA amplification have been described previously [21] . Gene expression was analyzed using an ABI Prism 7300 Sequence Detector and ABI Prism Sequence Detection Software version 1 . 4 ( Applied Biosystems ) . All PCR primers used for quantitative RT-PCR of TFs or κ0 , λ0 , and Vκ GLT are described in [21] , except for Obf1 ( forward 5′-CCTGGCCACCTACAGCAC-3′ , reverse 5′-GTGGAAGCAGAAA CCTCCAT-3′ , obtained from the Roche Universal Probe Library ) . Biotin-labeled cRNA was hybridized to the Mouse Gene 1 . 0 ST Array according to the manufacturer's instructions ( Affymetrix ) ; data were analyzed with BRB-ArrayTools ( version 3 . 7 . 0 , National Cancer Institute ) using Affymetrix CEL files obtained from GCOS ( Affymetrix ) . The RMA approach was used for normalization . The TIGR MultiExperiment Viewer software package ( MeV version 4 . 8 . 1 ) was used to perform data analysis and visualize results [45] . One-way ANOVA analysis of the five experimental groups of B cells was used to identify genes significantly different from wild-type VH81X Tg Rag1−/− pre-B cells ( p<0 . 01 ) . ChIP experiments were performed as previously described [76] using FACS sorted bone marrow pre-B cell fractions ( 0 . 3–2 . 0 million cells per ChIP ) . Antibodies against E2a ( sc-349 , Santa Cruz Biotechnology ) and Ikaros ( sc-9861 , Santa Cruz Biotechnology ) were used for immunoprecipitation . Purified DNA was analyzed by quantitative RT-PCR as described above . Primer sequences are available on request . 3C-Seq experiments were essentially carried out as described previously [21] , [41] . For 3C-Seq library preparation , BglII was used as the primary restriction enzyme and NlaIII as a secondary restriction enzyme . 3C-seq template was prepared from WT E13 . 5 fetal liver erythroid progenitors and FACS-sorted bone marrow pro-B cell or pre-B cell fractions ( see above ) from pools of 4–6 mice . In total , between 1 and 8 million cells were used for 3C-seq analysis . Primers for the Sis , iEκ , and 3′Eκ viewpoint-specific inverse PCR were described previously [21] . 3C-seq libraries were sequenced on an Illumina Hi-Seq 2000 platform . 3C-Seq data processing was performed as described elsewhere [41] , [77] . Two replicate experiments were sequenced for each genotype and viewpoint , and normalized interaction frequencies per BglII restriction fragment were averaged between the two experiments . For quantitative analysis , the Igκ locus and surrounding sequences were divided into three parts ( mm9 genome build ) : a ∼2 Mb upstream region ( chr6:65 , 441 , 978–67 , 443 , 029; 759 fragments ) , a ∼3 . 2 Mb Vκ region ( chr6:67 , 443 , 034–70 , 801 , 754; 1 , 290 fragments ) and a downstream ∼3 . 2 Mb region ( chr6:70 , 801 , 759–73 , 993 , 074; 1 , 143 fragments ) . For each cell type ( as described above ) sequence read counts within individual BglII restriction fragments were normalized for differences in library size ( expressed as “reads per million”; see [74] ) and averaged between the two replicates before further use in the various calculations . Very small BglII fragments ( <100 bp ) were excluded from the analysis . Fragments in the immediate vicinity of the regulatory elements ( chr6:70 , 659 , 392–70 , 693 , 183; 10 fragments ) were also excluded because of high levels of noise around the viewpoint , a characteristic of all 3C-based experiments . Vκ gene coordinates ( both functional genes and pseudogenes ) were obtained from IMGT [11] and NCBI ( Gene ID: 243469 ) databases . Vκ gene usage data ( C57BL/6 strain , bone marrow ) were obtained from [54] . ChIP-seq datasets were obtained from [21] ( Ctcf ) , [55] ( Ikaros ) , and [56] ( E2a , H3K4Me2 , and H3K4Me3 ) . Vκ genes were scored positive for TF binding sites or for a histone modification , if they were located on the same BglII restriction fragment ( corresponding to the 3C-Seq analysis ) . Rag-1−/− pro-B and Rag-1−/−;VH81X pre-B cells were isolated from femoral bone marrow suspensions by positive enrichment of CD19+ cells using magnetic separation ( Miltenyi Biotec ) . Cells were cultured for 2 wk in Iscove's Modified Dulbecco's medium containing 10% fetal calf serum , 200 U/ml penicillin , 200 mg/ml streptomycin , 4 nM L-glutamine , and 50 µM β-mercaptoethanol , supplemented with IL-7 and stem cell factor at 2 ng/ml . E2a−/− hematopoietic progenitors were grown as described previously [78] . Prior to 3D-FISH analysis , cells were characterized by flow cytometric analysis of CD43 , CD19 , and CD2 surface marker expression to verify their phenotype ( Figure S6 ) . 3D DNA FISH was performed as described previously [79] with BAC clones RP23-234A12 and RP23-435I4 ( located at the distal end of the Vκ region and at the Cκ/enhancer region , respectively; Figure 3A ) obtained from BACPAC Resources ( Oakland , CA ) . Probes were directly labeled with Chromatide Alexa Fluor 488-5 dUTP and Chromatide Alexa Fluor 568-5 dUTP ( Invitrogen ) using Nick Translation Mix ( Roche Diagnostics GmbH ) . Cultured primary cells were fixed in 4% paraformaldehyde , and permeabilized in a PBS/0 . 1% Triton X-100/0 . 1% saponin solution and subjected to liquid nitrogen immersion following incubation in PBS with 20% glycerol . The nuclear membranes were permeabilized in PBS/0 . 5% Triton X-100/0 . 5% saponin prior to hybridization with the DNA probe cocktail . Coverslips were sealed and incubated for 48 h at 37°C , washed , and mounted on slides with 10 µl of Prolong gold anti-fade reagent ( Invitrogen ) . Pictures were captured with a Leica SP5 confocal microscope ( Leica Microsystems ) . Using a 63× lens ( NA 1 . 4 ) , we acquired images of ∼70 serial optical sections spaced by 0 . 15 µm . The datasets were deconvolved and analyzed with Huygens Professional software ( Scientific Volume Imaging , Hilversum , the Netherlands ) . The 3D coordinates of the center of mass of each probe were transferred to Microsoft Excel , and the distances separating each probe were calculated using the equation: √ ( Xa−Xb ) 2+ ( Ya−Yb ) 2+ ( Za−Zb ) 2 , where X , Y , and Z are the coordinates of object a or b . Statistical significance was analyzed using a nonparametric Mann–Whitney U test ( IBM SPSS Statistics 20 ) . The p values<0 . 05 were considered significant . 3C-seq and microarray expression datasets have been submitted to the Sequence Read Archive ( SRA , accession number SRP032509 ) and Gene Expression Omnibus ( GEO , accession number GSE53896 ) , respectively . | B lymphocyte development involves the generation of a functional antigen receptor , comprising two heavy chains and two light chains arranged in a characteristic “Y” shape . To do this , the receptor genes must first be assembled by ordered genomic recombination events , starting with the immunoglobulin heavy chain ( IgH ) gene segments . On successful rearrangement , the resulting IgH μ protein is presented on the cell surface as part of a preliminary version of the B cell receptor—the “pre-BCR . ” Pre-BCR signaling then redirects recombination activity to the immunoglobulin κ light chain gene . The activity of two regulatory κ enhancer elements is known to be crucial for opening up the gene , but it remains largely unknown how the hundred or so Variable ( V ) segments in the κ locus gain access to the recombination system . Here , we studied a panel of pre-B cells from mice lacking specific signaling molecules , reflecting absent , partial , or complete pre-BCR signaling . We identify gene regulatory changes that are dependent on pre-BCR signaling and occur via long-range chromatin interactions between the κ enhancers and the V segments . Surprisingly the light chain gene initially contracts , but the interactions then become more functionally redistributed when pre-BCR signaling occurs . Interestingly , we find that the two enhancers play distinct roles in the process of coordinating chromatin interactions towards the V segments . Our study combines chromatin conformation techniques with data on transcription factor binding to gain unique insights into the functional role of chromatin dynamics . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"humoral",
"immunity",
"medicine",
"immune",
"cells",
"immunology",
"cell",
"differentiation",
"developmental",
"biology",
"molecular",
"development",
"epigenetics",
"chromatin",
"chromosome",
"biology",
"gene",
"expression",
"biology",
"genetics",
"of",
"the",
"immune",
... | 2014 | Pre-B Cell Receptor Signaling Induces Immunoglobulin κ Locus Accessibility by Functional Redistribution of Enhancer-Mediated Chromatin Interactions |
Many microbes are studied by examining colony morphology via two-dimensional top-down images . The quantification of such images typically requires each pixel to be labelled as belonging to either the colony or background , producing a binary image . While this may be achieved manually for a single colony , this process is infeasible for large datasets containing thousands of images . The software Tool for Analysis of the Morphology of Microbial Colonies ( TAMMiCol ) has been developed to efficiently and automatically convert colony images to binary . TAMMiCol exploits the structure of the images to choose a thresholding tolerance and produce a binary image of the colony . The images produced are shown to compare favourably with images processed manually , while TAMMiCol is shown to outperform standard segmentation methods . Multiple images may be imported together for batch processing , while the binary data may be exported as a CSV or MATLAB MAT file for quantification , or analysed using statistics built into the software . Using the in-built statistics , it is found that images produced by TAMMiCol yield values close to those computed from binary images processed manually . Analysis of a new large dataset using TAMMiCol shows that colonies of Saccharomyces cerevisiae reach a maximum level of filamentous growth once the concentration of ammonium sulfate is reduced to 200 μM . TAMMiCol is accessed through a graphical user interface , making it easy to use for those without specialist knowledge of image processing , statistical methods or coding .
Budding yeasts , such as Saccharomyces cerevisiae ( baker’s yeast ) , are able to change their pattern of growth on a solid substrate in response to the surrounding nutrient level . At sufficiently high nutrient levels , buds separate from the mother cell following cytokinesis to produce colonies that appear close to circular when viewed from above . When nutrient is not readily available , the cells reproduce via the pseudohyphal growth pattern , which is characterised by distal unipolar budding ( budding opposite to the birth scar ) , the elongation of cells , and a persistent connection between mother and daughter cell [1] . As a result of these changes , the colony develops a number of filaments that grow along and into the substrate . This growth mode has important consequences both ecologically and medically . For example , pseudohyphal growth increases the virulence of the pathogenetic yeast Candida albicans [2] . Owing to the widespread occurrence of yeasts in the production of foods such as bread , wine and beer , and the need to restrict the growth of drug-resistant yeast colonies on catheters and other medical equipment [3 , 4] , it is important to understand and classify strain-specific properties and growth characteristics . These features are usually investigated via two-dimensional top-down images of colonies grown on a solid medium [1 , 5–7] . The growth patterns observed in these experimental images are typically quantified using binary versions of the images , which indicate whether or not each pixel is part of the colony [7–9] . While it is possible to manually convert a single image to binary with sufficient accuracy using image analysis software , this task is difficult in studies of dimorphic growth , which may involve hundreds of images [7] , and is infeasible for large datasets produced using genome-wide mutant libraries , which consist of thousands of images [6] . The analysis of large datasets thus requires two elements: the automated conversion of colony images to binary , and robust statistics that enable quantification of the spatial patterns . The conversion of an image to binary requires each pixel to be placed into one of two categories ( in this case , the colony and background ) based upon a set of criteria . A variety of methods capable of performing this operation are available . The simplest approach to this task is thresholding , through which each pixel is categorised depending upon whether its intensity is greater than a given tolerance level . This approach is limited by the need to select a suitable tolerance , which may differ for each image and , for large datasets , must be chosen automatically to make processing feasible . There is a significant body of work regarding image thresholding and a variety of methods for achieving this are available , as illustrated in the reviews by Weszka [10] , Sahoo , Soltani and Wong [11] , and Sezgin and Sankur [12] . Common methods for selecting the tolerance include Otsu’s method [13] , the Ridler–Calvard method [14 , 15] , k-means clustering [16 , 17] , a watershed transformation [18 , 19] and DBSCAN [20] . Some existing techniques require particular lighting [21] or cells to be marked , such as by a fluorescent compound [22 , 23] , to facilitate the image analysis . Colony identification has been performed using specialised lighting techniques combined with multilevel thresholding [24 , COVASIAM] , through manual thresholding on multiple layers to create three-dimensional binary images [25 , COMSTAT] , and by using local thresholding around areas where colonies were expected to grow [26] . Methods have also been developed to identify individual cells using a mix of different processing methods [27 , 28 , CellProfiler] , edge detection [23 , 29] , Otsu’s method [21 , 30] and a combination of Otsu’s method and a watershed transformation [31] . Software [32 , CalMorph] and algorithms [19] tailored for yeast have also been produced for identifying and analysing individual cells . The production of binary images has been used to quantify images by examining the selected pixels in binary images produced using a range of thresholds [22] . Software for counting bacterial colonies using Otsu’s method has been developed for mobile phones [33 , Colonizer] . The choice of method is dependent upon the particular application and its computational cost . The demands of processing large datasets require that the method provides no greater accuracy than is needed in order to make the analysis tractable . The statistics required thus influence the choice of method , and consideration must be given to these in order to determine the appropriate accuracy for the image processing . Previous studies of yeast colonies have relied on a variety of commercial packages to quantify the images [6]; however , these produce a limited range of statistics , such as the colony area or the change in pixel intensity pre- and post-wash . Ruusuvuori et al . [8] developed the web-based application Yeast Image Analysis ( YIMMA ) for processing and analysing images of yeast colonies , which converts images to binary representations by first applying a global intensity threshold to the green channel of a colour image , followed by post-processing to clean the image . The images are quantified by 427 features , such as the area , perimeter , and fractal dimension . Cross-validation analysis found that only 6 features were required to classify colonies as either smooth or fluffy . A set of three spatial indices designed specifically to quantify the growth of filamentous yeast colonies [7] have been shown to provide useful information on the morphology of yeast colonies and microbial mats [34 , 35] . To facilitate the analysis of colony images , we have developed the software Tool for Analysis of the Morphology of Microbial Colonies ( TAMMiCol ) . This software converts images of microbial colonies to binary for analysis using either in-build statistics or by computing other statistics after export . The binary images are produced by applying a threshold to a greyscale image . Importantly , this threshold is determined efficiently and automatically for each image by exploiting the structure of microbial colony images , which provides an advantage over generic methods for image analysis . TAMMiCol is able to process images provided there is some contrast between the colony and background , so does not require prior marking of cells and thus may be used to analyse existing datasets , as demonstrated here . While fluorescent marking is not required , any enhancements to the contrast between the colony and background may improve performance . TAMMiCol provides several advantages over existing software used for converting images to binary , such as ImageJ [36] . Multiple images may be imported simultaneously and converted using a different threshold computed for each image without the need for the user to record macros . Post-processing steps are applied automatically and checks are performed on the binary images . TAMMiCol is able to compute specialised spatial indices [7] , which are shown to be sufficiently robust so that small differences between binary images produced manually and those produced automatically do not significantly alter the statistics . Critically , use of TAMMiCol does not require knowledge of image processing and all features are accessed through a graphical user interface , placing the ability to analyse images directly with experimentalists . This package thus makes it possible to produce useful statistics automatically from experimental images alone , and opens up an avenue to study large datasets in greater detail than has previously been possible . The images produced by this automated method are found to be of similar quality to images produced manually . Furthermore , because TAMMiCol is designed for images of microbial colonies , images produced using this method provide better agreement than images produced by standard image segmentation techniques . Through an analysis of new data using TAMMiCol , we find that colonies of the yeast S . cerevisiae reach a maximum level of filamentous growth once the concentration of ammonium sulfate is reduced to 200 μM . While largely applied to filamentous yeast colonies , this approach is shown to work for biofilms and other microbial colonies . TAMMiCol is free and publicly available from github . com/HaydenTronnolone/TAMMiCol . Installation files are available for both macOS and Windows . TAMMiCol is written in MATLAB using the App Designer package and , while not requiring a MATLAB license in order to operate , does need MATLAB Runtime , which contains necessary shared libraries and will be downloaded automatically during the TAMMiCol installation . Once TAMMiCol is downloaded , installation should be complete in under five minutes . TAMMiCol does not require additional plugins and is designed for microbial colony analysis , so provides seamless operation .
We consider a two-dimensional experimental image with dimensions Lx and Ly in the horizontal and vertical directions , respectively . The image is first converted to greyscale and , without loss of generality , we assume that the colony is darker than the background , as the same method may be applied when the colony is lighter by inverting the image . The colony is typically located near to the centre of the image , and so is assumed to lie within a rectangle of the same aspect ratio as the original image but 10% of the size , which ensures that some part of the colony is captured . The darkest pixel p in this central rectangle is assumed to be part of the colony and the corresponding intensity of this pixel denoted cp . Since p lies within the colony , pixels with intensities within some integer tolerance τ of cp that are near to p are thus also assumed to represent part of the colony . Accordingly , for each tolerance τ ∈ [τs , τf] for some chosen minimum tolerance τs and maximum tolerance τf , a binary image is produced containing all pixels with intensities in the interval [cp − τ , cp + τ] that form a contiguous region that contains p . By requiring that the colony comprises a connected region , other artefacts in the image , such as contaminants , are automatically removed . The binary images produced for each tolerance τ are quantified using the proportion χ ( τ ) of pixels selected for each tolerance , as illustrated in Fig 1 . The proportion χ ( τ ) is a non-decreasing function of τ , since increasing the tolerance can introduce additional pixels but not remove them . At low values of τ , the selected region will typically be a strict subset of the colony . As τ increases , more pixels from the colony are included in the selected region . Once τ becomes sufficiently large , the selected region will expand beyond the colony and start to include the background of the image . The transition as the selected region expands beyond the boundary of the colony appears as a rapid increase in the value of χ . Once τ is sufficiently large , the entire image is selected , corresponding to the maximum proportion χ = 1 . The tolerance at which this first occurs is denoted τu . Having created a binary image for each tolerance τ , it remains to choose which best represents the colony . The best image is assumed to occur just before parts of the background are included at the tolerance τb . This point is identified by optimising a piecewise linear fit to χ , as illustrated in Fig 1 . A critical tolerance τc is selected from the region after the rapid increase in χ . In general , we choose τc = τu; however , the fit may be adjusted by varying the value of τc . For each midpoint tolerance τm ∈ [τs + 1 , τc − 1] , the proportion χ is approximated by a piecewise linear function pm ( τ ) . This interpolant has two components that are defined on the intervals [τs , τm] and [τm + 1 , τc] . These are chosen to agree with χ exactly at the end points τs and τc , and at the two adjacent values τm and τm + 1 where the intervals meet . These conditions completely specify the four coefficients of the piecewise polynomial and may be written as p m ( τ s ) = χ ( τ s ) , p m ( τ m ) = χ ( τ m ) , p m ( τ m + 1 ) = χ ( τ m + 1 ) , p m ( τ c ) = χ ( τ c ) . For each choice of midpoint τm , we compute the mean error δ ( τm ) between the proportions χ and the interpolant pm . The optimal threshold τb is taken to be the value of τm corresponding to the minimum of δ ( τm ) . Although not typically needed , the tolerance may be adjusted manually . This may be required when analysing images comprising several disconnected pieces , which bring added difficultly as the increase in χ near the optimal choice of τ may be more gradual . When processing more than one image , the tolerance selection will be repeated independently for each image , so that image may thus be processed using a different tolerance . A detailed example of this process for sample 5 of the AWRI 796 50 μM dataset produced by Binder et al . is given in the Supplementary Material S1 Text . Three operations are performed on the selected image . Since it is assumed that the colony lies entirely within the image , it is expected that the filtered image will not contain any selected pixels around the boundary . If boundary pixels are selected , the tolerance is decreased automatically by one unit at a time until no pixels lie on the boundary . Next , any pixels with fewer than four neighbours in the cardinal directions are removed . While this removes a small number of pixels from the colony boundary , it ensures that any stray pixels located outside of the colony are not included in the binary image . Finally , if the user has indicated that the colony is connected , the largest connected piece in the binary image is identified and all other elements are removed from the image . The binary images are then saved as either CSV or MATLAB MAT files , which may be analysed using other software or by using the statistics incorporated into TAMMiCol , described in the next subsection . Filamentous yeast colonies may be quantified using the spatial indices introduced by Binder et al . [7] with some modifications , explained here briefly . To validate the binary images produced by TAMMiCol , we compare the statistics computed from the automated method with those computed from manually processed images , and examine the statistics produced from new large datasets that are infeasible to analyse manually . In the following definitions , the total number of occupied pixels is denoted ν and the maximum colony radius , as measured from the colony centroid , is denoted R . From these two quantities , the average density is calculated as ρ = ν/πR2 . To quantify the distribution in the radial direction , we count the number of pixels in nr concentric annuli centred on the colony . Instead of using annuli of equal width , as per Binder et al . [7] , the annuli are chosen to each have area A = πR2/nr . The number of pixels in the jth annulus is denoted cr ( j ) . This is scaled by the number of pixels expected to lie within the annulus if the pixels were distributed uniformly , yielding the scaled counts f r ( j ) = c r ( j ) ρ A = n r c r ( j ) ν . The function fr has mean value 1 , which is not true when using annuli of equal width . The value at which fr ( j ) first drops below 1 is called the complete spatial randomness ( CSR ) radius RCSR . Using this value , the radial distribution is characterised by the index I r = 1 - R CSR R ∈ [ 0 , 1 ] . This index measures non-uniform growth in the radial direction . The angular distribution is quantified by computing the angular distance between each pair of pixels . Since the large number of possible pairs ν ( ν − 1 ) /2 makes this computationally expensive , we repeatedly sample νΘ pixels chosen randomly from the population and average over the samples . Throughout this study , we take at most 103 pixels and calculate the average of 103 samples . The differences are grouped into nΘ bins with widths π/nΘ and the corresponding bin counts are denoted cΘ ( j ) . These counts are scaled by the expected count for a uniform distribution to give the scaled counts f Θ = 2 n Θ c Θ ( j ) ν Θ ( ν Θ - 1 ) , which again has mean 1 . Binder et al . [7] introduced the index IΘ = fΘ ( 1 ) , which is a measure of local aggregation . This index is largest when all the pairs lie in the first bin , in which case fΘ ( 1 ) = nΘ . An improved index is given by scaling by this maximum value , resulting in I Θ = f Θ ( 1 ) n Θ ∈ [ 0 , 1 ] . This index measures the aggregation of cells . The maximum variance of fΘ occurs when all the pairs lie in a single bin . In this case Var ( f Θ ) = n Θ - 1 . The spread of the cells is measured by the index I CSR = Var ( f Θ ) n Θ - 1 ∈ [ 0 , 1 ] . This index measures non-uniform growth in the angular direction . For each of the indices , larger values indicate greater variation in the morphology . For the purposes of this work , each index is calculated using 200 bins . In addition to the three indices described here , TAMMiCol produces several other values that quantify the morphology , which are described in the Supplementary Material S1 Text .
To examine the performance of the automated method , we validate the binary images produced using the filamentous yeast datasets produced by Binder et al . [7] and summarised in Table 1 . These datasets comprise 270 raw images of filamentous yeast colonies , along with binary versions of each image that were processed manually using image processing software . The binary images produced by TAMMiCol are available online [37] . All analysis by TAMMiCol was performed using the default settings . To first illustrate this method , we consider the experimental image of colony 5 from the AWRI 796 50 μM dataset after 233 hours of growth . The original image , the region selected by TAMMiCol and the corresponding proportions χ plotted against τ are shown in Fig 2 . The selected level and the value at which all pixels are selected are also marked on this plot . The overlaid image shows that TAMMiCol is able to separate the colony from the background with a high degree of accuracy . This is confirmed by the plot of the pixel proportions , which shows that the point just before the rapid increase in the number of pixels was correctly selected . The suitability of the automated method may be quantified by comparing the images produced by TAMMiCol to the images produced manually by Binder et al . [7] . For each pair of images , we counted both the number of pixels νu selected in the union of the manual and automated images ( pixels selected as part of the colony in at least one image ) and the number of pixels νi selected in the intersection ( pixels selected as part of the colony in both images ) . A graphical representation of this process for sample 5 of the AWRI 796 50 μM dataset produced by Binder et al . is given in the Supplementary Material S1 Text . The relative percentage difference between the images is defined to be d = ν u - ν i ν u × 100 ∈ [ 0 , 100 ] , which is the Jaccard distance expressed as a percentage . This represents the percentage of selected pixels that differ between the two images relative to the total number of pixels that are considered part of the colony by at least one of the methods . If both images have the same set of selected pixels then d takes the value 0 , while if there are no common selected pixels between the two images then d = 100 . If the pixels in both images are selected at random with probability 0 . 5 , then d = 200/3 ≈ 66 . 7 . The values of d for each image from the datasets described in Table 1 are shown in Fig 3 . For each dataset , the difference is typically less than 10% and many samples have percentage difference values around 5% or lower . Furthermore , the mean differences d ¯ over each dataset , given in Table 1 , are all less than 7% . This indicates that the automated and manual images are in close agreement . Extreme value distributions [38] and Wilcoxon’s rank-sum test [39] have been used previously to perform statistical tests to determine whether agreement indicated by d is due to chance alone . It has been noted , however , that statistical tests performed on Cohen’s kappa coefficient , which measures agreement in a similar fashion to d , rarely indicate that agreement is due to chance alone [40] . As this observation also applies to d , we do not report formal statistical tests on these results but instead note that they indicate a high level of agreement . To examine the performance of TAMMiCol , the trial datasets were also processed using a selection of standard methods implemented using MATLAB . The methods considered were: ( 1 ) Otsu’s method for threshold selection [13]; ( 2 ) the Ridler-Calver method [14] , which is the default algorithm used by the commercial image processing software ImageJ; ( 3 ) k-means++ clustering [16 , 41]; ( 4 ) a watershed transformation [18] using Meyer’s flooding algorithm [42]; and ( 5 ) DBSCAN [20] using natural patterns [43] . The watershed transformation ( method 4 ) was not able to produce viable images , while DBSCAN ( method 5 ) , which has computational complexity O ( N2 ) for the number of pixels N , proved infeasible for the image sizes considered here . Otsu’s method ( method 1 ) , the Ridler–Calvard method ( method 2 ) and k-means++ clustering ( method 3 ) produced viable images , which were compared with the manual images in the same manner as for TAMMiCol . The respective mean differences d ¯ Otsu , d ¯ RC and d ¯ kmeans for these methods are given in Table 1 . The mean differences for each dataset are lower for images produced by TAMMiCol than for images produced by either Otsu’s method , the Ridler–Calvard method or k-means++ clustering . As the images can be quantified using the in-built spatial indices , it is of interest to compare the values produced by the TAMMiCol and the manual images , which are plotted in Fig 4 . Both methods produce similar results , which indicates that the automated method provides sufficiently accurate images for the computation of the indices . Importantly , despite the differences in the indices , the automated and manual images generally agree on the relative order of the statistics for each of the datasets . Having validated TAMMiCol using existing data , we next demonstrate the computational efficiency of this software using a larger collection of new images . It is well known that colonies of S . cerevisiae produce filamentous growth when starved of nitrogen [1] . In a study of 1026 strains of S . cerevisiae , 56% displayed filamentous growth in low-nitrogen conditions , and large variations in morphology were observed [44] . Furthermore , changes in growth have been observed due to the nitrogen source used [45] . Much work has been devoted to identifying the signalling pathways responsible for the transition to filamentous growth [46] , while global gene-deletion assays have been performed to identify the genes that control filamentous growth [6] . While some studies have attempted to quantify the observed behaviour [6 , 7] , little quantitative information is available relating the nitrogen level and colony shape . To address this , colonies of AWRI 796 were grown on agar at five different ammonium sulfate concentrations between 50 μM and 500 μM at 30°C , and imaged daily for ten days , as summarised in Table 2 . The experiment was stopped at this point in order to avoid using images that had been distorted due to evaporation from the agar . Whereas the previous datasets considered comprised a total of 270 images and could be processed manually , the new datasets contain 690 images and , as such , manual processing of the images is infeasible . The images were thus processed using TAMMiCol only , with spatial indices computed from the resulting binary data . All analysis by TAMMiCol was performed using the default settings . Using a MacBook Pro running OS X 10 . 10 . 5 with a 2 . 5 GHz Intel Core i7 processor , each individual image was converted to binary and the statistics computed in approximately 20 seconds . This highlights how TAMMiCol permits analyses that were previously infeasible . The binary images produced by TAMMiCol are available online [37] . To compare the effect of ammonium sulfate concentration , the spatial indices Ir , IΘ and ICSR were averaged over each concentration , with the resulting mean values plotted in Fig 5 . In general , the indices show that filamentous growth increases with decreasing nutrient . After approximately 100 hours of growth , the colonies begin to show filamentous behaviour and the values for 350 μM and 500 μM become distinct from the other concentrations , which remain grouped closer together , suggesting that , for ammonium concentrations of 200 μM or less , the colonies reach a maximum level of filamentous growth . This indicates that a threshold concentration of ammonium is required for cells to grow in the yeast form . Below this , filamentation is triggered ( presumably as a response to nitrogen stress ) and some cells then switch to grow in the filamentous form . Other environmental conditions may affect the exact ammonium threshold required to trigger filamentous growth , such as the density of the medium and the concentrations of other nutrients . While all the examples thus far have considered filamentous yeast colonies , the methods presented here are applicable to a variety of other cases . To illustrate this , we consider an S . cerevisiae ( L2056 ) biofilm produced by Tam et al . [34] , and a colony of Bacillus subtilis produced by Fujikawa and Matsushita [47] , both of which are shown with the colony identified by TAMMiCol in Fig 6 . In both cases , TAMMiCol is able to identify the colony with a high degree of accuracy , demonstrating the versatility of the software . The statistics used here are also suitable for analysing these colonies [34 , 35] .
We have introduced the software TAMMiCol , which converts photographs of microbial colonies to binary images automatically and in a computationally efficient manner . The binary images are produced using thresholding with the tolerance chosen by exploiting the structure of the images . By tailoring the method for images of microbial colonies , TAMMiCol produces results comparable with manual image processing and better than those produced by standard image segmentation methods , while the graphical user interface gives experimentalists direct access to quantification methods without the need for specialist knowledge of image processing or coding . TAMMiCol is free and publicly available from github . com/HaydenTronnolone/TAMMiCol . Using this method , it is possible to analyse large datasets that would be infeasible to process manually , as each image may be converted to binary and quantified in approximately 20 seconds using a laptop computer . In comparison , it can take up to 15 minutes to manually process an image . While TAMMiCol has been demonstrated using examples containing up to 690 images , the same procedure makes it possible to quantify the morphology from datasets containing thousands of images , such as genome-wide deletion mutant libraries . Therefore , TAMMiCol provides the opportunity for future research on identifying specific genes responsible for measurable growth characteristics . Through an analysis of a new large dataset using TAMMiCol , we have shown that colonies of the yeast S . cerevisiae reach a maximum level of filamentous growth once the concentration of ammonium sulfate is reduced to 200 μM . TAMMiCol presents several advantages over existing software used for producing binary images , such as ImageJ [36] . The default method for producing binary images in ImageJ is through the IsoData algorithm , which is a version of the Ridler–Calvard method [14] tested here and which was outperformed overall by TAMMiCol . TAMMiCol automatically undertakes a number of steps in addition to thresholding to remove artefacts from the images . While similar steps could be performed using other software , this would need to be done manually by the user , which could be difficult without experience in image processing . TAMMiCol is designed for batch processing , while ImageJ requires the user to record a macro in order to process multiple images , and other software may lack this capability altogether . Finally , TAMMiCol is able to organise the binary output and produce statistics to describe the data . This means that TAMMiCol takes raw images as input and produces appropriate statistics in one action , so that data are available to the user without the need for specialist skills in image processing , data management or shape quantification . We believe that no other software provides the ease of use and combination of features available through TAMMiCol . While we have shown that the indices developed by Binder et al . [7] are able to quantify the spatial pattern , the binary images produced by this method are not limited to these measures . The processed images may be accessed by other image-analysis software and other features examined , such as the perimeter or number of connected pieces . Furthermore , the methods described here may be used to analyse a wide variety of microbial colonies and other images with similar features . This work thus opens an avenue to efficiently quantify numerous large datasets using either the indices provided or any other custom statistics . This has the potential to provide new insights that were previously unobtainable , and to motivate new experimental work that was previously unsupported by statistical analysis . While the current version of TAMMiCol has been designed to convert images of microbial colonies to binary , the methods employed here could be used to analyse a variety of images . This includes , but is not limited to , images of scratch assays , tumour spheroids and vegetation patterns . While TAMMiCol is expected to be able to convert other images , the method could be improved by employing alternative methods for identifying the best threshold that would provide additional user control , so that the best method could be selected . Future versions of TAMMiCol may be further generalised by incorporating a variety of algorithms to assist in selecting the best binary image , and by analysing individual colour channels rather than a greyscale image . While the indices computed by TAMMiCol have been designed to quantify microbial colonies , future versions may include additional indices and pair-correlation functions that would permit general-purpose analysis . | Many microbes are studied by examining the colony morphology via a two-dimensional top-down image . In order to quantify such images , we typically need to label each pixel as belonging either to the colony or the background , creating a binary image . This task is laborious when performed manually and proves infeasible for large datasets . To overcome this , we have developed the software Tool for Analysis of the Morphology of Microbial Colonies ( TAMMiCol ) , which automatically and efficiently converts colony images to binary . Multiple images may be imported and processed simultaneously , and TAMMiCol exploits the structure of the images to identify an appropriate threshold for the binary conversion of each image . The images produced by TAMMiCol , which take around 20 seconds each to process , compare favourably with images processed manually , which take anywhere up to 15 minutes , while TAMMiCol outperforms several standard image segmentation methods . After processing , the images may be exported as a CSV or MATLAB MAT file for further analysis , or may be quantified by TAMMiCol using the in-built statistics . Using TAMMiCol , we have found that colonies of S . cerevisiae reach a maximum level of filamentous growth once the concentration of ammonium sulfate is reduced to 200 μM . | [
"Abstract",
"Introduction",
"Design",
"and",
"implementation",
"Results",
"and",
"discussion",
"Availability",
"and",
"future",
"directions"
] | [
"medicine",
"and",
"health",
"sciences",
"chemical",
"compounds",
"pathology",
"and",
"laboratory",
"medicine",
"engineering",
"and",
"technology",
"signal",
"processing",
"pathogens",
"applied",
"mathematics",
"salts",
"microbiology",
"simulation",
"and",
"modeling",
"... | 2018 | TAMMiCol: Tool for analysis of the morphology of microbial colonies |
An essential feature of meiosis is Spo11 catalysis of programmed DNA double strand breaks ( DSBs ) . Evidence suggests that the number of DSBs generated per meiosis is genetically determined and that this ability to maintain a pre-determined DSB level , or “DSB homeostasis” , might be a property of the meiotic program . Here , we present direct evidence that Rec114 , an evolutionarily conserved essential component of the meiotic DSB-machinery , interacts with DSB hotspot DNA , and that Tel1 and Mec1 , the budding yeast ATM and ATR , respectively , down-regulate Rec114 upon meiotic DSB formation through phosphorylation . Mimicking constitutive phosphorylation reduces the interaction between Rec114 and DSB hotspot DNA , resulting in a reduction and/or delay in DSB formation . Conversely , a non-phosphorylatable rec114 allele confers a genome-wide increase in both DSB levels and in the interaction between Rec114 and the DSB hotspot DNA . These observations strongly suggest that Tel1 and/or Mec1 phosphorylation of Rec114 following Spo11 catalysis down-regulates DSB formation by limiting the interaction between Rec114 and DSB hotspots . We also present evidence that Ndt80 , a meiosis specific transcription factor , contributes to Rec114 degradation , consistent with its requirement for complete cessation of DSB formation . Loss of Rec114 foci from chromatin is associated with homolog synapsis but independent of Ndt80 or Tel1/Mec1 phosphorylation . Taken together , we present evidence for three independent ways of regulating Rec114 activity , which likely contribute to meiotic DSBs-homeostasis in maintaining genetically determined levels of breaks .
In most sexually reproducing organisms , meiotic recombination is initiated by programmed catalysis of DNA double strand breaks ( DSBs ) by Spo11 , an evolutionarily conserved type II topoisomerase-like transesterase [1] . In Saccharomyces cerevisiae , where the process is best understood , Spo11 activity requires nine additional proteins , five of which are meiosis specific ( Rec102 , Rec104 , Rec114 , Mei4 , and Mer2 ) , and four that are expressed during both meiosis and vegetative growth ( Rad50 , Mre11 , Xrs2 , and Ski8 ) [2] . These proteins interact with each other and/or with Spo11 to form a complex referred to as the Spo11- or DSB-complex , or DSB-machinery , and participate in the Spo11 transesterase reaction that leads to the formation of a DSB ( reviewed in [2] ) . Meiotic DSBs are essential for meiosis; nevertheless , each break represents a potentially lethal or mutagenic DNA lesion that must be repaired before the first meiotic division ( MI ) . As such , Spo11 catalysis is tightly regulated at the temporal , spatial , and quantitative levels . For instance , the catalysis does not normally take place until the locus has undergone replication [3] , [4] . When it occurs , DSB-catalysis takes place preferentially at loci referred to as DSB hotspots rather than randomly throughout the genome [5]–[7] . The number of breaks catalyzed per meiosis is also developmentally programmed; in yeast or mammals , the number is approximately 150–250 per meiosis , whereas in Drosophila , it is about 25 [6]–[10] . Maintaining the number of meiotic DSBs at the developmentally programmed level would require both positive and negative means of regulating break formation . Although much is known about the genetic requirements for DSB formation [2] , factors and mechanisms involved in monitoring the extent of breakage and/or limiting the number of breaks remain largely elusive . Recent studies suggested a role for the mammalian ATM kinase and its Drosophila and budding yeast homologs , tefu+ and TEL1 , respectively , in down-regulating meiotic DSB formation [8] , [9] , [11] . These proteins are members of the ATM/ATR family of conserved signal transduction kinases involved in fundamental DNA/chromosomal processes such as DNA replication , DNA damage repair , recombination , and checkpoint regulation [12] , [13] . They also play a key role ( s ) in many essential meiotic processes including interhomolog bias in DSB repair [14] , meiotic recombination checkpoint regulation [15] , and sex chromosome inactivation in mammals [16] . Here we present evidence that Rec114 , an evolutionarily conserved Spo11-accessory protein and an essential component of the meiotic DSB-machinery [2] , is a direct target of Tel1/Mec1 , the budding yeast ATM/ATR homologues . Several Spo11-accessory proteins are proposed to be anchored at the chromosome axes and interact transiently with DSB hotspots at chromatin loops to promote cleavage [17]–[21] . Tel1/Mec1 phosphorylation of Rec114 upon DSB formation down-regulates its interaction with DSB hotspots and leads to reduced levels of Spo11 catalysis . Further analyses showed two additional means of down-regulating Rec114: synapsis associated removal at the onset of pachytene , as previously observed [17] , [22] , and Ndt80-dependent turnover . We propose a model whereby multiple means of regulating Rec114 activity contribute to meiotic DSB homeostasis in maintaining the number of breaks at the developmentally programmed level .
Budding yeast Tel1 and Mec1 , like their mammalian counterparts , ATM and ATR , are serine/threonine kinases [23] . These kinases preferentially phosphorylate their substrates on serine ( S ) or threonine ( T ) residues that precede glutamine ( Q ) residues , so called SQ/TQ or [S/T]Q motifs . Many known targets of the ATM/ATR family proteins contain [S/T]Q cluster domains ( SCDs ) , defined as a region where three or more SQ or TQ motifs are found within a tract of 100 residues or less [24] . As a means to investigate a role of Tel1/Mec1 in regulating DSB formation , we explored the possibility that they might directly phosphorylate one or more of the nine Spo11-accesssory proteins mentioned above . Rec114 , an evolutionarily conserved meiosis specific chromosomal protein , was the most likely target with eight SQ/TQ consensus phosphorylation sites , seven of which are found in two clusters , referred to as SCD1 and SCD2 ( Figure 1A ) . Western blot analysis using polyclonal α-Rec114 antibodies [17] revealed the appearance of slower migrating Rec114 species ( Figure 1A ) . The putative phosphorylated isoform ( s ) of Rec114 was more prominent in a strain expressing a tagged version of REC114 , REC114-13xMYC ( Figure 1B “WT” ) . The tagged version also persisted for longer , showing that despite conferring full spore viability the tag changed some of Rec114's characteristics ( see below ) . In both REC114 and REC114-13xMYC strains , the slower migrating species became prominent by 4 hours , corresponding to meiotic prophase in the current experimental condition [14] . DSBs formed by Spo11 activates Tel1/Mec1 , which in turn , directly phosphorylate a number of targets including H2AX , Sae2/Com1 , ( the ortholog of human CtIP ) , Hop1 , and Zip1 [14] , [25]–[27] . To test whether the Rec114 phosphorylation was also dependent on meiotic DSBs , we assessed the effect of spo11-Y135F , a catalytically inactive allele of SPO11 [1] . The gel shift was not detected in protein from spo11-Y135F strains , indicting it is dependent on DSB formation ( Figure 1B ) . Next , we tested the dependence of the Rec114 mobility shift on TEL1/MEC1 . To this end , we assessed Rec114 migration patterns in a rad24Δ tel1Δ strain . In a rad24Δ tel1Δ strain , the Tel1/Mec1 signaling is down-regulated to a level comparable to that in mec1Δ tel1Δ cells kept viable by a suppressor mutation , sml1Δ; however , rad24Δ tel1Δ cells do not exhibit the severe meiotic progression defect observed in the latter [14] . We found that Rec114 mobility shift was reduced in a rad24Δ tel1Δ background ( Figure 1B ) . The reduction was also observed at the restrictive temperature in a tel1Δ strain carrying the temperature sensitive mec1-4 allele [28] ( Figure 1G ) . Defects in meiotic recombination or synapsis activate Tel1- or Mec1- checkpoint response [12] , [14] , [15] , [26] , [27] , [29] . In rad50S , mre11S ( “S” for separation of function ) , or com1Δ/sae2Δ backgrounds , Spo11 remains covalently bound to the break ends , preventing their further processing . Accumulation of unprocessed meiotic DSBs in these mutants triggers a TEL1-dependent checkpoint response [30]–[32] . Elimination of the meiotic recombinase Dmc1 , on the other hand , leads to accumulation of hyper-resected break ends that are loaded with single strand DNA ( ssDNA ) binding proteins and activates a MEC1-mediated checkpoint response [15] , [33] . During Tel1- or Mec1-checkpoint response , a number of targets , including Hop1 and Com1/Sae2 , remain hyper-phosphorylated , reflecting the increased kinase activity of Tel1/Mec1 . We found that both the extent and duration of Rec114 mobility shift seemed also enhanced in a rad50S or dmc1Δ background ( Figure 1C ) , consistent with the possibility that Rec114 might be a target of Tel1/Mec1 . To further address the role ( s ) of Tel1/Mec1 in Rec114 mobility shift , we examined its migration pattern in a strain expressing a rec114 allele , rec114-8A , where all of the S or T residues of the eight Tel1/Mec1 consensus sites were replaced by a non-phosphorylatable alanine ( A ) . We found that Rec114 mobility shift was abolished in a rec114-8A dmc1Δ strain ( Figure 1D ) , indicating that the observed shift is due to a modification ( s ) at one or more of the eight Tel1/Mec1 consensus sites . To confirm in vivo phosphorylation of Rec114 at a specific residue ( s ) during normal meiosis , we generated phospho-specific antibodies against three of the eight ATM/ATR consensus sites in Rec114 . T175 and S187 were chosen based on their biological relevance ( Table 1; see analysis below ) ; S265 was selected using a software tool that predicts kinase-specific phosphorylation sites ( GPS 2 . 1; Supporting Online Material ) . Using these phospho-specific antibodies , we performed Western blot analyses on samples taken from strains expressing either WT or the non-phosphorylatable allele , rec114-8A . The results showed that each of the three phospho-specific antibodies generated signals in the WT samples but not the rec114-8A , confirming in vivo phosphorylation of Rec114 at these three sites ( Figure 1E ) . Finally , we demonstrated that purified Mec1 could directly phosphorylate one or more of the three confirmed in vivo Rec114 phosphorylation sites in vitro ( Figure 1F ) . Taken together , we conclude that Rec114 is a DSB dependent target of Tel1/Mec1 during normal meiosis . To investigate function ( s ) of Tel1/Mec1 phosphorylation of Rec114 , the effect of mutating the S or T residues of the eight Tel1/Mec1 consensus sites was examined . We began the analysis with two rec114 alleles , rec114-8A or rec114-8D , where the eight S or T were mutated to either a non-phosphorylatable alanine ( A ) or to a phospho-mimetic aspartic acid ( D ) residue , respectively . Spore viability of rec114-8A diploids was comparable to that of REC114 in all genetic backgrounds tested ( Table 1 ) . rec114-8D , in contrast , conferred haploinsufficiency and synthetic interactions with mutations that confer either a reduction in DSB-catalysis ( e . g . spo11-HA and spo11-DA ) [34] or sensitivity to such reduction ( e . g . pch2Δ ) [35] ( Table 1 ) . Thus , constitutively mimicking Tel1/Mec1 phosphorylation might be deleterious to meiosis . Alternatively , the effect might be due to protein misfolding caused by the introduction of eight closely spaced negative charges , which might have led to its degradation . Although we cannot rigorously rule out the latter , it appears unlikely , given that chromatin bound Rec1148D is more abundant than Rec114 ( see analysis below ) , and also because replacing as few as two ( T175 and S187 ) of the eight consensus sites with a phosphomimetic residue confers a rec114-8D like phenotype with respect to haploinsufficiency and synthetic interaction with spo11-hypomorphic alleles ( Table 1 ) . Notably , T175 and S187 of Rec114 are confirmed in vivo phosphorylation sites ( Figure 1E ) . The synthetic spore lethality interaction between rec114-phosphomimetic and spo11-hypomorphic alleles , which are known to confer sublethal reductions in crossover ( CO ) levels [34] ( Table 1 ) , suggested that the combined effects of the mutations may result in a lethal deficit in CO-formation . To test this , we assessed the effect of rec114-8D on CO-levels at the well characterized HIS4-LEU2 artificial meiotic recombination hotspot ( Figure 2A ) [36] . rec114-8D conferred a delay in the accumulation of COs , and about 25% reduction in the final level of COs; in rec114-8A , the level of COs was comparable to WT but they appeared earlier ( Figure 2BC ) . A reduction in CO-levels can result from either insufficient DSB levels and/or a defect ( s ) in CO homeostasis [34] . CO homeostasis refers to the notion that CO-levels tend to be maintained at the expense of noncrossovers ( NCOs ) , and is , in part , based on the observation that strains expressing spo11-hypomorphic alleles exhibited only a modest reduction in the levels of COs despite the fact that their DSB levels , assessed in a rad50S background , were significantly lower than WT [34] . To determine whether the reduction in CO-levels in a rec114-8D strain was due to a defect in break formation and/or CO homeostasis , we measured DSB levels in a rec114-8D com1Δsae2Δ or rec114-8D rad50S strain using pulsed field gel electrophoresis ( PFGE ) /Southern analysis ( Figure 2D; data not shown ) . The results showed that rec114-8D confers a dramatic reduction in the levels of DSBs on three different chromosomes examined , ChrIII , V , and VIII ( Figure 2E; Figure S1 ABC; data not shown ) . We conclude that the modest reduction in CO-levels in a rec114-8D strain is likely due to a reduction in DSB levels , and that the observed synthetic interaction between rec114-phosphomimetic and spo11-hypomorphic alleles ( Table 1 ) may result from additive impact of the two mutations on insufficient DSB-catalysis . The above observations suggest that Tel1/Mec1 phosphorylation of Rec114 , mimicked in rec114-8D , down-regulates DSB formation . If so , the absence of the phosphorylation in rec114-8A should lead to an increase in DSB levels , assuming that no other mechanism was acting redundantly . Indeed , a substantial increase could be observed for break sites near YCL064C or YCR048W on ChrIII ( Figure 2EF ) . The extent of the increase was comparable to that observed in tel1Δ , a mutant reported to cause an increase in DSB levels [11] . Since Rec114 is a target of Tel1 and/or Mec1 ( above ) , the latter suggests that Rec114 is likely to be a key target in mediating Tel1 negative regulation in DSB levels . Unlike rec114-8D , whose negative effect on break levels was obvious at all break sites analyzed on ChrIII , V , and VIII , we were only able to document the much subtler positive effect of rec114-8A or tel1Δ on ChrIII with this technology ( Figure 2EF; Figure S1D–E and data not shown ) . The dramatic effect of rec114-8D suggests that phosphorylation of some or all of the sites mutated is sufficient to strongly reduce Spo11 catalysis . The comparably modest increase in rec114-8A mutants , where Rec1148A is insensitive to Tel1/Mec1 negative control via phosphorylation at these sites , suggests that Rec1148A might mainly cause repeated cleavage by the same activated DSB machine near the break on the same chromatid , which would hardly increase the DSB signals measured by Southern; alternatively , it may point to the existence of additional mechanism ( s ) limiting break formation , and that it/they is/are yet to be discovered . Unexpectedly , we found that the negative effect of rec114-8D on break level was notably attenuated in a dmc1Δ background compared to rad50S or com1Δ/sae2Δ ( Figure 2G; data not shown ) . In a rec114-8D dmc1Δ strain , DSB levels reached about 75% of a REC114 dmc1Δ . In a RAD50 DMC1 background , the effect of rec114-8D was intermediate , between rad50S/com1Δ/sae2Δ and dmc1Δ ( Figure S2 ) . These observations show that the control of DSB formation is likely multi-layered and that feedbacks in addition to that by Rec114 phosphorylation exist . As an independent means of assessing the effect of Rec114 phosphorylation on DSB levels , we performed a genome-wide Spo11-chromatin immunoprecipitation ( ChIP ) on CHIP assay ( here on referred to as ChIP-chip ) , which confers greater resolution and offers easier normalization than a Southern blot based analysis ( e . g . [7] , [37] ) . In constructing the required strains for the analysis , we took into account the potential genetic interaction between various epitope tags of Spo11 and rec114 alleles as suggested by reduced spore viability of strains expressing tagged versions of either protein ( Table 1; data not shown ) . We introduced the untagged versions of REC114 , rec114-8A , or rec114-8D alleles into a rad50S strain expressing SPO11-18xMYC . Unlike spo11-6xHIS-3xHA , the SPO11-18xMYC did not affect spore viability of rec114-8D strains ( data not shown ) . Spo11-myc ChIP was performed without the use of formaldehyde ( FA ) cross-linking to enrich for Spo11 proteins that had remained covalently bound to the break ends upon DNA-cleavage . To ensure the highest degree of comparability between the three REC114/rec114 allele backgrounds , the experiments were performed strictly in parallel for all steps from culturing to the final analysis . The resulting profiles of covalently bound Spo11 in the three backgrounds reproduced the published DSB hotspot profiles [7] with great precision ( Figure 3A ) . A small fraction of signals , typically near telomeres and within pericentric regions , however , are not DSB specific , but identical among the three profiles ( Figure 3A , areas denoted by * ) ; these were used to superimpose the profiles ( decile normalization , [17] , Materials and Methods ) . Importantly , the three aligned profiles differ in the amplitude of hundreds of sharply defined positions in an almost invariable pattern: Spo11 signal in rec114-8A is higher than in wild type , while Spo11 in rec114-8D is strongly reduced ( Figure 3A; Figure S3 ) . The results of statistical evaluation of the differences in these peaks is presented in Figure 3C . The following prediction was tested in this analysis: If DSB formation was indeed reduced in rec114-8D relative to rec114-8A , then the ratio of the Spo11 profiles of rec114-8A over rec114-8D , ( hereon referred to as 8A/8D ) , should define DSB sites . In fact , the correlation between DSB hotspots and the 8A/8D peaks should be greater than that of not-normalized profiles . Indeed , profiles of these ratios identify near 100% of the published DSB hotspots ( eg . Figure S3 A , D ) . When peaks of the ratio of these profiles were compared to the mapped hotspots at a resolution of 600 bp , >97% of the 1200 strongest Spo11 8A/8D peaks matched one of the 3600 DSB sites [7] , ( p<10−40 , Figure S4A ) . The same was true for smaller selections; 62% of 500 strongest 8A/8D sites matched one of the 500 strongest DSB sites ( p<10−40 , Figure 3C ) , while 76% of 100 8A/8D matched 100 DSBs ( p<10−20 , Figure S4B ) . More detailed results showing the cumulative curves of distances compared to a null hypothesis ( random ) are provided in Figures 3Ci and Figure S4A , B . Although there are some peaks in the Spo11 profiles , where 8D>8A , less than 1% of the 500 strongest 8D/8A match the 500 DSBs , a strong anti-correlation ( p<10−6 ) that excludes that there is significant 8D>8A at DSB sites ( data not shown ) . Even for the smaller difference between WT and 8A , WT/8A produces a clear anti-correlation ( Figure 3Ci ) . Being independent of decile or any other normalization , this analysis indicates that Spo11 catalysis at nearly all known hotspots is attenuated in the phospho-mimicking rec114-8D background . Furthermore , the degree of attenuation is roughly proportional to the hotspot strength in that the 100 strongest DSB peaks correspond to the 100 strongest Spo11 8A/8D peaks , whereas the 500 strongest DSB peaks to the 500 strongest Spo11 8A/8D peaks . Analysis of the smaller differences between Spo11 profiles in rec114-8A and in REC114 by 500×500 comparison ( 500 hottest DSB hotspots against 500 strongest 8A/WT peaks ) also produced a significant , although somewhat weaker , correlation ( p<10−40 , Figure 3Ci ) . We thus confirm with high significance , that Spo11 signals in the non-phosphorylatable rec114-8A are more abundant than in the wild type background , at least for the 500 strongest hotspots genome wide . The effect of rec114 mutations on the extent of Spo11 catalysis was confirmed further by qPCR analysis at a strong DSB site ( YCR047C , Figure 3Aii ) . Taken together , these results strongly suggest genome-wide down-regulation of Spo11 catalysis by phosphorylation of Rec114 , at least in the rad50S background . Rec114 is a meiotic chromosome axis protein whose recruitment to the chromosomes is essential for Spo11 catalysis [17] , [20] , [22] . To test whether Tel1/Mec1 phosphorylation might down-regulate Spo11 catalysis by affecting Rec114's association with certain chromosomal positions , we performed genome wide Rec114 ChIP-chip analysis in strains expressing untagged versions of Rec114 , Rec1148A or Rec1148D using a polyclonal antibody raised against Rec114 [17] . The analysis of Rec114 ChIP-chip after 4 hours in SPM showed enrichment of Rec114 at chromosome axes located nearby strong DSB hotspots ( Figure 3Bi ) as shown previously [17] , [21] . Similar to the Spo11 profiles , the three Rec114 profiles became perfectly superimposed after decile normalization for many DSB-unspecific low signal peaks ( Figure 3Bi ) . Within DSB-rich domains of ChrIII , signals at axis sites were strongest for Rec1148D , followed by Rec114 and then Rec1148A at axis sites . This relationship was confirmed by qChIP at one axis site over a meiotic time course ( Figure 3Bii ) . Thus , Rec114-axis association appears to correlate negatively with DSB levels . We conclude that the reduction in DSB levels in a rec114-8D strain is not due to defects in Rec1148D -axes interaction . RMM and other Spo11 accessory proteins are proposed to be anchored at the chromosome axes and interact transiently with DSB hotspots at chromatin loops to promote cleavage [17]–[21] . Given the apparent excess of Spo11-accessory proteins relative to the number of breaks catalyzed ( e . g . [20] ) , such transient interaction is expected to manifest as small peaks near hotspots interspersed in a landscape of prominent axis signals . Indeed , for the hyperactive Rec1148A protein , nearly all of the strong DSB hotspots show small peaks overlapping the hotspots ( Figure 3Bi , at 211 . 7kb; Figure 3Biii , v , Figure S5 ) . These DSB associated peaks are stronger in Rec1148A than in wild type and are typically absent in Rec1148D . At strong hotspots , the profiles reversed their order noted above and become Rec1148A>Rec114>Rec1148D , although Rec1148D strongly dominates at the immediately adjacent axis sites ( Figure 3Biii , v , Figure S5 ) . Among the 35 strongest hotspots ( as defined in [7] ) , 33 of them presented Rec1148A>Rec1148D ( p<1 . 6×10−17 ) , and all but one overlapped with local Rec1148A maximum in the DSB cluster ( e . g . Figure 3Biii , iv , v ) . Comparing Rec114 association with a DSB site ( YCR047C ) and its neighboring axis site as a function of time , we observed that the extent of increase at the DSB site ( Figure 3Bvi ) is greater than the increase at the axis site ( Figure 3Bii ) . Furthermore , the time dependent increase in the hotspot associated Rec114 exhibited Rec1148A>Rec114>Rec1148D ( Figure 3Bvi ) . Similar to arguments of the previous section , the following prediction was tested: If more Rec1148A bound to DSB sites than Rec1148D , peaks of the ratio of the profiles Rec1148A/Rec1148D ( 8A/8D ) should map to DSB sites . Analysis shows that the majority of DSB-sites coincide with 8A/8D peaks ( Figures S3 B , E ) . Indeed , comparison of the 500 strongest peaks and 500 hottest hotspots revealed a highly significant correlation ( Figure 3C , p<10−37 ) . Interestingly , 8A/WT and WT/8D peaks also exhibit significant correlations with DSB sites ( p<10−19 , 98% confidence interval of a random model plotted ) suggesting the relation: 8A>WT>8D at DSB sites . Inversion of the DSB anti-correlated 8D profile also lead to the observed positive correlation of WT/8D ( Figure 3Cii , ‘1/8D’ red circles ) , albeit with a weaker correlation than the 8A/8D ( p<10−7 ) and WT/8D ratios ( p< . 04 ) , lending solid statistical support to the interpretation Rec1148A>Rec114>Rec1148D at the 500 strongest DSB hotspots . Selecting just 100 strongest sites produced similar significances , while selecting more hotspots ( 3600 ) results in loss of significance , as the effect of 8A becomes insignificant compared to the effect of 1/8D for weak hotspots ( Figure S4 ) . The parallel analysis of mutations with opposite effects on DSB hotspot binding provided an opportunity to unequivocally demonstrate genome-wide associations of Rec114 with DSB sites . In addition , these mutants reveal that interaction between Rec114 and DSB sites are negatively regulated by Tel1/Mec1 phosphorylation of Rec114 . The effects of Rec114 phosphorylation on its steady state protein levels were assessed by Western blot analysis ( Figure 4 ) using the α-Rec114 antibody [17] . In a rec114-8A culture , a reduction in the protein levels , most notable at 5 and 6 hours , was observed ( Figure 4A ) . In rec114-8D , protein persists longer , until the 6 and 8 hour time points . Thus , phosphorylation of Rec114 appears to increase not only its axis-association but also its steady state levels . Ndt80 is a meiosis specific transcription factor required for pachytene exit and resolution of joint molecules ( JMs ) . Some meiotic DSBs persist in an ndt80Δ background , suggesting its involvement in curtailment of break formation [38] , [39] , or a failure to repair some DSBs . To determine whether Ndt80 affected the stability of Rec114 , we repeated the same Western blot analysis in an ndt80Δ background . Results revealed that Rec114 becomes stabilized in a REC114 ndt80Δ strain for at least 12 hours after transfer to SPM ( Figure 4B ) , while it rapidly declined in NDT80 after 5 hours ( Figure 4A ) . Thus , timely Rec114 degradation requires Ndt80 . ndt80Δ also prevented the degradation of Rec1148A and Rec1148D ( Figure 4B ) , suggesting that the observed differences in steady state protein levels in the mutants ( Figure 4A ) might be caused by differential timing of Ndt80 activation . All Spo11-accessory proteins examined to date , including Rec114 , are recruited to the chromosomes before the initiation of meiotic recombination , and remain chromosome-associated until Zip1 dependent homolog synapsis takes place [17] , [20] , [22] , [40] . Zip1 is an evolutionarily conserved component of the central region of the synaptonemal complex ( SC ) , and is required for homolog synapsis and meiotic recombination [41]–[43] . In early meiotic prophase , there is little overlap between Rec114 and Zip1; at later stages , Rec114 foci become less abundant and dimmer in synapsed chromosome regions but remain bright in unsynapsed regions of the same nucleus [17] , [22] . These observations suggest that synapsis might promote the removal of Rec114 and its associated proteins Mei4 and Mer2 . Combining this with the current observation that the extent of Rec114-axis association is affected by its phosphorylation status ( Figure 3B ) raised the possibility that Rec114 phosphorylation might affect the timing of synapsis . To address this , we performed co-immunostaining analyses of Rec114 and Zip1 using polyclonal antibodies raised against each protein ( Supplementary Online Information ) . The experiment was conducted in an ndt80Δ background to exclude any influence by the NDT80 dependent Rec114 degradation ( above ) . Rec114 in the ndt80Δ background behaved as reported [17] , [22] , with Rec114 foci peaking at mid prophase just before the onset of synapsis , with little or no overlap between Rec114 and Zip1 staining ( Figure 5A , C ) . The fraction of nuclei containing Rec1148A -foci decline more rapidly than Rec114 , while that of Rec1148D containing nuclei remain abundant until at least 6 hours in SPM ( Figure 5Di , ii ) , consistent with synapsis being affected by the status of Rec114 phosphorylation ( Figure 5D iii ) . These observations show that synapsis-associated dissociation of Rec114 is Ndt80 independent . Depending on the Rec114 allele and the associated DSB frequency , synapsis occurs earlier or later , entailing earlier or later Ndt80 independent Rec114 removal . In ∼30% of ndt80Δ cells , some strong Rec114 foci persisted up to 6 hours into meiosis ( Figure S6 ) , consistent with the stabilization of the protein in ndt80Δ ( Figure 4 ) . Most of the Rec114 positive ndt80Δ cells exhibited dimming and/or disappearance of signal along SCs and a persistent polycomplex ( PC ) ( Figure S6i , ii ) , in agreement with ‘SC-decay’ in ndt80 mutants [38] . All prominent Rec114 foci were on Zip1-free areas or in PC ( Figure S6i , iii ) , suggesting that they might be aggregates of stripped Rec114 that cannot be degraded in an ndt80 background . The abundance of residual Rec114 present at these late time points is consistent with the aberrantly late DSBs observed in ndt80Δ [39] .
Three independent studies have implicated a role of the ATM kinases in down-regulating Spo11 catalysis [8] , [9] , [11] . The evidence presented here implicates Rec114 as a physiologically relevant Mec1 and/or Tel1 target in this regulation . Results show a robust reduction of function for phospho-mimicking Rec1148D and a subtler increase in function in the rad50S background for Rec1148A . Why are these effects not symmetric ? One trivial explanation could be that introducing eight aspartic acid residues into Rec1148D may render the protein partially non-functional , aside from mimicking phosphorylation . However , Rec1148D is a stable protein that binds well to chromatin , excluding general protein stability- , nuclear import- or chromatin-binding defect for Rec1148D . Furthermore , since Rec1148D behaves similar to Rec1142D ( T175D , S187D ) in terms of reduction of DSBs , inferred based on reduced spore viability in a spo11-HA background ( Table 1 ) , and two aspartic acid exchanges are less likely to strongly damage the protein , we favor an interpretation involving constitutive phospho-mimicking as explanation . If so , phosphorylation is sufficient to tune down DSB formation ( e . g . Rec1148D or Rec1142D ) , while other effects might prevent the observation of a strong increase in break levels under constitutive “on” conditions ( e . g . Rec1148A ) . Several models ( e . g . [14] , [17] , [49] ) propose that a first negative feedback may be locally restricted to the activated DSB-machine and its surrounding chromatin loops . Phosphorylation of Rec114 would be ideally suited to mediate such a control . However , repeated cleavage of the already broken chromatid is not expected to lead to an increase of the DSB signal . Cleavage of hotspots on the intact sister chromatid could be responsible for the 20–30% increase observed by the ChIP-chip analysis in the rad50S background . Increased DSB formation in Rec1148A , even if only moderate , identifies Rec114 as a rate limiting target of negative feedback at least in the com1Δ/sae2Δ or rad50s background . On the other hand DSB formation is strongly impeded in Rec1148D ( or Rec1142D ) , suggesting that phosphorylation affects a critical function of Rec114 . Importantly , phosphomimicking Rec1148D shows a reduced interaction with DSB-hotspots suggesting a plausible mechanism explaining its reduced activity . Budding yeast Rec114 physically interacts with Mei4 and Mer2 , two other components of the DSB-machinery , to form a complex referred to as RMM ( Rec114-Mei4-Mer2 ) [20] , [22] . RMM foci become abundant in early meiotic prophase and the proteins accumulate on DNA sequences , which organize the chromosome axis upon condensation; when chromosomes synapse , RMM foci become dimmer and eventually disappear [17] , [22] , an observation confirmed here in a strain expressing untagged Rec114 ( Figure 5 ) . The strong correlation between the appearance of Zip1-lines and the disappearance of Rec114 foci from the synapsed regions of the chromosomes suggests that the process of synapsis itself could be removing RMM foci . Synapsis dependent removal of RMM occurs independently of Tel1/Mec1 phosphorylation and Ndt80 ( and thus protein degradation ) , consistent with being an independent mechanism of down regulating DSB levels . During normal meiosis , steady state levels of meiotic DSBs decrease to the background level as cells proceed beyond pachytene with complete cessation of DSBs depending on Ndt80 [39] . The Ndt80 dependent Rec114 degradation reported here ( Figure 4 ) presents a plausible mechanistic explanation for the latter . Normally , Ndt80 activation is coupled to meiotic DSB-repair and synapsis . In response to defects in either process ( e . g . in a rad50S , dmc1Δ or zip1Δ background ) , Tel1/Mec1 prevent Ndt80 activation by hyper-phosphorylating Hop1/Mek1 [50] . Hop1 is an evolutionarily conserved meiotic chromosome axis protein that functions as a meiotic paralog of Rad9 in the Tel1/Mec1 signaling cascade [14] . Hyper-phosphorylation of Hop1 , in turn , activates the checkpoint function of Mek1 , a meiotic chromosome axis associated serine/threonine kinase , and a meiotic paralog of Rad53/Chk2 [14] . Importantly , the Tel1/Mec1-Hop1/Mek1-Ndt80 signaling pathway appears to also regulate normal meiotic progression [12] , [38] , [51]–[53] , raising the possibility that the observed earlier or delayed onset of NDT80 dependent Rec114 turnover in rec114 phospho-mutants might be under Tel1/Mec1 regulation ( Figure 6 ) . The term “meiotic DSB homeostasis” was originally introduced to refer to the phenomenon , whereby the accumulated DSB frequency in a chromosomal region appeared to be maintained at a constant level [54] . Here , we expand the meaning to include that the break frequency might be regulated not only at the regional , but also at genome wide level . A sophisticated system controlling chromosome synapsis and recombination is expected to operate , at least , at two levels: First , the DSB machinery needs to be “informed” about the success of a particular DSB catalysis . This local negative feedback should be limited to the immediate environment and should prevent repeated cleavage of the already broken chromatid near the break . One manifestation of this local down regulation would be DSB interference , or “competitive inactivation” of weaker hotspots , by a nearby strong hotspot [54]–[57] . We show evidence that phosphorylation of Rec114 is a key step in communicating DNA breakage to the DSB machinery via Mec1 and/or Tel1 . Second , nucleus wide ( global ) signaling of successful completion of homolog synapsis and meiotic DSB repair should precede irreversible global inactivation of the DSB machinery . We present evidence for two feedback based mechanisms of such regulation: synapsis dependent removal and Ndt80 dependent degradation of Rec114 . Linking the progression of synapsis to down regulation of DSB formation conveniently ensures that enough DSBs have been formed to guarantee successful homology search . However , synapsis alone does not appear to lead to irreversible inactivation of the DSB machinery . This ultimate decision is instead linked to the exit from pachytene , when the activation of Ndt80 provides some guarantee that early prophase events have successfully completed . Evidence here and elsewhere indicates that cells have the means to prevent excessive DSB formation via negative feedback [8] , [9] , [11] . But what happens when the break level is too low ? A key feature of these nested feedback loops ( Figure 6 ) is that a reduced Spo11 activity , independent of its cause , will delay the completion of synapsis and the onset of Ndt80 activation , and thus provide more time to accumulate breaks , even at a low rate . For instance , delayed synapsis and Ndt80 activation in the DSB poor rec114-8D mutant , would delay RMM inactivation , likely prolonging its active lifespan and raising the level of DSBs eventually produced . Indeed , Rec1148D remained longer and in greater abundance at chromosome axes than Rec114 or Rec1148A . DSB homeostasis could also account for the apparent ‘catching up’ of break levels in a rec114-8D dmc1Δ strain ( Figure 2G ) . For instance , defective recombination and synapsis in the mutant would severely compromise Zip1 dependent RMM removal , while dmc1Δ activation of Mec1-Hop1/Mek1 checkpoint response would prevent Ndt80 dependent Rec114 degradation , thus allowing Rec1148D to remain active . The fact that break levels in rec114-8D rad50S remain low , apparently unable to catch up , suggests that DSB repair beyond rad50S arrest point ( e . g . endonucleolytic removal of Spo11 followed by break resection ) might be required to activate the synapsis and/or Ndt80 based feedback loops ( Figure 6 ) . DSB homeostasis may contribute significantly to the relatively mild effect on spore viability of mutants ( e . g . spo11-hypomorphs ) with a low rate of DSB formation that was up to now solely attributed to CO homeostasis . But clearly , postponing the inactivation of the DSB machinery in response to problems in synapsis and break repair helps to provide more DSBs , on which CO homeostasis can act to ensure correct chromosome segregation .
Standard yeast manipulation procedures and growth media were utilized . All strains are of the SK1 background; relevant genotypes of the strains are listed in Table S1 . The myc13 tag from a REC114-MYC13-HYGRO plasmid ( pNS2 ) was removed to generate pJC15 , an integration plasmid without an epitope tag . Specific [S/T]Q to AQ or DQ mutations were introduced into either pNS2 or pJC15 utilizing the QuickChange Multi Site-Directed Mutagenesis kit ( Stratagene ) . The entire open reading frame ( ORF ) of each allele was sequenced to ensure that the allele did not contain any incidental mutation ( s ) . Each rec114 allele was introduced into a rec114Δ::KanMX4 haploid strain ( RCY336/337 ) , where the endogenous REC114 gene was replaced by a kanamycin resistant gene . Transformants were identified based on their ability to grow on hygromycin plates but not on kanamycin . Southern blot and PCR analyses were performed on candidate colonies to confirm integration of a single copy of a specific rec114-HygroMX4 allele at the endogenous locus , replacing the rec114Δ::KanMX4 allele . Correct rec114 haploid transformants of each allele were taken through standard yeast genetics manipulation to generate corresponding rec114 homozygous diploid strains suitable for meiotic analyses . Three of the eight S/T[Q] consensus sites in Rec114 , T175 , S187 and S256 , were selected for generation of phospho-specific antibodies . T175 and S187 were chosen based on the fact that replacing these residues with a non-phosphorylatable alanine ( A ) confers haploinsufficiency and synthetic interaction with spo11 hypomorphic alleles ( Table 1 ) . S256 was chosen because it was one of the six residues within Rec114 that were predicted to be the most likely ATM/ATR phosphorylation sites ( GPS2 . 1 software [58] ) . Specificity of each phospho-specific antibody was confirmed by Western blot analysis of rec114 strains , each expressing a rec114 allele missing a specific phosphorylation site ( s ) . Induction of synchronous meiosis is carried out according to the established protocols [17] , [59] . All pre-growth and meiotic time courses were carried out at 30°C except for mec1-4ts tel1Δ sml1Δ meiosis , where the culture was kept at 23°C and shifted to 30°C 2 hours after transferring into sporulation medium ( SPM ) . GST-REC114 and GST-rec114-8A plasmid-construction and protein expression were carried out as described [60] . To purify Mec1-myc18 from yeast cells , 500 ml of logarithmically growing cell cultures were subjected to 1 hour incubation with 0 . 1% methyl methanesulfonate ( MMS ) followed by Immunoprecipitation using Goat anti-myc-agarose antibodies ( AbCam ) . Mec1-myc immunoprecipitates were mixed with reaction cocktail containing kinase buffer , cold ATP , and either GST-Rec114 or GST-Rec1148A . The mixtures were incubated at 30°C for 25 minutes and subjected to electrophoresis on SDS gels . Gels were transferred onto a nitrocellulose membrane and subjected to Western blot analyses using anti-Rec114 or phospho-specific antibodies . Whole-cell extracts ( WCE ) were prepared from cell suspensions in 20% trichloroacetic acid ( TCA ) by agitation with glass beads . Precipitated proteins were solubilized in SDS-PAGE sample buffer , and appropriate dilutions were analyzed by SDS-PAGE and Western blotting . Antibodies for Western blotting were mouse monoclonal anti-myc ( 1∶1000 , AbCam ) , rabbit polyclonal anti-Rec114 ( 1∶1000 ) , anti-Phospho-Rec114-S187 , anti-Phospho-Rec114- T175 , anti-Phospho-Rec114- S265 ( 1∶1000 , Cambridge Research Biomedicals ) , goat anti-mouse IgG conjugated to horseradish peroxidase ( 1∶10 , 000; Sigma-Aldrich ) , and donkey anti-rat IgG conjugated to horseradish peroxidase ( 1∶10 , 000; Sigma-Aldrich ) . Southern blot analysis following Pulse Field Gel electrophoresis ( PFGE ) using DNA prepared in agarose plugs or standard agarose gel electrophoresis were performed as described [61] . Exception was that the PFGE gels shown in Figure 2G and Figure S1A were run with the following modifications: initial switch time; 15 sec – final switch time; 32 . 5 sec , in order to better separate large chromosomes . For quantifying the level of DSBs , only the signals associated with breaks proximal to the probe was utilized to maximize the detection of chromosomes that acquired more than one break ( see [3] for discussion ) . Rec114 and Spo11-myc chromatin immunoprecipitation ( ChIP ) , quantitative PCR ( qPCR ) , and microarrays hybridization/analysis were performed as described [17] . Surface spread meiotic chromosomes were prepared as described [14] . Staining was performed as described [14] with the following primary antibodies: rabbit polyclonal anti-Rec1141 ( 1: 100 , F . Klein , MFPL ) , mouse monoclonal anti-HA ( 12CAS , 1∶100 , S . Ley , NIMR ) , mouse monoclonal anti-MYC ( 9E10 , 1∶100 , S . Ley , NIMR goat polyclonal anti-Zip1 ( 1∶50 , SantaCruz Biotechnology ) . Secondary antibodies ( Invitrogen ) were used at a 1∶500 dilution: chicken anti-mouse Alexa-488 , anti-goat Alexa-488 , chicken anti-rabbit Alexa-594 . Chromosomal DNA was stained with 1 ug/ml 4 , 6-diamino-2-phenylimide ( DAPI ) . Images were recorded and analyzed using a Deltavision ( DV3 ) workstation from Applied Precision Inc . with a Photometrics CoolSnap HQ ( 10–20 MHz ) air cooled CCD camera and controlled by Softworx image acquisition and deconvolution software . For comparing proportions ( e . g . matching versus non-matching peaks ) significance values were computed using Fisher's exact test ( http://www . langsrud . com/fisher . htm ) . data was done as described [17] . Briefly , CEL files were converted using Affymetrix's “tiling array software” ( TAS ) and expressed as ChIP relative to WCE ( whole cell extract ) or as ChIP relative to another ChIP . The output of TAS ( intensities per chromosome position ) was smoothed using bandwidths between 250 and 1000 bp ( ksmooth , statistic package R ) and plotted for all chromosomes . For microarrays “decile normalization” was used . It is often a good choice for automatic background correction [17] . For the profiles to be compared in this work , a single correction factor was determined per profile F = 1/ ( 0 . 1 percentile ) . These factors were usually very close to 1 . After multiplying all profile intensities with their correction factors , they were precisely superimposing on all background peaks ( compare Fig . 3A 0–30 kb , and 115–155 kb , see brackets with asterisk ) . Peaks were called automatically as described previously [17] . To quantify the overlap between peaks of one profile and the meiotic hotspot map published by [7] , we used their simplified list of hotspots , organized in 3600 blocks , listing start and end of each hotspot array and the number of mapped 5′-ends detected . We defined the distance of a peak to its nearest hotspot as the distance to the nearest edge of a hotspot block ( even when the peak mapped inside the block ) . Most of these arrays have a very narrow width , ( median 180 bp , mean 252 bp , 0 . 9 quantile 507 bp ) . In order to simulate random distributions , we generated random positions corresponding to the numbers of peaks to be tested and mapped them relative to the hotspot-blocks , the same way as the experimental data . ( Simulations were repeated 100 times to obtain the 2% and 98% percentiles plotted ) . For example , for the comparison of the 500 highest peaks with the 500 strongest hotspots , the hotspots were sorted according to strength , the peaks were sorted according to strength and then the top 500 of each list were compared . For each of the 500 highest peaks , the distance to the nearest hotspot-block was determined and the distances accumulated and plotted . | Meiosis is a specialized cell division that underpins sexual reproduction . It begins with a diploid cell carrying both parental copies of each chromosome , and ends with four haploid cells , each containing only one copy . An essential feature of meiosis is meiotic recombination , during which the programmed generation of DNA double-strand-breaks ( DSBs ) is followed by the production of crossover ( s ) between two parental homologs , which facilitates their correct distribution to daughter nuclei . Failure to generate DSBs leads to errors in homolog disjunction , which produces inviable gametes . Although DSBs are essential for meiosis , each break represents a potentially lethal damage; as such , its formation must be tightly regulated . The evolutionarily conserved ATM/ATR family proteins were implicated in this control; nevertheless , the mechanism by which such control could be implemented remains elusive . Here we demonstrate that Tel1/Mec1 down-regulate meiotic DSB formation by phosphorylating Rec114 , an essential component of the Spo11 complex . We also observed that Rec114 activity can be further down-regulated by its removal from chromosomes and subsequent degradation during later stages in meiosis . Evidence presented here provides an insight into the ways in which the number of meiotic DSBs might be maintained at developmentally programmed level . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"molecular",
"cell",
"biology",
"genetics",
"biology"
] | 2013 | Budding Yeast ATM/ATR Control Meiotic Double-Strand Break (DSB) Levels by Down-Regulating Rec114, an Essential Component of the DSB-machinery |
Chikungunya virus ( CHIKV ) has dispersed in the Americas since 2013 , and its range of distribution has overlapped large forested areas . Herein , we assess vector competence of two sylvatic Neotropical mosquito species , Haemagogus leucocelaenus and Aedes terrens , to evaluate the risk of CHIKV to initiate a sylvatic cycle in the continent . Haemagogus leucocelaenus and Ae . terrens from the state of Rio de Janeiro , Brazil were orally challenged with the two CHIKV lineages circulating in the Americas . Fully engorged females were kept in incubators at 28±1°C and 70±10% humidity and examined at 3 and 7 days after virus exposure . Body ( thorax plus abdomen ) , head and saliva samples were analyzed for respectively determining infection , dissemination and transmission . Both Hg . leucocelaenus and Ae . terrens exhibited high infection and dissemination rates with both CHIKV isolates at 7 dpi , demonstrating that they are susceptible to CHIKV , regardless of the lineage . Remarkably , Hg . leucocelaenus expectorated infectious viral particles as rapidly as 3 days after the infectious blood meal , displaying higher values of transmission rate and efficiency than Ae . terrens . Nevertheless , both species were competent to experimentally transmit both CHIKV genotypes , exhibiting vector competence similar to several American Aedes aegypti . These results point out the high risk for CHIKV to establish a sylvatic transmission cycle in the Americas , which could be a serious health issue as CHIKV would become another zoonotic infection difficult to control in the continent .
Several arboviruses of public health importance such as yellow fever virus ( YFV ) , Chikungunya virus ( CHIKV ) and , more recently , Zika virus have spread from Africa to other continents . Coincidently in their historical cradle in Africa , these arboviruses are transmitted between non-human primates ( NHP ) by Aedes mosquitoes , mostly belonging to the Old World Stegomyia and Diceromyia subgenera [1] . Due to the spread of the anthropophilic mosquito Ae . ( Stg . ) aegypti outside Africa and the transit of viremic people , likely perpetrated by the globalization of trades and travel , these viruses have invaded several continents and are considered a most important alarming public health threat [2] . Owing to the domestic , synanthropic and anthropophilic behavior of Ae . aegypti , these arboviruses exploit urban and periurban ecosystems limiting transmission between Ae . aegypti and humans , originally in Africa and then the secondarily invaded continents such as Asia and the Americas [1 , 2] . To date , the only documented exception is YFV ( Flaviviridae: Flavivirus ) , which after migrating from Africa to the Americas , has spread into the forest in the tropical and subtropical areas . There , the YFV found susceptible NHPs capable of developing high viremia to infect Neotropical wild canopy-dwelling mosquitoes of the Haemagogus and Sabethes genera , initiating a sylvatic cycle [3] . Thereafter , YFV became a zoonotic arbovirus in the Americas as it was originally in Africa . The urban transmission of YFV has been eradicated from the Americas since the first half of the 20th century , but each year YFV has victimized dozens of people in South America due to infections acquired from the sylvatic cycle . As YF vaccine is not adequately supplied and vaccination coverage is insufficient , the risk of YF transmission by the bite of infected wild mosquitoes remains a constant threat [3] . Moreover , as the control of the YFV enzootic circulation would be highly challenging , the enzootic spillover to an urban cycle has more than ever been feared due to great infestation by competent mosquito vectors—Ae . aegypti and Ae . albopictus—colonizing habitats near sylvatic foci [4 , 5 , 6] . Chikungunya virus ( Alphavirus , Togaviridae ) was first isolated in Tanzania in 1952 [7] . Three genotypes sharing a common ancestor in tropical Africa have been described: West African , East Central South African ( ECSA ) and Asian[8 , 9 , 10] . Like the YFV , CHIKV emerged from an enzootic cycle maintained between NHPs and sylvatic African mosquitoes , namely Aedes furcifer , Aedes taylori , Aedes africanus , Aedes luteocephalus and Aedes neoafricanus [11] , establishing rural and urban cycles where Ae . aegypti ensured the inter-human transmission . Viremic people contributed to expand CHIKV territory to Asia , causing epidemics during the 1960’s and , since 2004–2005 , a pandemic covering the Indian Ocean region , Asia , Mediterranean Europe and Central Africa [10 , 12] . These outbreaks were effected by the emerging Indian Ocean lineage ( IOL ) , a monophyletic lineage descendant from the ECSA phylogroup , which contains a mutation in the envelope protein ( E1-Ala226Val ) that enhances viral transmission by the mosquito Ae . albopictus [9 , 10 , 13 , 14] . In the Americas however , autochthonous CHIKV transmission was only described in late 2013 in the Caribbean , the starting point of a large epidemic in the Americas [15 , 16 , 17] . From December 2013 to August 2014 , nearly 660 , 000 cases were reported in the New World , and autochthonous transmission was confirmed to occur in 33 American countries and territories , 27 of which were in the Caribbean , while only French Guiana and Brazil reported CHIKV transmission in continental South America [15] . If in 2014 the CHIKV epidemic was primarily in the Caribbean , in 2015 the virus was identified in multiple countries of Central and South Americas with 30 countries or territories reporting CHIKV cases . In 2015 , Colombia alone recorded 51 . 3% of the 693 , 000 cases in the Americas [18] . In 2016 , South America reported 89 . 2% of the CHIKV cases in the Americas where the virus was detected in 42 countries or territories , including Brazil with nearly 76% of the 347 , 647 suspected cases recorded in the Americas [19] . Remarkably in 2015 and 2016 , the distribution of CHIKV cases has progressively moved towards the inland and the forested areas resulting in a significant overlap with the area of the sylvatic YFV transmission cycle [20] . Thus , it is plausible that CHIKV viremic people infected in the Ae . aegypti urban cycle were bitten by sylvatic primatophilic mosquitoes in the neighboring forests or forest fringe . Since these mosquitoes are competent species transmitting a viral strain capable of being amplified by American NHPs or other vertebrates , CHIKV may initiate a sylvatic cycle as did the YFV in the past . Actually , the tropical American forest is presently more receptive to CHIKV transmission than ever due to the expanding and frequent CHIKV epidemic waves recently reported in zones contiguous to the wild . Therefore , to assess the potential risk for CHIKV to establish a sylvatic transmission cycle in the Americas , we experimentally evaluated vector competence of two sylvatic primatophilic mosquito species , Haemagogus leucocelaenus ( Dyar & Shannon ) and Aedes terrens ( Walker ) for two CHIKV isolates belonging to the lineages currently circulating , ECSA and Asian genotypes [17] .
Female mosquitoes used in this study derived directly from field-collected eggs as Hg . leucocelaenus and Ae . terrens have never been successfully colonized in laboratory [21 , 22] . Eggs were collected in Parque Natural Municipal de Nova Iguaçu ( PNMNI ) , an Atlantic forest conservation area in the State of Rio de Janeiro , Brazil ( 22°46’45”S 43°27’23”W ) , with 20 ovitraps suspended in the forest canopy at a height of 5-16m ( median = 10m ) . Each ovitrap had three wooden paddles that were fortnightly changed from June to November 2016 . The paddles were allowed to dry at room temperature , examined for egg number and stored in an insectary ( 26±1°C; 70±10% RH ) until use . Eggs were hatched by immersing the paddles in dechlorinated tap water for two consecutive days . Larvae were reared in pans ( ~50 larvae/pan measuring 25x25x10cm ) containing 1 liter of dechlorinated tap water , supplemented with yeast powder and shed leaves , renewed every 2–3 days . Adults were morphologically identified [23] , kept in 30×30×30-cm mesh cages maintained in an insectary ( 28±1°C; 80±10% RH; 14h:10h light:dark cycle ) and supplied with both 10% sucrose and honey solutions . Female mosquitoes were challenged with two CHIKV isolates belonging to two distinct lineages: CHIKV 05 . 115 ( CHIKV_115 ) isolated from La Réunion in 2005 belonging to the ECSA lineage [24] , and CHIKV_20235 ( CHIKV_SM ) isolated from Saint-Martin Island in 2013 and belonging to the Asian lineage [16] . They are phylogenetically related to strains circulating in Brazil and other American countries ( 17 , 25–28 ) . CHIKV_115 and CHIKV_SM were isolated from human serum on Ae . albopictus C6/36 and Vero cells respectively , and viral stocks were produced following three passages in the respective cell lineage , then harvested and stored at -80°C until used for the mosquito experimental infection assays [16 , 24] . Virus isolates were provided by the French National Reference Centers for Arbovirus at the Institut Pasteur in Paris and in Marseille . Batches of 60 6–8 day-old female mosquitoes were isolated in feeding boxes and starved for 24 h , then exposed to the infectious blood-meal containing final viral titers of 107 . 5 PFU/mL ( CHIKV_115 ) and 106 . 5 PFU/mL ( CHIKV_SM ) , which correspond exactly to the same titers , passage and stocks used by Vega-Rua et al . [29 , 30] to assess vector competence of American Ae . aegypti and Ae . albopictus populations . The infectious meal consisted of a mixture of two parts of washed rabbit erythrocytes and one part of the viral suspension . Females were fed through a pig-gut membrane and the infectious blood-meal was maintained at 37°C . Mosquito feeding was limited to 1 hour . Only fully engorged females were incubated at 28°C constant temperature , 80% RH and 14h:10h light:dark cycle , with unlimited access to 10% sucrose solution [31] . As expected for sylvatic , not colonized mosquitoes , the artificial blood-feeding rates under experimental conditions here were low ( <10% ) . Thus when available , samples of around 20 mosquitoes of each species were examined at 3 and 7 days after virus exposure , abbreviated as “dpi” . Females were individually processed as follows: abdomen and thorax ( herein after referred to as body ) were examined to estimate viral infection rate , head for viral dissemination and saliva for viral transmission [31] . For the determination of viral infection and viral dissemination rate , each mosquito body and head were respectively ground in 500 μL and 250μL of Leibovitz L15 medium ( Invitrogen ) supplemented with 2% fetal bovine serum ( FBS; Eurobio ) and centrifuged at 10 , 000 x g for 5min at +4°C for further inoculation onto monolayers of Ae . albopictus C6/36 cell ( Institut Pasteur , Paris ) culture in 96-well plates [29 , 31] . After 1 h incubation of homogenates at 28°C , 150 μL of 2 . 4% CMC ( carboxymethyl cellulose ) in Leibovitz L15 medium supplemented with 10% FBS was added per well . After 3 days of incubation at 28°C , cells were fixed with 10% formaldehyde , washed and revealed with hyperimmune ascetic fluid specific to CHIKV as the primary antibody and Alexa Fluor 488 goat anti-mouse IgG as the second antibody ( Life Technologies ) . Presence of viral particles was assessed by detection of focus forming units ( FFU ) . To estimate viral transmission , mosquito saliva was collected in individual pipette tips containing 5 μL FBS for 30 min as previously described [32] . Then , FBS containing mosquito saliva was expelled into 45 μL of Leibovitz L15 medium for titration in Ae . albopictus C6/36 cell culture in 96-well plates and stained as described above . Viral load in saliva was expressed as FFU/saliva . Infection rate ( IR ) refers to the proportion of mosquitoes with infected body among tested females . Disseminated infection rate ( DIR ) corresponds to the proportion of mosquitoes with infected head among the previously detected infected mosquitoes ( i . e , abdomen/thorax positive ) . Transmission rate ( TR ) represents the proportion of mosquitoes with infectious saliva among mosquitoes with disseminated infection . Transmission efficiency ( TE ) represents the proportion of mosquitoes with infectious saliva among the initial number of females tested [29] . We used the Wilcoxon signed rank test to compare the viral load in the saliva . Significant difference was established when p-values were lower than 0 . 05 . Data analyses and graphics were done with PRISM 5 . 0 software ( GraphPad Software , San Diego-CA , USA , 2007 ) . The Institut Pasteur animal facility has been accredited by the French Ministry of Agriculture to perform experiments on live animals in compliance with the French and European regulations on care and protection of laboratory animals ( directive 2010/63/EU ) . This study was approved by the Institutional Animal Care and Use Committee ( IACUC ) at the Institut Pasteur and by the Institutional Ethics Committee on Animal Use ( CEUA-IOC license LW-34/14 ) at the Instituto Oswaldo Cruz . Mosquito collections in the Atlantic forest were approved by local environmental authorities ( PNMNI license 001/14-15; SISBIO-MMA licenses 37362–2 and 012/2016 ) . This study did not involve endangered or protected species .
Both Hg . leucocelaenus and Ae . terrens exhibited high infection and dissemination rates with both CHIKV isolates at 7 dpi , demonstrating that they are CHIKV susceptible , regardless of the lineage ( Fig 1A–1C ) . Indeed , 99 . 7% Hg . leucocelaenus and 85 . 7% Ae . terrens were already infected with the CHIKV_115 at 3 dpi , when 66 . 6 and 60% had disseminated infection , respectively . As expected , viral dissemination increased from 3 dpi to 7dpi ( Fig 1A and 1B ) , exceeding 90% in both species when infected with the ESCA isolate ( CHIK_115 ) . Even with the Asian genotype ( CHIK_SM ) delivered at a lower dose , the two mosquito species also presented high IR ( 94 . 7% in Hg . leucocelaenus and 84 . 6% in Ae . terrens ) and DIR ( 61 . 1% and 81 . 8% , respectively ) ( Fig 1C ) . Most importantly , both Hg . leucocelaenus and Ae . terrens were competent to transmit CHIKV of both lineages circulating in the Americas at 7 dpi . Moreover , Hg . leucocelaenus was able to transmit infectious viral particles as rapidly as 3 dpi ( Fig 1A ) . It usually exhibited higher TE and TR than Ae . terrens regardless of the CHIKV lineage and time of incubation ( Fig 1A–1C ) . Nonetheless , one week after ingesting the infectious meal with the CHIKV_155 , TE and TR varied from 60 to 66 . 6% in Ae . terrens and reached values as high as 69 . 5 to 76 . 2% in Hg . leucocelaenus . When considering the CHIKV_SM delivered at a lower dose to mosquitoes , TR was still high ( 63 . 6% ) in Hg . leucocelaenus while it dropped to 11 . 1% in Ae . terrens at 7 dpi ( Fig 1B and 1C ) . At 7 dpi , the saliva viral load ranged from 0 . 2 to 2 . 5 log10 for Hg . leucocelaenus and from 0 . 2 to 1 . 6 log10 for Ae . terrens when infected with the CHIKV_115 , and from 0 . 2 to 2 . 2 log10 for Hg . leucocelaenus when infected with the CHIKV_SM . When infected with the CHIKV_115 , the saliva viral load ( Fig 1D ) did not differ between species regardless of the incubation period ( p = 1 . 00 and p = 0 . 151 , for 3 and 7 dpi , respectively ) . However , the saliva viral load was higher in Hg . leucocelaenus than Ae . terrens when infected with CHIKV_SM ( p = 0 . 01 ) . When evaluating the same mosquito species , the saliva viral load did not differ between the two CHIKV isolates , but when infected with the CHIKV_115 , the expectorated saliva viral load was higher in Hg . leucocelaenus at 7 dpi than at 3 dpi ( p = 0 . 008 ) .
We demonstrated for the first time that two sylvatic primatophilic mosquito species from the Americas are competent to transmit CHIKV belonging to the two lineages in circulation . High TRs were detected for both Hg . leucocelaenus and Ae . terrens one week after ingesting infectious blood meals containing viral titters close to viremia of CHIKV-infected patients [33] . Importantly , infectious viral particles were detected as rapidly as 3 dpi in the saliva of Hg . leucocelaenus challenged with the ESCA isolate ( CHIK_115 ) which is the major lineage circulating in the current Brazilian epidemics , including the area of Rio de Janeiro where the tested mosquito population was originated [17 , 25 , 26 , 27 , 28] . Moreover , values of vector competence estimated herein for both Hg . leucocelaenus and Ae . terrens were similar to those described for Ae . aegypti and Ae . albopictus populations from tropical Americas and the Caribbean , challenged with the same viral strains and titers by Vega-Rua et al . [29 , 30] . Such a comparison however , should be considered with caution due to the low number of mosquitoes examined , especially for the Ae . terrens . Both Hg . leucocelaenus and Ae . terrens are arboreal , tree hole breeding mosquitoes that can bite not only NHPs in the forest canopy , but also humans at the ground level [34–42] . Besides , both mosquitoes are capable of flying long distances linking isolated patches of the forest and bite people in the open fields , particularly in the ecotone between the wild and the man-modified environment [23 , 43 , 44] . For instance , Hg . leucocelaenus frequently bite humans both in primary and secondary growing forests as well as in the peridomicile located far from forests [45 , 46] . All together , these patterns of behavior favor the zoonotic transmission of arboviruses by these mosquitoes to humans and vice-versa . Indeed , Hg . leucocelaenus has been proven to play an important role in the transmission of sylvatic YFV as well as other human-infecting arboviruses across the Americas [36 , 39 , 47 , 48] . This mosquito has been incriminated as the primary vector in the recent YFV epizooties and epidemics reported in Southern and Southeastern Brazil [49 , 50] . Notably , both Hg . leucocelaenus and Ae . terrens have a large geographical distribution in the New World . Ae . terrens has been reported from Mexico to northern Argentina ( Argentina , Bolivia , Brazil , Colombia , Costa Rica , Ecuador , French Guiana , Guatemala , Guyana , Mexico , Panama , Paraguay , Suriname and Venezuela ) , and Hg . leucocelaenus from Panama to northern Argentina and Uruguay ( Argentina , Bolivia , Brazil , Colombia , French Guiana , Guyana , Panama , Paraguay , Peru , Suriname , Trinidad and Tobago , Uruguay and Venezuela ) [35 , 36 , 51] . The ability to experimentally transmit the two lineages of CHIKV circulating in the Americas described herein for Hg . leucocelaenus and Ae . terrens might become a pivotal factor facilitating the establishment of CHIKV in a zoonotic cycle in the Americas . The current CHIKV outbreaks in the continental tropical Americas covering a large geographical range of competent sylvatic mosquitoes enhance the chances of CHIKV to approach the forests . Indeed , the transit of CHIKV viremic people combined with the vector competence of Hg . leucocelaenus and Ae . terrens , the large distribution of vectors and their biological and behavioural features suggests potential for CHIKV to become zoonotic in the continent . Additionally , high densities of Ae . albopictus populations , experimentally CHIKV transmission competent , in the transition zone between urban and forest environments possibly favor this phenomenon as well [29 , 52] . However , the establishment of a CHIKV sylvatic transmission cycle in the Americas depends upon the susceptibility of local NHPs to this virus , which has not been assessed yet . Wild NHPs were detected naturally infected with CHIKV in Malaysia , which suggested that a sylvatic , zoonotic transmission cycle could also occur in Asia [53] . Therefore , if Neotropical NHPs can amplify CHIKV and produce sufficient viremia to infect mosquitoes , the establishment of a sylvatic CHIKV cycle in the Americas could occur . The establishment of a CHIKV sylvatic transmission cycle in the New World would have immediate public health consequences as , so far , there are no efficient methods to control the enzootic circulation of any arboviruses . As learned from the case of YFV , the control of a potential zoonotic transmission of CHIKV in the Americas would be extremely challenging . An effective fight against tree hole harboring , canopy feeding mosquito vectors is not feasible , and there is still neither antiviral treatment nor a licensed vaccine to prevent infection in the case of CHIKV [54] . Thus , surveillance programs need to be organized in the continent to determine whether CHIKV has initiated a sustainable zoonotic transmission , which should search for natural infections in NHPs and enzootic vectors , together with the investigation of neutralizing antibodies in NHPs and other sentinel vertebrates as well as people living near forests , especially in Stegomyia-free sites . Besides , mitigating CHIKV epidemics in suburban and rural areas intimately linked to the forested habitats is crucial to prevent virus establishment in the wild in the tropical Americas , if it is not already too late . | Chikungunya is a mosquito-borne-viral disease of African origin that has spread in the Americas since its first detection in 2013 . The vector of Chikungunya virus ( CHIKV ) in the Americas is the mosquito Aedes aegypti . Due to this vector domestic behavior , CHIKV transmission is limited between this mosquito species and humans in urban and suburban American areas . However , since 2015 the distribution of CHIKV has moved towards the inland and the forested areas in the tropical Americas . The recent reports of CHIKV epidemic waves in zones intimately linked to the wild exemplify the potential of CHIKV initiating a zoonotic cycle in the continent since local sylvatic mosquitoes can be infected and transmit the virus . We experimentally demonstrated that two widely distributed American sylvatic primatophilic mosquito species , Haemagogus leucocelaenus and Aedes terrens , are highly susceptible and competent to transmit the two CHIKV lineages currently circulating in the continent , 7 days after an infectious blood meal , Hg . leucocelaenus mosquitoes presenting infectious viral particles in their saliva as rapidly as 3 days exposure . We concluded that there is a definite risk for CHIKV to establish a sylvatic cycle in the tropical Americas if local non-human primates can amplify the virus to infect wild primatophilic mosquitoes . | [
"Abstract",
"Introduction",
"Materials",
"and",
"methods",
"Results",
"Discussion"
] | [
"invertebrates",
"medicine",
"and",
"health",
"sciences",
"ecology",
"and",
"environmental",
"sciences",
"body",
"fluids",
"pathology",
"and",
"laboratory",
"medicine",
"togaviruses",
"chikungunya",
"infection",
"viral",
"transmission",
"and",
"infection",
"pathogens",
... | 2017 | High risk for chikungunya virus to initiate an enzootic sylvatic cycle in the tropical Americas |
Dosage compensation equates between the sexes the gene dose of sex chromosomes that carry substantially different gene content . In Drosophila , the single male X chromosome is hypertranscribed by approximately two-fold to effect this correction . The key genes are male lethal and appear not to be required in females , or affect their viability . Here , we show these male lethals do in fact have a role in females , and they participate in the very process which will eventually shut down their function—female determination . We find the male dosage compensation complex is required for upregulating transcription of the sex determination master switch , Sex-lethal , an X-linked gene which is specifically activated in females in response to their two X chromosomes . The levels of some X-linked genes are also affected , and some of these genes are used in the process of counting the number of X chromosomes early in development . Our data suggest that before the female state is set , the ground state is male and female X chromosome expression is elevated . Females thus utilize the male dosage compensation process to amplify the signal which determines their fate .
When the sex chromosomes carry substantially different gene numbers , dosage compensation is necessary to equalize gene expression between the two sexes . In the three best studied model systems Drosophila , C . elegans and mammals where males are XY and females XX , this involves targeting X-specific components which modify the chromatin and transcription of X-linked genes . In each of these cases the end result is different; Drosophila upregulates transcription of the male X by about two-fold , C . elegans downregulates transcription of both X chromosomes in the hermaphrodite by approximately half , and mammals generally shut down transcription of one of the two female X chromosomes ( reviewed in [1] ) . As it is the Drosophila male which requires dosage compensation , mutation of genes strictly dedicated to this process results in male lethality . The first male specific lethal identified , maleless ( mle; [2] ) , is indeed involved in dosage compensation as are the next identified male lethals , msl-1 and msl-2 [3] . msl-3 identified by Uchida et al . [4] and males absent on the first ( mof; [5] ) complete the proteins collectively known as the male specific lethals ( msls; reviewed in [1] , [6] , [7] ) . In addition to these proteins , two RNAs on the X chromosome ( the roX RNAs ) , which are not present in females , are also essential for dosage compensation [8] . Although roX1 and roX2 show no sequence similarity and do not have an open reading frame that could encode a significantly sized protein , they function redundantly; either roX is adequate for function , while loss of both RNAs is required for a failure in dosage compensation and male lethality [9] . The MSL proteins and roX RNAs function as a complex , coating the male X chromosome; the X chromosome is hypertranscribed and MOF acetylates histone H4 on lysine 16 . Finally , a protein that appears to be part of the dosage compensation complex ( DCC ) but is required by both sexes is the JIL histone H3 kinase . JIL is also enriched on the male X chromosome but its loss leads to lethality in both sexes [10] . In 1980 , Skripsky and Lucchessi [11] reported that females heterozygous for a Sex-lethal ( Sxl ) null allele , Sxlf1 , and homozygous for mle showed morphological characteristics indicative of sex transformations . Sxl is the Drosophila sex determination master switch , which is on in females but off in males . The Sxlf1/+; mle/mle sex transformation result was confirmed and extended by Uenoyama et al . [12] who observed similar effects with two different mle alleles as well as msl-2 and msl-3 . This argued that this phenomenon was not unique to mle , but likely a general property of the msls . These results , a requirement of male specific genes in females , present a paradox . First , homozygous msl− females show no sex transformations and are fully viable [3] . Second , besides controlling differentiation , a key function of Sxl is to turn off the male dosage compensation system to prevent hypertranscription of the two female X chromosomes , which would otherwise lead to female lethality . As a splicing and translation regulator , Sxl alters the splicing and inhibits translation of msl-2 mRNA so preventing assembly of the DCC [13]–[17] . The absence of MSL-2 also destabilizes MSL-1 and MSL-3 assuring inactivation of the dosage compensation machinery . The initial activation of Sxl is transcriptional , at the Sxl ‘establishment’ promoter , SxlPe [18] . In cycle 12 of embryogenesis , SxlPe responds to activating X linked genes ( known members: sisterless-a ( sis-a ) , sisterless-b ( sis-b ) , runt ( run ) and unpaired ( upd ) ) , in conjunction with positive maternal factors such as Daughterless , balancing their dose against the negative effect of genes on the autosomes ( deadpan ( dpn ) , the only identified member ) and maternal factors such as Groucho ( Gro ) and Extramacrochetae ( hereafter collectively referred to as the X∶A ratio; reviewed in [19] ) . Protein from SxlPe transcripts alters the splicing of transcripts from the ‘maintenance’ promoter , SxlPm , first transcribed in cycle 14 in both sexes . In the absence of Sxl protein , default splicing includes a translation terminating exon into the transcripts from SxlPm . As male embryos do not activate SxlPe , Sxl protein is absent and a splicing change on SxlPm transcripts is only effected in females . Females thus set in motion a splicing autoregulatory feedback loop which serves to maintain Sxl expression , and sexual identity , through the rest of the life cycle [20] . Returning to the paradox of a female requirement of male specific genes , one explanation is that XX embryos with only one copy of Sxl fail to reliably activate the gene . These XX cells would be male and are presumably eliminated , due to the gene imbalance from inappropriate dosage compensation . However , when one or more of the msls is mutant , these masculinized XX cells might survive since assembly of the male DCC is prevented . The resulting clones grow but are sexually transformed , so accounting for the observed sex transformations .
The above viability results prompted us to analyze whether key activators of Sxl - the numerator genes sis-a and sis-b , would show a similar interaction with the msls . Figure 1B shows that the effect of a sis-a , sis-b double mutant chromosome is more extreme than a Sxl null , when crossed to mothers mutant for each of the msls . The greater effect of the sis genes is not surprising , given they function in a dose sensitive process to activate Sxl and so determine female sex . What is surprising is that the msls interact with the numerators to promote female viability . To test whether the loss of a single numerator gene could also affect females , we performed crosses with reduced dose of either sis-a or sis-b . Since msl-3 showed the strongest overall interaction , this msl was examined . Figure 2A shows that sis-a as well as sis-b alone affected females , with sis-b having the stronger effect . The sis gene interactions suggest that very early steps in the female sex determination process are compromised . Testing two Sxl alleles , an early ( Sxlf9 ) versus a late ( SxlM1 , f12 ) defective allele , indicated that the early defective allele had an effect , almost as strong as sis-a alone , while the late defective allele did not . These data are consistent with the view that early , dose sensitive events in female sex determination are influenced by the msls . The late Sxl transcripts may not turnover or be as dose sensitive as the early transcripts , so a 50% reduction may not be sufficient to sensitize the females . sis-a and sis-b are zygotic in their role in female sex determination . To determine whether the effect observed with the msls was maternal or zygotic , reciprocal crosses to Figure 1B were performed . Under these conditions , halving the dose of each of the four msls , including msl-2 , reduced female viability ( Figure 2B ) . The zygotic effect was generally weaker than the maternal . A maternal effect of msl-2 is surprising given that the protein is not detected in females [13] , [15] , [17] . We note that a maternal effect of msl-2 was also described by Uenoyama et al . [12] . msl-2 RNA is deposited into the egg ( Flybase microarray data; http://www . flybase . org/ ) , so the strength of the zygotic effect is presumably influenced by the amount of maternal protein/RNA of each of the msls . As the msls , particularly MSL-3 and MOF , have been shown to bind to both autosomal and X-linked genes where they might perform an unknown role , we wondered whether the entire male DCC , including MOF and the roX RNAs , influenced female viability . With the numerator gene dose compromised , halving mof dose had an effect , as did roX1 which was much stronger in effect than roX2 ( Figure 2B ) . Since the roX RNAs function redundantly , the impact of roX1 and the weaker interaction of roX2 can be explained by the fact that first expression during embryogenesis is later for roX2than for roX1 [9] . Combined , these results indicate that the msls affect an early event and that the entire male DCC is required for promoting female viability . The foregoing suggests an event early in Sxl expression is altered by the DCC . To directly assess the effect of the DCC on Sxl transcription , in situs were performed with Sxl probes specific for either the early or late transcripts . Embryos from homozygous mutant mle1 , msl-1L-60 , msl-21 , msl-31 or roX1ex6 , roX2− double mutant mothers , mated to heterozygous msl males were analyzed . For the roX1 , roX2 double mutant embryos , the roX1− , roX2− males have a duplication of roX2+ on their Y chromosome so only the females are roX1− , roX2− . In wild type embryos , SxlPe is not activated until cycle 12 , its expression becomes stronger in cycles 13 and 14 before it rapidly ceases expression early in cycle 14 . For all the msls about half the embryos showed weaker than normal expression of SxlPe , as judged by the size and intensity of the in situ dots on their X chromosomes ( Figure 3 ) . The fraction was higher in the roX1− , roX2− cross where all the females are expected to be mutant . These data indicate that the entire DCC complex is used to upregulate transcription from SxlPe . If , as the data suggest , the primary reason for female lethality is the failure to activate Sxl , a constitutive allele ( such as SxlM1 ) which bypasses the X∶A ratio should rescue them . Since msl-3 showed the strongest interaction in the genetic tests , we determined whether the presence of SxlM1 could rescue the lethality of sis-a , b or Sxl dose reduction in embryos from mothers homozygous for msl-31 . The rescue ( Figure 4 ) of 72 . 8% and 98 . 7% of the females by SxlM1 for sis-a , b or Sxl dose reduction , respectively , demonstrates that female lethality is primarily caused by the inadequate expression of Sxl . We next examined whether transcripts from the maintenance promoter , SxlPm , were affected . As shown in Figure 5 , this promoter was also affected by loss of DCC components . For msl-1 , msl-2 and msl-3 about 50% of the embryos , presumably the homozygotes , showed weaker expression . For mle and the roX1− , roX2− double mutants almost all the females ( 2-dots/cell embryos ) showed weaker than normal expression . As noted for SxlPe , most of the females are mutant for roX1− , roX2− ( excepting the few non-disjunction embryos that also receive the Y with a duplicated roX2+ gene ) , however , only 50% of the embryos are homozygous for mle1 . The mle1 data suggest the maternal contribution of MLE is important for SxlPm expression , an effect that appears distinct from the loss of the DCC since for SxlPe only half the embryos were affected . This may be an outcome of SxlPm relying more heavily on maternal MLE compared to the other msls . Alternatively , because MLE also affects the stability of roX1 RNA , which has a larger role in sex determination than roX2 ( Figure 2B ) , the effect of mutating MLE may be amplified as it not only eliminates the maternal MLE but also reduces the levels of roX1 RNA , acting as a double mutation . Despite this unexplained effect on SxlPm by maternal MLE , the data together indicate that both Sxl promoters are susceptible to the DCC and suggest that Sxl , which resides on the X , is a dosage compensated gene . Consistent with the idea that transcription elongation and not initiation is altered by the DCC [21] , transcription from both Sxl promoters , which are regulated by different factors , is affected . Although the in situs of Ore R embryos did not show the distinctly different classes we observed with the msl embryos , to control for the possibility that the quality of the in situs was responsible for generating a poor signal in half the embryos from msl mutant mothers , in situs for SxlPm transcripts were simultaneously performed with a distinguishable second probe - the segmentation gene hairy which has a striped pattern of expression . As seen in Figure 6 , embryos from msl mutant mothers that are at the same developmental stage as Ore R embryos have comparable hairy stripes but poor SxlPm signal , indicating that the poor signal is not an artifact of the in situs but an effect of the msls on Sxl transcription . The in situs are qualitative and the nuclear dots detect transcription directly off the chromosomes , indicating only high levels of transcription . For a better measure , we performed quantitative RT-PCR analysis on 2–3 h and 2 . 5–3 . 5 h embryos for SxlPe and SxlPm expression , respectively . Embryos were from homozygous mutant mothers for msl-21 , msl-31 or roX1ex6 , roX2− double mutants , mated to heterozygous msl males . As for the in situs , in the roX1− , roX2− embryos only the females are doubly mutant as males have a duplication of roX2+ on their Y . RNA levels were normalized to tubulin levels and compared to Ore R embryos which were set to 1 . Figure 7 shows that SxlPe is expressed at lower levels than Ore R embryos in all three msl genotypes . The median for msl-21 embryos was slightly above , for msl-31 and roX1− , roX2− embryos the median slightly below half of Ore R . The medians for SxlPm were also close to half , except for the roX1− , roX2− genotype which was closer to 0 . 7 . SxlPm is transcribed in both sexes and all the males have a functional roX2 gene in the roX1− , roX2− embryos . In these embryos , SxlPm gave a value of 0 . 7 , suggesting males are transcribing SxlPm at close to normal levels while the females express SxlPm at close to half . This would suggest that functional roX2 RNA is present mid-way through cycle 14 , a little earlier than in situs can detect [9] . A value close to 0 . 5 for both promoters ( excepting SxlPm for roX1− , roX2− ) was a little surprising given the in situ results which show about half the embryos have close to normal levels of transcription . It suggests that the DCC may be upregulating the expression of Sxl by a little more than two-fold , not unlike the roX genes [22] . Alternatively , and not mutually exclusive , it may also indicate that at 2–3 h of development most of the DCC is assembled primarily from maternal reserves and the presence of one wild type chromosome in half the embryos ( from the heterozygous fathers ) makes a small contribution . With respect to SxlPe , the qRT-PCRs score embryos whose average age is slightly younger than the in situs , at cycles 13 and 14 ( 2 . 75–3 . 25 h ) . Close examination of those in situs shows few embryos in early cycle 13 with uniform , wild type levels of SxlPe expression . However , when the membranes begin to drop between the nuclei later in cycle 13 , the class with more uniform expression resembling wild type , is more readily observed ( data not shown ) . By late cycle 13 and cycle 14 , the zygotic contribution of the wild type chromosome from the heterozygous fathers must begin , and the two different classes are more readily apparent in cycle 14 embryos ( Figure 3 ) . For SxlPm , the data are more consistent with the DCC having slightly greater than a two-fold effect . Effects of the DCC were also scored for some of the sex determination genes in the 2–3 h collections from homozygous msl-21 or msl-31 mothers . msl-31 embryos show sis-a , like Sxl , with a median expression close to 0 . 5 , while run and dpn gave medians close to 1 as expected for non-dosage compensated genes ( sis-b could not be reliably scored as it has an anti-sense transcript , CG32816 ) . upd appeared reduced to ∼0 . 7 but this was not statistically significant and the data showed greater variability than for the other genes . This may be because upd begins expression later ( cyc 13; [23] ) , and half the embryos are beginning to perform normal dosage compensation . Also , there are 2 DCC high-affinity sites ( see Discussion below ) relatively close to upd . These are predicted to make upd less sensitive to the loss of MSL-3 , since MSL-3 is required for spreading of the DCC from its initial entry sites . For msl-21 embryos , the upd median did drop to ∼0 . 5 , consistent with the loss of MSL-2 having a greater effect than MSL-3 for genes with close DCC entry sites . run did not show a significant change from wild type . However , unlike for the msl-31 embryos , sis-a was slightly elevated relative to wild type , while dpn mRNA , at a low level of significance , showed a small decrease . As the msl-31 embryos show that sis-a is dependent on the DCC , these latter data suggest that besides dosage compensation , MSL-2 may have an additional role , one that perhaps affects mRNA stability . MSL-2 affects the steady state levels of the roX RNAs [24]; such an activity could explain the greater variability in the values we measured for msl-21 embryos . To test this , in situs of sis-a mRNA were performed to determine if over time , the mRNA levels would show a change consistent with accumulation . Indeed , we found this to be the case ( Figure S1 ) , suggesting that in the case of sis-a MSL-2 may serve to destabilize its RNA . During the early cycles , embryos from msl-21 mothers had signal which was generally weaker than wild type , but by cycle 12 when the message has its highest accumulation in wild type [25] , the accumulated levels in the msl-21 embryos were even higher . While alternative explanations , e . g . repression of the sis-a promoter by MSL-2 are also plausible , this effect would have to occur at some but not all stages of sis-a transcription and be independent of the DCC , as loss of MSL-3 shows the predicted 2-fold drop in sis-a mRNA levels . Despite the suggestion of an additional role beyond dosage compensation for MSL-2 , the qRT-PCR data show that the 2 Sxl promoters are expressed at approximately half their normal levels by the loss of the DCC . Expression of other X-linked genes also appears to be similarly affected , very clearly evident in the msl-31 embryos . This indicates the DCC functions relatively early , and may also affect the handful of genes known to be expressed during these early stages of embryonic development [26] , [27] . The data argue for a role of the male DCC in females , a function not ascribed to it , and the complex has not been detected in female embryos [28]–[30] . Our data suggest that prior to the full activation of Sxl there is a brief window of male dosage compensation in females , after which Sxl protein is predicted to shut down MSL-2 expression , and destabilize the entire DCC . Not all anti-MSL antibodies have been reported to detect the complex at this early stage , even in males ( see [30] ) . Given this limitation , we used an anti-MSL-1 antibody from the Lucchesi lab which has high sensitivity and enhanced the signal with an M3TAP construct [31] . These embryos were co-stained with anti-Sxl antibodies and closely examined around the cellular blastoderm stage . Figure 8 shows that there is indeed a very brief stage , in mid cycle 14 , when it is possible to simultaneously detect both Sxl and the DCC in females . The ant-MSL-1 signal in the female nuclei is not as bright and generally covers an area of DNA larger than in males , presumably the two X chromosomes .
Our data indicate that some of the earliest expressed genes on the X , the numerators as well as Sxl , are dosage compensated . Dosage compensation is a chromosome wide phenomenon , and , at the least , the effect of the msls can be detected as early as cycle 13 . Previous work timed the DCC in males at cellular blastoderm ( Stage 5 , [30] ) and early gastrulation ( Stage 6 , [29] ) . Our data ( qRT-PCRs and in situs ) suggest dosage compensation sets in earlier , by 2–3 h in development and appears to initially rely on maternal stores and the zygotic expression of the roX RNAs ( roX1 primarily ) . As discussed by McDowell et al . [30] antibody sensitivity sets the limit for the prior studies . The present studies relied on different assays , which may account for the difference . Indeed , we were also unable to detect convincing signal in males , which is normally stronger , much before blastoderm by antibody staining ( Figure 8 ) . It is also possible that early in development the DCC is harder to detect directly as there are fewer genes being transcribed , so less of the complex may have assembled onto the X chromosome before cycle 14 , when the mid-blastula transition occurs and there is a large transcriptional burst . The zygotic expression of roX1 RNA has been placed at around 2 h of embryogenesis [9] , consistent with the effects we observe . Targeting of the DCC to the X chromosome , rather than the autosomes , is thought to rely on transcription marks , sequence elements ( ∼150 MREs – MSL recognition elements and ∼130 HAS – high affinity sites ) , and other unknown elements [35] , [36] . The identified sequence element set is still incomplete since the two data sets show an overlap of 69%; it is predicted that the X chromosome may have as many as 240–300 elements ( reviewed in [7] ) . Examination of the published MRE and HAS shows the closest element to Sxl ∼139 Kbp 5′ of the gene . This distance is on the large side , although it should be noted that all elements which target the DCC to genes on the X remain to be identified; as an example , the white gene has its closest known MRE/HAS 93 Kbp away but its mini form in transgenes , which does not include this site , is clearly dosage compensated . Finally , ChIP data ( modENCODE , Flybase ) show Sxl with strong H4K16 acetylation marks , a modification dependent on the DCC . ChIP data for JIL-1 kinase also suggest the DCC is at Sxl . None of the other sex determination genes , other than upd ( two 3′ elements at ∼5 . 6 and 6 . 8 Kbp away ) had an element within 10 Kbp ( sis-a ∼26 Kbp , sis-b ∼38 Kbp ) , consistent with the observation that the msls involved in spreading the DCC from its entry sites on the X ( MSL-3 , MOF and the roX RNAs ) , are required for their elevated expression . upd , the exception , showed greater sensitivity to the loss of MSL-2 than MSL-3 , as might be expected for dosage compensation which is less dependent on spreading . An interesting correlation is that run which is not compensated , had its closest elements ∼343 ( 5′ ) and ∼273 Kbp ( 3′ ) away , further than the rest of the other known key sex determination genes . By using the DCC before the female state is established , Sxl capitalizes on the default male state . Transcription from SxlPe is amplified , an effect unique to females as males do not transcribe from SxlPe . Determination of female identity is thus consolidated . As expression of Sxl protein levels is established , Sxl protein subsequently shuts down the DCC and eliminates the very difference in gene dose between the sexes which set in motion , as well as augmented , its own activation . Implicit , is that before the establishment of Sxl expression , each X-chromosome in females is transcribed at 2X levels , as in males , and our qRT-PCR data of some of the key sex determination genes would support this view . The conventional X∶ A ratio would then be 4∶2 rather than 2∶2 , and in males 2∶2 rather than 1∶2 ( Figure 9 ) . In that there is a 2-fold difference between the sexes , this scenario is mathematically the same . However , there are practical and functional implications . An X∶ A ratio that is transiently 4∶2 rather than 2∶2 in females , would have some of the X-linked genes which activate SxlPe at twice the level of their putative counteracting autosomal or denominator genes . In a screen which seeks suppression of a female-specific lethal condition due to a decrease in numerator elements , it would require the equivalent of two autosomal genes to be mutated to reestablish an X∶ A ratio favorable for female survival . Obtaining two mutations in genes functioning in the same process at once is unlikely , which may have skewed the outcome of screens which sought to identify zygotic autosomal genes . It may not be a coincidence that the only autosomal acting component identified is dpn [37] , [38] . As both Dpn and Run bind the co-repressor Gro [39] , [40] but have opposing effects on SxlPe , it has been speculated they may antagonize each other [39] , [41] . Screens may have repeatedly identified dpn as it would be counteracting a gene expressed at its chromosomal equivalent , since run is not upregulated by the male DCC . On a more general level , our data suggest an upregulation of transcription of the Drosophila X , and may reflect a universal requirement of elevated X chromosome expression to avoid monosomy . Recent microarray analysis of mouse ES cells indicates that mammalian dosage compensation is more complex than previously thought: there is higher expression of the X chromosome relative to the autosomes giving them equivalence , i . e . chromosome per chromosome the X is overexpressed by about two-fold relative to each autosome [42]–[44] . As differentiation proceeds , females lose expression of one of their X chromosomes , silencing it through inactivation . Put in other words , the mammalian X chromosome is not monosomic in expression but rather is hyperactive , and the process of dosage compensation appears to shut down elevated X chromosome transcription in females . ( Hyperactive X chromosome expression in C . elegans has also been suggested [42] , so dosage compensation in the hermaphrodite would then serve to lower the X chromosomes to match autosome levels ) . In this regard , Drosophila would not be very different from mammals except that rather than inactivating one of the female X's , Sxl inactivates the mechanism which upregulates X chromosome specific expression . In all cases , dosage compensation avoids tetrasomy of the X . What the components are which specifically upregulate the mammalian or C . elegans X chromosome—the Drosophila male DCC counterpart—remain to be determined .
Flies were reared under uncrowded conditions on standard cornmeal medium . All crosses were done at 25°C; Ore R was the wild type control . Progeny were counted out to 8 days from the first day of eclosion . Description of genes can be found in Flybase ( http://www . flybase . org/ ) . These were done as in Erickson and Cline [25] . The Sxl early ( 407 nt ) and late ( 1039 nt ) transcript specific probes were generated by the primers , respectively: 5′ GTTCCACTCGTGACAAGTCC 3′and 5′ GTTTCTAAGCAGATCCCG 3′; 5′ GCGAAACGTGCACACTGC 3′ and 5′ GGGCGATGCTTGCATGTTGC 3′ ( T7 promoter sequence removed ) . For hairy , the primers 5′ CCAGAACCTGCTGCTCAT TCG 3′ and 5′ GGGAAAGCGGCTA ACCTCGTTC 3′; for sis-a the primers 5′ CAAAATGCACTACGCCGACG 3′ and 5′ GCATCGTGTCCAACATGACG 3′ were used . All in situs were repeated at least once . Each batch was done simultaneously with an Ore R control , and had sufficient embryos so that several representatives of each cycle could be examined . M3TAP embryos [31] were stained for Sxl ( mouse ) and MSL-1 ( rabbit ) as previously described [45] . To enhance the MSL signal , the M3TAP was first bound ( blocked ) by the same anti-rabbit fluorescent secondary used for the anti-MSL-1 primary before addition of the primary antibody . Embryos were collected on apple juice agar plates for one hour and aged for the appropriate time . They were washed off the plate , dechorionated with 50% chlorox , washed extensively with 1x PBST and frozen at −80°C . RNA was extracted from the frozen embryos using tri-reagent as per manufacturer's protocol . An additional phenol extraction was performed on the purified RNA , followed by DNAse treatment . A PCR test was performed on the RNA to confirm the lack of DNA , after which 4 ug of the RNA was reverse transcribed ( RT ) with AMV RT at 50°C for 15 min followed by 1 . 5 h at 42°C . A small amount ( 2 ng ) of Sxl primer ( 5′ CGT GTC CAG CTG ATC GTC GG 3′ ) was added to the oligo-dT mix ( 100 ng ) per RT , as the stage specific 5′ exons of Sxl are distant from the polyA tail . The quantitative PCRs were performed in triplicate on a Bio-Rad iQ5 thermocycler; Ct values that showed a difference of greater than 0 . 5 from the other two replicates were discarded . For each genotype a minimum of 3 separate RNA samples was analyzed . PCR products were between 200 and 300 bp; primers for SxlPe 5′ CTGTTCGACCATGTCGTCCTA C and CTA CCACCGCTGCCCAGCGAC , SxlPm 5′ GTGGTTATCCCCCATATGGC 3′ and 5′ CTA CCACCGCTGCCCAGCGAC 3′ , sis-a 5′ CGTATACGCACCGTATCGCGG 3′ and 5′ GCATCGTGTCCAACATGACG , runt 5′ CGACGAAAACTACTGCGGCG 3′ and CCAGCCAAGCGGGATTCAGC , upd 5′ GAAAGCGGAACAGCAACTGG 3′ and 5′CAGGAACTTGTAGTTGTGCG 3′ , dpn 5′ CCGATTATGGAGAAACGTCGC 3′ and 5′ CTGAGCCGCTGACGAACACC . Statistical data analysis was completed using Microsoft Excel and GraphPad Prism . | When substantially different , sex chromosomes present the challenge of not only gene dose inequity between the sexes , in the heterogametic sex where one chromosome ( frequently the Y ) carries few genes , but also an inequity relative to the autosomes which are diploid . Dosage compensation refers to the process which equates gene dose between the sexes . Recent results , however , indicate that the mammalian X chromosome avoids monosomy and has a level of expression that is two-fold relative to the autosomes . Hyperactive X chromosome expression in Caenorhabditis elegans has also been suggested , and dosage compensation in the hermaphrodite appears to lower expression of the X chromosomes to match autosome levels . We find that , before the female state is set in Drosophila , the X chromosomes may also express their genes at the two-fold male level and that this level of expression is used to female advantage to consolidate their sex determination . Together , the results suggest that elevated X chromosome expression may be the norm , and that the various dosage compensation processes different organisms utilize reflect a mechanism to counteract an initial hyperactive X chromosome state . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"developmental",
"biology/molecular",
"development",
"molecular",
"biology/transcription",
"initiation",
"and",
"activation"
] | 2010 | Requirement of Male-Specific Dosage Compensation in Drosophila Females—Implications of Early X Chromosome Gene Expression |
An epidemic of Zika virus ( ZIKV ) illness that occurred in July 2007 on Yap Island in the Federated States of Micronesia prompted entomological studies to identify both the primary vector ( s ) involved in transmission and the ecological parameters contributing to the outbreak . Larval and pupal surveys were performed to identify the major containers serving as oviposition habitat for the likely vector ( s ) . Adult mosquitoes were also collected by backpack aspiration , light trap , and gravid traps at select sites around the capital city . The predominant species found on the island was Aedes ( Stegomyia ) hensilli . No virus isolates were obtained from the adult field material collected , nor did any of the immature mosquitoes that were allowed to emerge to adulthood contain viable virus or nucleic acid . Therefore , laboratory studies of the probable vector , Ae . hensilli , were undertaken to determine the likelihood of this species serving as a vector for Zika virus and other arboviruses . Infection rates of up to 86% , 62% , and 20% and dissemination rates of 23% , 80% , and 17% for Zika , chikungunya , and dengue-2 viruses respectively , were found supporting the possibility that this species served as a vector during the Zika outbreak and that it could play a role in transmitting other medically important arboviruses .
Outbreaks of arboviral disease have been documented in islands of the western Pacific including The Federated States of Micronesia ( FSM ) and Palau . Multiple dengue outbreaks have been reported in the western pacific [1]–[4] with an outbreak of dengue 4 virus occurring in Palau in 1995 after a 7 year absence of dengue on this island [5] . This first outbreak of dengue 4 in the Western Pacific also affected FSM the same year [6] . Additional dengue outbreaks occurred more recently in FSM during 2004 and 2012–13 [7] , [8] . In 2007 , an outbreak of acute febrile illness characterized by rash , conjunctivitis , fever , and arthralgia was reported on the island of Yap in the Federated States of Micronesia . While dengue was originally suspected , clinicians noted differences from classical dengue fever and collected serum from acutely-ill individuals for diagnosis . Chikungunya virus ( CHIKV ) was also considered as the clinical presentation was representative of CHIKV infection and an ongoing epidemic of CHIKV was occurring in Southeast Asia . However , Zika virus ( ZIKV ) nucleic acid was detected in 14% of the samples tested and no evidence of alternate etiologies was identified [9] . Zika virus ( ZIKV ) is a member of the family Flaviviridae . Presence of the virus in human specimens has been demonstrated by virus isolation ( samples from Africa and Asia ) and antibody presence ( Asia ) [10]–[15]; however , only a handful of clinical disease cases were described in the literature prior to this 2007 outbreak [16]–[19] . Since the outbreak in Yap , additional ZIKV outbreaks have been documented in Gabon in 2007 and in French Polynesia in 2013 [20] , [21] . Mosquito vectors from which virus has been identified include ( among others ) Aedes africanus , Aedes luteocephalus , Aedes aegypti , and Aedes albopictus ( all belonging to the subgenus Stegomyia ) [15] , [21]–[25] . However , little else is known regarding the natural ecology of the virus . Because this was the first documentation of the virus in Oceania , understanding the biological transmission of the virus was a public health priority . A team including epidemiologists , clinicians , entomologists , and public health personnel investigated the outbreak with the objectives of characterizing the epidemiology , course of clinical illness , and ecological factors contributing to the epidemic and transmission of the virus . Household surveys were performed to obtain serum specimens , to obtain clinical and epidemiological data , to identify risk factors for infection , and to collect entomological specimens for the purpose of determining the most probable epidemic vector [9] . The entomological studies included both immature ( larval and pupal ) and adult surveys to determine the species present on the island , contributions of distinct container types in mosquito maintenance , and to perform virus isolation . This report describes the entomologic findings from the field collected material as well as subsequent laboratory studies assessing the vector capacity of the likely outbreak vector .
The Federated States of Micronesia are located in the Western Pacific Ocean northeast of Papua New Guinea ( Figure 1 ) . Yap State is the westernmost state and is comprised of a main island group consisting of four closely associated islands situated at 9° North and 138° East . It is approximately 6 km wide by 15 km long with a population of 7 , 391 persons during the 2000 census . The climate is tropical with warm temperatures and rainfall reported throughout the year . Mosquitoes were collected and household surveys performed between July 4 , 2007 and July 16 , 2007 at 170 randomly selected homes in 9 out of the 10 municipalities , representing 16% of the total households on the island . The outbreak was estimated to have begun in April , 2007 and continued through July of 2007 [9] . Adult mosquito sampling was carried out using three collection methods . Host seeking mosquitoes were collected using light traps , resting mosquitoes were collected using vacuum aspiration , and mosquitoes looking for oviposition habitat were collected with gravid traps . Gravid and light traps ( light only ) were set in the evening at three sites in the state capital city of Colonia from July 4–9 , and July 12–16 . Collection bags from the light and gravid traps were recovered daily for 9 days in the early morning . A battery operated backpack or handheld mechanical aspirators were used to collect mosquitoes resting in and around random houses where serosurveys were being performed during daytime hours , July 9–15 . All collected mosquitoes were identified morphologically using keys from Bohart [26] , and Rueda [27] . They were then sorted by sex , species , collection method , and collection period and placed into cryovials at a temporary laboratory set up in Colonia . Specimens were frozen at −20°C on-site; they were later transported to the CDC at Fort Collins CO , USA where storage was at −70°C until processing . All indoor and outdoor water containing receptacles at the randomly selected households [9] were inspected for mosquito larvae and pupae . Live larvae observed in receptacles were collected and identified to species and allowed to emerge to confirm identification . All pupae found were collected and reared to adulthood . Key habitat information was recorded and larval indices ( Breteau and household [28] ) were calculated from the collected data for each of the species observed . Each pool of mosquitoes ( not exceeding 40 individuals ) were placed into a 1 . 7 mL polypropylene tube ( Eppendorf , Hauppauge , NY ) and ground with a pestle ( Kontes ) and 500 µl of Dulbecco's minimal essential medium ( DMEM ) ( Gibco ) supplemented with 10% fetal bovine serum ( FBS ) , 100 U/mL of penicillin and streptomycin , 1 U/mL of fungizone and gentamycin . The homogenized mosquitoes were then centrifuged at 15 , 000 g for 1 min . Triturate was then transferred to a new tube and frozen at −70°C . 100 µl of thawed triturate was then plated onto a 96-well cell culture plate ( Corning ) . 50 µl of a Vero cell suspension was then added to the same well and placed into an incubator at 37°C and 5% carbon dioxide . The cell and homogenized mosquito mixture was then monitored daily for cytopathic effect ( CPE ) for 10 days . Medium from those wells presenting signs of CPE ( presumptive positives ) were removed and placed at −70°C until use [29] . The viruses used for laboratory mosquito infections were obtained from the Arbovirus Reference Collection at CDC , Fort Collins , CO ( Table 1 ) . As no ZIKV field strain was obtained , we used the prototype strain for ZIKV laboratory infections . Viral RNA was isolated using the QiaAmp viral RNA protocol ( Qiagen ) . Total RNA was extracted from 50 µl of cell supernatant ( CPE positive pools ) or 100 µl of mosquito homogenate ( artificial infections ) and eluted from the kit columns using 60 µl of elution buffer . The RNA was stored at −70°C until use . Both reverse-transcription PCR ( RT-PCR ) and real-time RT-PCR assays were utilized to detect viral nucleic acid . The Titan one-step RT-PCR ( Roche ) kit was paired with the primers FU1 and cFD3 for detection of Zika [30] , [31] . Briefly , 5 µl of sample RNA was added to the kit components and 400 nM of primers . The manufacturer's protocol was followed with no modifications . The reactions were analyzed by gel electrophoresis . The real-time PCR assay was used on both the presumptive positive pools and the experimentally infected mosquitoes . The previously described Zika ( 800 series set ) and chikungunya virus specific primer and probe sets were used [32] , [33] . The DENV-2 oligonucleotide set were designed with the Primer Select software program ( DNASTAR ) ( 1085 CCAAACAACCCGCCACTCTAAG , 1244c TTTCCCCATCCTCTGTCTACCATA , and TaqMan probe 1145 FAM-AACAGACTCGCGCTGCCCAACACA-BHQ1 ) and were based on the published GenBank full-length sequences . All real-time assays were performed by using the QuantiTect probe RT-PCR reagent kit ( Qiagen ) . Briefly , a 50µl total reaction volume consisted of kit components , 10 µl of RNA , 400 nM of each primer , and 150 nM of probe . The reactions were subjected to 45 cycles of amplification in an iQ5 Real-Time PCR detection system ( BioRad ) following the manufacturer's protocol . The limits of detection for DENV , ZIKV , and CHIKV assays were found using the previously described techniques [34] and were cycle threshold ( Ct ) values of 37 . 7 , 36 . 1 , and 38 . 0 respectively , which is equivalent to approximately 1 . 0 plaque forming unit/mL . In addition , each run included a standard RNA curve . The standard curve was completed by serially diluting the virus stock and extracting the RNA from each dilution , according to the previously mentioned RNA extraction protocol , while simultaneously titrating each dilution in a standard plaque assay . A curve correlation coefficient of ≥0 . 950 and a 90–100% PCR efficiency was used to validate each detection assay . Mosquito eggs were collected at selected houses in Yap using oviposition cups . Briefly , black , plastic cups were lined with seed germination paper [35] and filled approximately half full with water . Cups were placed under foliage near selected homes ( 2–4 feet above the ground ) and collected after 3–5 days . Field collected egg liners were wrapped in moist paper towels , sealed in Ziploc-style bags , and transported to the insectary at the Center for Disease Control and Prevention ( CDC ) , Fort Collins Colorado for colonization . The eggs were washed with a 10% bleach solution prior to hatching in a pan of tap water to eliminate surface fungal and bacterial contaminants . Larvae were supplied with either a liver powder solution or mouse pellets as appropriate for the developmental stage and identified to species as 4th instar . All larvae collected were identified as Aedes ( Stegomyia ) hensilli . Pupation occurred between days 5–7 post hatching . Pupae were removed from the larval pans and allowed to emerge into 1 ft3 adult mosquito cage ( BioQuip ) . In order to produce the next generation , adults were provided an anesthetized mouse as a blood meal source and the engorged females were provided with an oviposition site ( seed germination paper ) to deposit their eggs . The process was repeated in order to get sufficient numbers of experimental mosquitoes . In addition , species verification was performed on F2 adult mosquitoes . Three to four day-old adult Ae . hensilli mosquitoes ( F12–15 ) were fed on blood meals containing ZIKV , CHIKV , or DENV-2 . The blood meals contained equal parts of virus , FBS with 10% sucrose , and sheep blood ( Colorado Serum CO ) washed with phosphate-buffered saline and packed by centrifugation . A Hemotek feeding system ( Discovery Workshops ) was used to deliver the blood meal to the mosquitoes for 1 hour at 37°C . The fully engorged females were separated and placed into a humidified environmental chamber ( Thermo Scientific ) and held at 28°C for 8 days until processing . Blood meal titer was determined by plaque assay to determine input titer . After the 8 day holding period , mosquitoes were cold anesthetized and decapitated with the heads and bodies placed into separate 1 . 7 mL tubes ( Eppendorf ) . A 400 µl aliquot of Dulbecco's minimal essential medium ( DMEM ) ( Gibco ) supplemented with 10% fetal bovine serum ( FBS ) , 100 U/mL of penicillin and streptomycin , 1 U/mL of fungizone and gentamycin was added to each tube and the sample was homogenized using a micropestle ( Kontes ) . The supernatant was clarified by filtration through a 0 . 2 µM syringe filter ( Pall ) and stored at −70°C until use [36] . Virus presence was determined using the virus isolation method as described above . An infected mosquito exhibited a virus positive body [percent infected = ( number positive bodies/total number of mosquitoes processed ) X 100] while those with disseminated infections were the infected individuals with virus in the head [percent disseminated = ( number of positive heads/number of positive bodies ) X 100] . Quantities of viral RNA were determined using real-time RT-PCR ( above ) and correlated with viral titer .
Adult mosquitoes were captured using three different collection methods ( light trap , gravid trap , and vacuum aspirations ) . A total of 879 mosquitoes were collected in 84 trap nights . Additionally , 475 individuals collected as larvae and/or pupae were reared to adults for confirmatory identification and processing . Nine species were identified in these collections ( Table 2 ) . The most abundant adult species collected was Aedes hensilli ( 41 . 2% ) followed by Culex quinquefasciatus ( 28 . 1% ) . All other species each comprised less than 10% of the total collection . All adult mosquitoes ( field collected adults and those reared from immatures ) were processed and subjected to virus isolation efforts . No viable virus was recovered from any of these mosquitoes . From 170 randomly surveyed households ( July 4–16 , 2007 ) , 1366 water holding habitats were identified . Larvae and/or pupae were collected from 586 of these containers and 85% of surveyed households had at least one infested habitat; individual habitats sometimes contained more than one species ( Figure 2 ) . The most prevalent containers with larvae or pupae were discarded cans followed by coconut shells ( Table 3 ) . Proportionally , containers including tires , tarps , floats , and bamboo had high percentages of immatures but several of these container types were found only infrequently ( Table 4 ) . Containers such as water barrels , used to collect rainwater , while proportionally fewer in number than other containers , were actually major contributors to mosquito production due to the sheer number of larvae and pupae present ( e . g . thousands of immature mosquitoes per water barrel in comparison with cans or shells which typically contained fewer than 10 individuals each ) . In total , ten different species were identified from the larval collections . Ae . hensilli was both the most abundant and most prevalently identified immature species being found in 83% of the infested containers distributed all over the island ( household index of 81 . 2 and Breteau index of 282 . 9 ) . Because no virus was found in any of the field collected material , laboratory infections were performed on the most common mosquito collected , Ae . hensilli , to determine if this species could have served as the epidemic vector . Cohorts of Ae . hensilli were infected with three different viruses during these studies: 1 ) ZIKV - to determine if this was the likely vector during the outbreak; 2 ) CHIKV- to ascertain whether Ae . hensilli could serve as a vector for this virus which was expanding through SE Asia and was considered as a possible etiology of the outbreak prior to ZIKV diagnosis; 3 ) DENV - as Ae . hensilli was previously postulated as the vector of the 1996 dengue outbreak in Yap [5] . Cohorts of 3–4 day old adults were provided infectious blood meals with titers of at least 4 . 9 log10 pfu/mL . Mosquitoes provided the lowest dose of ZIKV were resistant to infection with only 7% becoming infected ( Table 5 ) . However , at least 80% of those receiving a slightly higher dose became infected . Curiously , only 13–23% of those developed disseminated infections . Only a small percentage of mosquitoes exposed to DENV-2 became infected ( 0–21% ) and few of these had virus dissemination . In contrast , Ae . hensilli was found to be exceptionally sensitive to CHIKV with infection and dissemination rates greater than 60% and 80% respectively .
The 2007 outbreak of ZIKV in Yap prompted the investigation of vectorial capacity of the predominant local mosquito to transmit this virus and other related viruses that are present or threaten to affect FSM and other Western Pacific island countries . Yap State , the western-most part of FSM , has previously been affected by arboviral outbreaks [6] but the discovery of ZIKV on the island highlighted the risk of epidemics due to agents previously unknown to the area . During the entomological investigations , collection of mosquito larvae and pupae from over 15 distinct container types revealed a wide range of habitats , both natural and artificial , that could support development of a variety of mosquito species . Because the island extensively imports products via cargo ships , introduction of exotic species that could utilize the variety of habitats is a strong possibility . This could allow further novel arboviral introduction events on the island . For example , Ae . albopictus could easily be or have been introduced to the island due to the proximity and intense air and sea traffic with Guam and Mariana Islands where this species is widespread [37] . None were found during this study . The overwhelmingly predominant mosquito species found on the island was Ae . hensilli . This mosquito was previously speculated to be the vector of DENV during the 1995 outbreak in Yap State as it was the only Aedes ( Stegomyia ) present on some affected islands [6] . However , like in this outbreak , no isolations were made from field-collected mosquitoes and no arboviruses have ever been reported from this species so incrimination as a vector could not be biologically confirmed . The collection of additional mosquitoes may have allowed virus isolation from field material but repeated strong rainstorms limited the number of adults collected . As in the previous dengue outbreak , Ae . hensilli is the most probable outbreak vector due to its high density , widespread distribution on the island , and its tendency to bite humans . Although transmission studies may have helped clarify vector status , laboratory infection studies reported here further suggest that this is a probable vector due to the high infection rates with ZIKV . While there is an admittedly suboptimal dissemination rate to indicate vector status for ZIKV , there has been documentation of other Aedes ( Stegomyia ) mosquitoes serving as outbreak vectors even with low susceptibility to infection or dissemination . For example , Ae . aegypti , which has been reported to be relatively resistant to infection to yellow fever virus , has nevertheless been implicated in outbreaks of yellow fever [38] . Vector status of Ae . hensilli for DENV-2 is more difficult to assert based on the laboratory data indicating less than 20% infection rates with virtually no dissemination . However , susceptibility to viruses in at least 2 distinct arboviral genera ( flavivirus and alphavirus ) suggests that this species could possibly serve as a vector of other medically important arboviruses typically transmitted by Aedes ( Stegomyia ) species ( e . g . yellow fever and chikungunya viruses ) . It could also serve as a vector of arboviruses in large population centers where the mosquito is found [39] . Aedes hensilli has a limited known distribution consisting of FSM , Palau , and Singapore [39] suggesting that these additional areas might also be potentially at risk due to arboviral pathogens vectored by this species . Since little is known of the biology or zoonotic transmission of ZIKV , it is also possible that other Scutellaris group species ( among others ) could be possible vectors of the virus . This is supported by the findings that ZIKV has previously been associated with Ae . africanus [23] , [40] , [41] , Ae . luteocephalus [42] , and Ae . aegypti [15] mosquitoes . There are numerous Scutellaris group mosquitoes from island ecologies including Aedes cooki , Aedes polynesiensis , Aedes palauensis , Aedes rotumae , and Aedes scutellaris , and others , some of which have been implicated in arboviral transmission [42]–[48] . The range of the Scutellaris group mosquitoes should be considered as possible vectors of ZIKV in islands of the Pacific and elsewhere . Aedes hensilli was found to be very susceptible to infection by CHIKV . This finding was interesting as the strain of CHIKV selected was a Central/East African genotype strain associated with the Indian Ocean lineage but not possessing the valine residue at E1that has been linked to increased infectivity in Ae . albopictus [49] . A strain without this mutation was specifically selected to evaluate the susceptibility of Ae . hensilli to a virus that may not have been adapted to alternate Scutellaris group mosquitoes . However , the high degree of susceptibility to CHIKV even without the valine reside at position 226 is not completely unexpected as distinct populations of Ae . albopictus have historically shown significant susceptibility to CHIKV [50] . The ability of Ae . hensilli to be infected with CHIKV again , like with ZIKV , indicates that geographic areas with less well characterized Scutellaris group mosquitoes should consider alternate species to be potential vectors of introduced arboviral diseases . | Arthropod-borne viruses ( arboviruses ) cause significant human morbidity and mortality throughout the world . Zika virus , which is reported to be transmitted by Aedes ( Stegomyia ) species mosquitoes , caused an outbreak on the island of Yap , in the Federated States of Micronesia in 2007 . This was the first described outbreak of Zika in Oceania , which has had several arbovirus outbreaks in the past . Diagnosing the outbreak was difficult due to the similarity in clinical symptoms between disease caused by Zika virus and other viruses . This work describes the efforts to identify the mosquito species that were responsible for transmission of the virus . While no virus was isolated from any species of mosquito collected during the current study , the predominant species found was Aedes hensilli and through the complementary laboratory studies , this mosquito was implicated as a probable vector for Zika virus . In addition , this species was found to be susceptible to both the medically important dengue-2 and chikungunya viruses . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"public",
"and",
"occupational",
"health",
"infectious",
"diseases",
"tropical",
"diseases",
"medicine",
"and",
"health",
"sciences"
] | 2014 | Aedes hensilli as a Potential Vector of Chikungunya and Zika Viruses |
Deedu ( DU ) Mongolians , who migrated from the Mongolian steppes to the Qinghai-Tibetan Plateau approximately 500 years ago , are challenged by environmental conditions similar to native Tibetan highlanders . Identification of adaptive genetic factors in this population could provide insight into coordinated physiological responses to this environment . Here we examine genomic and phenotypic variation in this unique population and present the first complete analysis of a Mongolian whole-genome sequence . High-density SNP array data demonstrate that DU Mongolians share genetic ancestry with other Mongolian as well as Tibetan populations , specifically in genomic regions related with adaptation to high altitude . Several selection candidate genes identified in DU Mongolians are shared with other Asian groups ( e . g . , EDAR ) , neighboring Tibetan populations ( including high-altitude candidates EPAS1 , PKLR , and CYP2E1 ) , as well as genes previously hypothesized to be associated with metabolic adaptation ( e . g . , PPARG ) . Hemoglobin concentration , a trait associated with high-altitude adaptation in Tibetans , is at an intermediate level in DU Mongolians compared to Tibetans and Han Chinese at comparable altitude . Whole-genome sequence from a DU Mongolian ( Tianjiao1 ) shows that about 2% of the genomic variants , including more than 300 protein-coding changes , are specific to this individual . Our analyses of DU Mongolians and the first Mongolian genome provide valuable insight into genetic adaptation to extreme environments .
Prehistoric Mongolian ancestry can be traced to the Gobi and Mongolian steppes in Northeastern Asia , yet Mongolian lineages are found among present-day inhabitants of regions as far west as Eastern Europe [1]–[3] . This vast genetic signature is largely attributed to population movements during the time of Genghis Khan [3] , [4] , whose efforts to unite Eurasian tribes during the 13th century helped to establish one of the largest contiguous empires in human history ( 12 million square miles ) [5] , [6] . Assimilation with resident inhabitants throughout this vast and varied region has resulted in a complex mixture of genetic variation and cultural practices in Mongolians that may promote survival in novel environments [7] , [8] . A better understanding of Mongolian genetic diversity will provide important insights into genetic variation throughout northern Eurasia and genetic adaptation in highly variable environments . Through much of their history , Mongolians have survived the harsh conditions of northern latitudes , including seasonal cold and drought and a highly restricted diet , and they might have genetically adapted to these conditions . More than 500 years ago , the nomadic Deedu ( “at high altitude” ) Mongolians ( referred to as “DU Mongolians” hereafter ) migrated from the Mongol steppes to the northeastern section of the Qinghai-Tibetan Plateau [5] , [6] , [9] . In this new environment ( ∼3000 meters above sea level ) , they have been further challenged by hypoxic conditions . Adaptations to the challenge imposed by hypoxia have been studied in highland Tibetan populations , and putatively adaptive genetic factors that are associated with both hypoxia and metabolic factors have been reported [10] . In addition , adaptations to hypoxia and cold have also been reported in high-altitude deer mice , highlighting regulatory shifts in energy utilization and thermogenic capacity [11] . In humans , putatively adaptive genetic factors in highland Tibetans are also associated with metabolic factors that may be linked with cold tolerance [10] . While it is unlikely that genome-wide selection analyses would identify selective events that occurred since their arrival into the Qinghai-Tibetan Plateau about 25 generations ago , we hypothesize that shared ancestry and recent genetic admixture from neighboring long-term highland residents play a crucial role in the adaptation of DU Mongolians to this region . By examining genome-wide SNPs in a DU Mongolian population and whole-genome sequence data from a DU Mongolian individual , we are able to explore Mongolian genetic variation at an unprecedented level of detail . Our analysis identifies hypoxia , antimicrobial , and metabolic selection candidates and reveals novel variation in the genomes of this unique population .
To assess the population structure of DU Mongolians and their relationship with surrounding populations , we examined ∼860 , 000 SNPs genotyped in 369 individuals from 10 Eurasian populations ( Figure 1A ) . A pairwise FST analysis ( Table 1 ) indicates the DU Mongolians are most closely related to the Buryat Mongolians ( FST = 0 . 0075 ) , followed by the Kyrgyzstani population and the Tibetan population from the Tuo Tuo River ( TTR ) area of the Qinghai-Tibetan Plateau ( FST = 0 . 0105 and 0 . 0114 , respectively ) . A neighbor-joining tree from the pairwise FST values showed similar patterns , with DU Mongolians positioned outside of the clade that includes Tibetans , Han Chinese and Japanese ( Figure 1B ) . To further examine the relationships among individuals , we performed principal components analysis ( PCA ) using pairwise genetic distances among all individuals . Two-thirds ( 67 . 8% ) of the variance in the dataset can be explained by the first principal component ( PC1 ) , separating populations from Europe and West Asia from populations from East and Central Asia ( Figure 2A ) . PC2 only accounts for <1% of the variance , separating Mongolians , Tibetans , HapMap CHB and JPT individuals , and Kyrgyzstanis . When only East Asian populations were examined , DU Mongolians are most closely related to Buryat Mongolians ( Figure 2B ) . Two Tibetan populations ( from the TTR and Maduo regions ) , HapMap CHB , and HapMap JPT are all clearly separated from the Mongolian populations . Although the majority of the DU Mongolian individuals form a relatively distinct cluster , several individuals show more similarity to other East Asian populations , such as Tibetans and Han Chinese , suggesting possible gene flow among these populations . Next we used the program ADMIXTURE to assess the genetic composition of DU Mongolians and neighboring populations . The ADMIXTURE analysis estimates the ancestry of each individual , assuming 3–6 ancestral populations ( K ) ( Figure 3 ) . At K = 3 , ancestral components corresponding to European/West Asian , Tibetan/DU Mongolian , and East Asian/Buryat Mongolians are recognized . At K = 4 , a Mongolian component including DU and Buryat Mongolians is recognized . DU Mongolian genomes appear to be mixtures of ancestral Mongolian and Tibetan components . These two components are closely related and separate from other East Asians ( HapMap CHB/JPT ) . On average , DU Mongolian genomes are composed of 44% of the Mongolian component ( standard deviation [sd] 11% ) , 52% of the Tibetan components ( sd 6% ) , and 3% East Asian component ( sd 8% ) . In contrast , the Buryat Mongolian genomes are composed primarily of Mongolian ( 75% , sd 10% ) and East Asian ( 25% , sd 9% ) components . This result suggests that unlike Buryat Mongolians , DU Mongolians share substantial genetic components with Tibetans , either due to recent admixture or more ancient shared ancestry . At K = 6 , Maduo and TTR Tibetans were recognized as two separate groups , and DU Mongolian genomes appear as mixtures of TTR Tibetan and Mongolian components . Some of the DU Mongolian individuals also contain more than 5% East Asian components . The Kyrgyzstanis from Central Asia have the most genetic admixture , and about half of their ancestry is accounted by the ancestral Mongolian component at K = 4 to 6 . We performed iHS and XP-EHH analyses to identify candidate regions targeted by positive selection in the DU and Buryat Mongolian populations . In total , 96 regions ( 68 containing genes; 28 non-genic regions ) comprising 162 candidate genes were identified in both DU and Buryat Mongolian populations in at least one of the two selection analyses ( Table S1 ) . We hypothesize that selection candidates identified in both Mongolian populations would be associated with factors common to ancestral Mongolian groups at northern latitudes ( see [12] for review ) such as metabolic adaptations to temperature and diet . Indeed , a number of metabolic genes are identified in the overlapping candidate gene sets ( e . g . , ADRA2A , MYOF , and CYP26A1/C1 ) . In addition , we identified a gene cluster containing six beta-defensin genes ( DEFB125 , DEFB126 , DEFB127 , DEFB128 , DEFB129 , and DEFB132 ) among the selection candidates . It is known that genes related to immune system and microbial defense are highly variable [13] . In particular , this defensin gene cluster varies in copy number in humans and might have undergone copy number expansion in East Asian populations [14] . Considering the geographic proximity of Mongolians and Tibetans , as well as their shared cultural practices and environmental conditions , we also examined the overlap of selection targets among these populations . A total of 153 selection regions containing 399 genes were identified as significant ( p<0 . 02 ) in at least one Mongolian and one Tibetan population ( Table S2 ) . Among the top selection candidates is PPARG ( peroxisome proliferator-activated receptor gamma ) , which has been hypothesized as a “thrifty” gene for dietary adaptation [15] , [16] . PPARG may also be associated with metabolism-related traits involving cold tolerance and energy utilization , such as thermogenesis and brown adipose tissue [17] . A number of selection candidate genes reported in previous studies of Asian populations were also found in the Mongolian-Tibetan overlapping candidates . For example , EDAR ( the ectodysplasin A receptor gene is significant in both the DU Mongolians and the TTR Tibetans ( iHS , p = 0 . 004 and 0 . 016 , respectively ) . EDAR is involved in the development of hair , teeth , and exocrine glands [7] and has been highlighted as a strong selection candidate in several Asian populations [18] , [19] . A total of six regions ( 15 total candidate genes ) identified in our analyses are significant in all four Mongolian and Tibetan populations ( Table S2 ) . The number of shared regions is significantly higher than the expected number ( ∼0 . 3 , p = 0 ) under a null hypothesis of independence . This highly refined list of selection candidates , which contains genes related to muscle metabolism and angiogenesis ( MYOF , MAP2K5 ) and tumor suppression ( MCC ) , yields valuable insight for future examination of adaptation in both Mongolian and Tibetan ethnic groups . We next compared putatively selected regions ( p<0 . 02 ) in DU Mongolians with genes that have been previously associated with high-altitude adaptation in Tibetans [20]–[25] . Three Tibetan high-altitude selection candidates are also highly significant in DU Mongolians , including the hypoxia inducible pathway gene EPAS1 [20]–[22] , [24]–[27] ( Figure 4 ) ; the PKLR gene [20]–[22] , which encodes liver- and erythrocyte-expressed pyruvate kinase; and the cytochrome P450 family member CYP2E1 [20] . These hypoxia-related genes are also associated with metabolic processes [17] , [23] that may be involved in orchestrating important physiological responses at high altitude . To control for the effect of admixture among the DU Mongolians , we also performed two SNP-based selection scans in addition to the haplotype based selection analysis . Results from the population branch statistic ( PBS ) and multiple regression analysis ( MR ) are shown in Tables S3 and S4 , respectively . Among the high-altitude candidate genes examined , EPAS1 showed significant signal ( p<0 . 005 ) in the PBS analysis . PPARG , a selection candidate identified among the top 2% of iHS selection candidates in Tibetans and DU Mongolians , is also significant in the DU Mongolian PBS analysis ( p<0 . 005 ) . Before DU Mongolians migrated to the Qinghai-Tibetan plateau , Tibetans had lived on the plateau for thousands of years and had adapted genetically to high altitude [28] , [29] . On average , Tibetans exhibit relatively lower hemoglobin concentration ( [Hb] ) compared to other groups at comparable high altitude [30] , [31] , and this unique characteristic is associated with adaptive genetic factors [20]–[22] . The ADMIXTURE results support shared ancestry and possible gene flow from Tibetans to DU Mongolians ( Figure 3 ) , which in turn suggests the hypothesis that Tibetan high-altitude adaptive genes have been transferred to the DU Mongolian population . To test this hypothesis , we measured [Hb] in DU Mongolians and compared the results to those of Tibetans and Han Chinese from a previous study [32] ( Table 2 ) . The [Hb] of the Tibetan and Han Chinese samples were collected at an altitude ( 2664+/−258 m ) , comparable to the region where the DU Mongolian samples were collected ( ∼3000 m ) . When the individuals are divided into two age groups , the average [Hb] of DU Mongolian females is similar to Tibetans and significantly lower than Han Chinese in the 16–40 year age group ( p<0 . 006 ) , but is similar to Han Chinese and significantly higher than Tibetans in the 41–60 year age group ( p<0 . 007 ) . Previously reported haplotypes that have undergone selection in Tibetans [20] are not associated with [Hb] in this sample ( Table S5 ) . To identify potential functional variants in DU Mongolians , and to compare patterns of genetic variation from whole-genome sequencing with those based on SNP microarrays , we performed whole-genome sequencing on a DU Mongolian male ( Tianjiao1 ) who was born and raised on the Qinghai-Tibetan Plateau . Relative to the reference sequence , a total of 3 , 803 , 076 variants were identified in Tianjiao1 , including 3 , 353 , 824 single nucleotide variations ( SNVs ) , 198 , 230 deletions , 180 , 121 insertions ( a total of 378 , 351 indels ) , and 70 , 901 complex substitutions and multiple nucleotide polymorphisms ( Table 3 ) . In addition , 170 copy number variations ( CNV ) and 1439 high-confidence structural variants ( SV ) were identified . Using whole-genome data , we determined the mitochondrial and Y haplogroups of Tianjiao1 ( Table S6 ) . The Tianjiao1 mtDNA sequence is assigned to clade H18 [33] , which has been previously reported in Europeans [34] , [35] and is also present in the Arabian Peninsula and Caucasus regions . H18 is estimated to have originated ∼13 , 500 years ago [36] . The Tianjiao1 Y chromosome produced 1 , 897 variant calls and can be assigned to haplogroup Q1a1 based on the current Y-DNA haplotype tree [37] . The Q1a1 ( M120 ) Y-chromosome lineage has been reported previously at low frequencies in populations from Mongolia , central and northern Asia , the Lhasa region of Tibet , and in Han Chinese [38] , [39] . We compared the Tianjiao1 genome to whole genome sequences of 54 unrelated individuals sequenced using the same sequencing platform and variant discovery pipeline . These 54 individuals ( referred to as “CGI54 panel” hereafter ) were selected from 11 populations across the world and are a good representation of overall genomic diversity in humans . The total number of variants in the Tianjiao1 genome is similar to that of other Eurasian individuals and is lower than African individuals in the CGI54 panel ( Figure S1 ) . Tianjiao1 shared most variants with CHB individuals and least with LWK ( Luhya in Webuye , Kenya ) individuals ( Figure 5A ) . A neighbor-joining tree based on pairwise genetic distances demonstrates the affinity of Tianjiao1 to other East Asian individuals ( Figure 5B ) . This whole-genome comparison result is consistent with our PCA results ( Figure 2 ) . In our ADMIXTURE analysis ( K = 4 ) , the Tianjiao1 genome appears to be a mixture of three ancestral components and is estimated to contain 26% , 56% , and 17% Mongolian , Tibetan , and HapMap CHB/JPT ancestry , respectively . To assess the functional impact of variants in this sequence , we annotated the effects of the variants using the terms defined in Sequence Ontology [40] . The number of variants in each category is shown in Table 4 . We found there are 1 , 467 , 962 genic variants , including 19 , 964 coding sequence variants . Among the coding variants , 9 , 216 are non-synonymous substitutions . There are also 62 stop-gain and 28 stop-lost SNVs . Among indels , 160 are in-frame and 134 are frame-shifting . Within introns , 61 mutations are located at the splice donor sites and 45 are located at splice acceptor sites ( Table 4 ) . We compared these functional variants with the Human Gene Mutation Database ( HGMD ) to identify variants that are previously known to be associated with disease phenotypes . A total of 43 HGMD mutations are present in the Tianjiao1 genome , including 9 homozygous autosomal mutations and one hemizygous mutation on chromosome X . All but one of these variants are known SNVs that are present in samples from the 1000 Genomes project . The only nonsense mutation ( chr17:15142830C->G ) that is specific to Tianjiao1 is located in the PMP22 gene ( peripheral myelin protein 22 [HGNC 9118] ) and has been associated with Charcot-Marie-Tooth disease ( CMT1A , [MIM 118220] ) [41] . Previous studies of DNA sequences from apparently healthy individuals have yielded similar results [42]–[45] . A comparison of the variants found in Tianjiao1 with those of the CGI54 panel and the 1000 Genomes Project shows that 31 , 216 variants are specific to Tianjiao1 . Of these , 449 are present in the coding regions , including 139 synonymous SNVs , 255 non-synonymous SNVs , 5 nonsense SNVs , 34 frame-shifting indels , and 17 in-frame indels ( Table 4 ) . Using the program SIFT [46] , we predicted the functional impact of non-synonymous SNVs and coding indels specific to Tianjiao1 . In addition to the 5 stop-gain SNVs , 77 non-synonymous SNVs were predicted to be “Damaging” or “Possibly damaging” ( Table S7 ) . Among the 51 coding indels , 8 are within the first 10% of the transcript and 23 were predicted to cause nonsense-mediated decay of the transcript ( Table S8 ) . Finally , we examined Tianjiao1-specific coding variants in the putatively selected regions that are shared by Mongolians ( Table S1 ) and between Mongolians and Tibetans ( Table S2 ) . Among all regions , two Tianjiao1-specific heterozygous nonsynonymous SNVs were identified in genes PAOX and EPHB6 , respectively . The mutation in the PAOX gene ( p . D274G ) was predicted to be “Damaging” while the mutation in EPHB6 ( R413L ) was predicted to be “Tolerated” by SIFT ( Table S7 ) . Within these regions , no coding indels , CNVs , or SVs were identified in the Tianjiao1 genome .
Our results show that DU Mongolians form a distinct population compared to other Eurasian populations and , among our samples , have closest genetic similarity to Buryat Mongolians . The ADMIXTURE analysis suggests that DU Mongolians share appreciable amounts of ancestry with neighboring Tibetan populations . This ancestral component is distinct from other East Asian populations . Genetic and archaeological evidence indicates that regions of the southern portion of the Tibetan Plateau were first occupied during the Late Pleistocene ( ∼30 , 000 years ago ) [25] , [47] , [48] . Approximately 3 , 750–6 , 500 years ago , additional populations migrated from the east into the northeastern section of present-day Qinghai Province and the easternmost regions of Tibet [47] . The ancestry shared among Qinghai Mongolian and Tibetan populations examined here may be attributed to the latter more recent migration into the northeastern section of the Qinghai-Tibetan Plateau . Because of the shared ancestry and possible gene flow between DU Mongolians and Tibetans , one intriguing question is whether high-altitude adaptive alleles have been transferred to the DU Mongolians after their migration to the Qinghai-Tibetan plateau . DU Mongolians specifically exhibit strong signals of selection for genes previously reported as candidates for high-altitude adaptation in neighboring Tibetan populations , including EPAS1 , PKLR , and CYP2E1 ( p<0 . 02 in DU Mongolian and Tibetan populations ) . The endothelin receptor type A ( EDNRA ) selection candidate gene reported by Simonson et al . [20] is significant in Buryat but not DU Mongolians , suggesting an adaptive role that may not be strictly hypoxia-specific . Our comparison of selection candidate genes among Buryat and DU Mongolians and two Tibetan populations yields candidate genes that may be related to metabolic factors involved in adaptation to cold , arid conditions and a relatively restricted diet ( Table S2 ) . Notably , the PPARG gene , previously hypothesized to be associated with metabolic adaptation in human populations [12] , [15] , [16] , exhibits a strong signal of selection in Mongolian and Tibetan populations and warrants further investigation . The individual variant-based selection scans , PBS and MR , showed some overlap in selection candidate regions with our haplotype based scan , although the top regions are largely different . For example , the EPAS1 gene was identified as a strong selection candidate gene in DU Mongolians by the XP-EHH and PBS tests but not in the iHS and MR tests , reflecting differences in these approaches . Because of the shared recent ancestry of these populations , it is unclear if these selection targets are the outcome of more ancient selective events common to ancestors of these groups or whether Mongolians and Tibetans exchanged favorable genes in more recent history . While we did not detect an association between putative selected Tibetan haplotypes and [Hb] in DU Mongolians , the difference between DU Mongolian and Han Chinese [Hb] phenotypes at an altitude of 3000 m suggests that DU Mongolians might share Tibetans' phenotype of decreased [Hb] in a hypoxic environment . Our whole-genome sequence and genome-wide SNP analyses provide the first genomic-level insight into a Mongolian population , augmenting our current understanding of human genetic variation . Further studies of this unique population will elucidate the evolutionary underpinnings of adaptive signals shared among other Mongolian populations in addition to neighboring Tibetan groups . Efforts focused on Mongolian genomics beyond the single genome analysis presented here will also provide greater insight into direct targets of selection that are likely associated with important physiological and metabolic traits .
Blood samples from all DU Mongolian subjects were drawn at an altitude of 3000 meters ( m ) . All the subjects are nomads who permanently live at an altitude of 3000 m–4300 m in the Qinghai Keke Xili area . DNA was extracted from whole blood samples using the Qiagen Gentra Puregene Blood Kit ( Qiagen Inc . , Valencia , California , USA ) . Informed consent was obtained for all participants according to guidelines approved by the Institutional Review Board at the High Altitude Medical Research Institute , Qinghai Medical College ( Xining , Qinghai , China ) . A healthy , 58-year-old DU Mongolian male who was born and raised at an altitude of 3250 m on the Qinghai-Tibetan Plateau provided a blood sample from which DNA was extracted for whole-genome sequencing . The sample collection was carried out in Salt Lake City , Utah , and whole-genome sequencing was undertaken at Complete Genomics , Inc . Both procedures were performed with Institutional Review Board approval under the University of Utah's Clinical Genetics Research Program ( IRB #7551 ) . Forty-nine individual DNA samples were genotyped using Affymetrix 6 . 0 SNP Array technology ( >900 , 000 SNPs ) at Capital Bio Corporation ( Beijing , China ) . We used default parameters for the Birdseed algorithm ( version 2 ) to determine genotypes for all samples ( Affymetrix , Santa Clara , CA , USA ) and the ERSA program to detect cryptic relatedness between subjects [49] . When a pair of individuals was estimated to be more closely related than first cousin , one member of the pair was excluded from the analyses . Based on these criteria , seven individuals were removed , leaving a total of 42 individuals for subsequent analysis . The genotype data were then combined with data from other Northern Eurasian individuals who were previously genotyped using the same platform [10] , [20] , [50] . The final dataset contains ∼860 , 000 SNPs that were genotyped in 369 individuals from 10 Northern Eurasian populations . SNP genotypes of the final dataset are available on our website ( http://jorde-lab . genetics . utah . edu/ ) under Published Data . Between-population FST estimates , neighbor-joining tree construction , pairwise allele-sharing genetic distance calculation , and principal components analysis ( PCA ) were performed using MATLAB ( ver . r2011b ) as previously described [50] , [51] . Genome-wide admixture estimates were obtained with the ADMIXTURE algorithm ( version 1 . 02 ) [52] . To eliminate the effects of SNPs that are in linkage disequilibrium , we first filtered out SNPs that had pairwise r2>0 . 2 within 50 SNP windows using PLINK [53] as recommended by the authors of ADMIXTURE . The pruned data set contains 142 , 888 SNPs . The XP-EHH ( cross-population extended haplotype homozygosity ) [19] , iHS ( integrated haplotype score ) [54] , and PBS ( population branch statistic ) selection scans were performed as previously described [20] , [21] . For XP-EHH and PBS selection scans , our test statistic was the maximum score in each 200 kb region [19] . For XPEHH , we used the HapMap CHB and JPT populations as reference populations and calculated XP-EHH at each site using default settings ( http://hgdp . uchicago . edu/Software/ ) . For PBS , we used the HapMap CHB and JPT as the first reference population and HapMap CEU as the second . We determined statistical significance for each region from the empirical distribution of the test statistic and selected regions that are significant at the 0 . 02 level as our candidates . To identify selected loci at intermediate frequencies , we employed the iHS statistic . We summed the integrated EHH in both directions from each SNP until EHH was less than or equal to 0 . 10 and calculated the iHS score as the log ratio of iHH single-site scores standardized by the population-derived allele frequency . After excluding regions with <5 SNPs , we determined the iHS test statistic for each 200 kb region as the fraction of SNPs in which the absolute value of iHS was >2 . 0 . The expected number of overlapping regions in Mongolian and Tibetan populations was determined using simulation assuming independence of the regions , and the significance level of the observation was determined using the empirical distribution from the simulation . The multiple regression ( MR ) analysis was performed as described in Alkorta-Aranburu et al . [55] to identify unusually divergent SNP loci in the DU Mongolian population . Specifically , the observed allele frequencies at all genotyped SNPs in six populations ( excluding the Buryat Mongolians and two Tibetan groups ) were used to predict the allele frequencies at those loci in DU Mongolians as a linear combination of frequencies in the six populations . The MR Score of a SNP is defined as the Studentized ( scaled ) residual of the observed vs . predicted allele frequency for the SNP . For each gene , we computed an MR Factor as the proportion of SNPs within its transcribed region plus 10 kb on either side that are in the tail of the MR Score distribution divided by the proportion of tail SNPs for all such gene regions . As in Alkorta-Aranburu et al . [55] , we considered three tails of the MR Score distribution ( 5% , 1% and 0 . 5% ) and computed MR Factors for each level . The MR Factors and empirical p-values ( transformed rank statistics ) for the top 2% of gene regions with at least 10 genotyped SNPs are shown in Table S4 . To control for potential batch effect in genotype calling , we combined our samples and HapMap samples and performed genotype calling on whole dataset . SNPs that showed less than 50% concordance rate in any given population between the original genotypes and the recalled genotypes were excluded from the PBS and MR analyses . [Hb] and hematocrit were determined from venous blood samples using the Mindray Hematology Analyzer ( BC-2300 , Shenzhen , China ) . Of the 42 unrelated samples , we were able to obtain [Hb] from 30 non-smoking , healthy individuals . The [Hb] values of 326 Tibetan and 335 Han Chinese female individuals ( between ages 16 and 60 ) collected at a comparable altitude ( 2664 m ) were obtained from a previous study [32] . To determine if [Hb] in DU Mongolians is associated with haplotypes that have undergone selection in Tibetans , we first defined the selected Tibetan haplotypes using three core SNPs in each region as previously described [20] and estimated the copy number of each haplotype in each DU Mongolian individual . We then used a stepwise linear regression model ( MATLAB ver . r2011b ) to determine the association between [Hb] and the inferred Tibetan haplotype copies of each genic region in DU Mongolians . Whole-genome sequencing was performed by Complete Genomics , Inc . ( CGI ) . The whole-genome sequencing yielded ∼1 . 6×1011 bp of total sequence , with an average coverage of 51 . 7× for the entire genome . The raw sequences were mapped to the NCBI reference human genome build 37 ( hg19 ) , and variant calling was performed by CGI using their variant-calling pipeline ( Software version 1 . 11 . 0 . 18 ) . Overall , both alleles were determined for 96 . 2% of the genome , one of the two alleles for 0 . 7% of the genome , and neither allele for 3 . 2% of the genome . The whole-genome variant calls of individuals in the CGI54 panel were obtained from the CGI website ( ftp://ftp2 . completegenomics . com/ ) . This panel contains 54 unrelated individuals , including all members of the CGI diversity panel ( 46 individuals ) , parents in the YRI and the PUR trios ( four individuals ) , and four grandparents in CEU Pedigree_1463 . The 1000 Genomes phase 1 variant calls were downloaded from the 1000 Genomes website ( ftp://ftp . 1000genomes . ebi . ac . uk/vol1/ftp/phase1/ ) . To validate the mtDNA variants in the Tianjiao1 genome , we performed Sanger sequencing on the HVS1 region and resequenced the entire mitochondrial genome using the Ion Torrent platform . All methods produced identical variant calls and confirmed the whole-genome sequencing results . The functional impact of the genomic variants was assessed using the Variant Annotation Tool ( VAT ) in the Variant Annotation , Analysis and Search Tool ( VAAST ) package [56] . Tianjiao1 variants that are present in the CGI54 panel individuals were excluded using the Variant Selection Tool ( VST ) in the VAAST package , and variants that overlap the 1000 Genomes phase 1 variants were excluded using tabix ( http://samtools . sourceforge . net ) [57] . The functional significance of coding SNVs and indels was assessed using SIFT [46] ( SIFT Human Coding SNPs: http://sift . jcvi . org/www/SIFT_chr_coords_submit . html; SIFT Human Coding indels: http://sift . jcvi . org/www/SIFT_chr_coords_indels_submit . html ) . | Throughout history , Mongolians have survived the harsh conditions of northern latitudes , including seasonal cold , drought , and a restricted diet . Approximately 500 years ago , nomadic Deedu ( DU; “at high altitude” ) Mongolians migrated from the Mongolian steppes to the northeastern highlands of the Qinghai-Tibetan Plateau . Using high-density SNP data , we demonstrate that present-day DU Mongolians share genetic ancestry with other Mongolians and with Tibetans . High-altitude selection candidate genes previously identified in the latter population ( EPAS1 , PKLR , CYP2E1 ) , and PPARG , a gene long hypothesized to play a role in metabolic adaptation , are among the strongest adaptive signals in DU Mongolians . Furthermore , we show that hemoglobin concentration , associated with high-altitude adaptation in Tibetans , is intermediate in DU Mongolians compared to Tibetans and Han Chinese at comparable altitudes . Whole-genome sequence from a DU Mongolian shows that approximately 300 protein-coding changes are specific to this individual . Our analyses provide new perspectives on genetic variation and adaptation to extreme environments . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"natural",
"selection",
"genetics",
"population",
"genetics",
"biology",
"genomics",
"evolutionary",
"biology",
"genomic",
"evolution",
"evolutionary",
"genetics"
] | 2013 | Genomic Analysis of Natural Selection and Phenotypic Variation in High-Altitude Mongolians |
Burkholderia pseudomallei is a water and soil bacterium and the causative agent of melioidosis . A characteristic feature of this bacterium is the formation of different colony morphologies which can be isolated from environmental samples as well as from clinical samples , but can also be induced in vitro . Previous studies indicate that morphotypes can differ in a number of characteristics such as resistance to oxidative stress , cellular adhesion and intracellular replication . Yet the metabolic features of B . pseudomallei and its different morphotypes have not been examined in detail so far . Therefore , this study aimed to characterize the exometabolome of B . pseudomallei morphotypes and the impact of acute infection on their metabolic characteristics . We applied nuclear magnetic resonance spectroscopy ( 1H-NMR ) in a metabolic footprint approach to compare nutrition uptake and metabolite secretion of starvation induced morphotypes of the B . pseudomallei strains K96243 and E8 . We observed gluconate production and uptake in all morphotype cultures . Our study also revealed that among all morphotypes amino acids could be classified with regard to their fast and slow consumption . In addition to these shared metabolic features , the morphotypes varied highly in amino acid uptake profiles , secretion of branched chain amino acid metabolites and carbon utilization . After intracellular passage in vitro or murine acute infection in vivo , we observed a switch of the various morphotypes towards a single morphotype and a synchronization of nutrient uptake and metabolite secretion . To our knowledge , this study provides first insights into the basic metabolism of B . pseudomallei and its colony morphotypes . Furthermore , our data suggest , that acute infection leads to the synchronization of B . pseudomallei colony morphology and metabolism through yet unknown host signals and bacterial mechanisms .
Colony morphology variants are described for many bacterial pathogens that are able to induce pneumonia including Pseudomonas aeroginosa [1] Staphylococcus aureus [2] and Burkholderia cepacia complex [3] . It is also a long known phenomenon of the Gram negative water and soil proteobacterium Burkholderia pseudomallei , which is the causative agent of melioidosis [4 , 5] . This disease occurs predominantly in Northern Australia , Southeast Asia , China and Taiwan but additional cases and environmental isolates of B . pseudomallei have been reported from several regions between latitude 20°N and 20°S [6–8] . Clinical manifestations are highly diverse including soft tissue lesions , abscess formation , sepsis and pneumonia , with the latter being the most frequent clinical presentation of this disease [4 , 8] . The isolation of isogenic B . pseudomallei morphotypes out of patients indicates a significant role for these morphotypes in adaptation during human melioidosis [9] . In vitro studies have shown that the appearance of various B . pseudomallei morphotypes can be linked to starvation stress , iron limitation , growth temperature and presence of antibiotics [9] . However a clear connection between morphotype formation and metabolism as described for S . aureus or Pseudomonas fluorescence has not been established so far [10–12] . Functional studies indicate that infection relevant parameters like adhesion and intracellular replication differ between the morphotypes [9 , 13] . Additionally , a higher susceptibility to reactive oxygen species , overexpression of the arginine deaminase system and flagellin was observed [13 , 14] . However , the question , if morphotype formation affects the bacterial metabolism should be addressed , since in many intracellular pathogens metabolism is closely connected to virulence and depends on host derived nutrients ( reviewed in [15 , 16] ) . Metabolic adaptation of B . pseudomallei to the host environment includes the expression of metabolic genes for alternative carbon sources and the downregulation of TCA-cycle , glycolysis and oxidative phosphorylation [17 , 18] . Yet gene expression data only can indicate metabolic activities and to our knowledge actual metabolic investigations have not been carried out on B . pseudomallei so far . We therefore aimed in this study to gain a deeper insight into the diversity of colony morphology variants of B . pseudomallei on a metabolic level . Thus we investigated the uptake of amino acids , glucose and other carbon sources as well as the secretion of metabolites into the extracellular space . Thereby , we found so far unknown shared metabolic characteristics of B . pseudomallei morphotypes and morphotype specific secretion patterns . Our intention was also to address the question whether these metabolic features are affected by the process of infection in a murine macrophage cell culture model and in an acute pneumonia mouse model . Altogether our data indicate that , despite some variation , metabolic principles are shared among B . pseudomallei colony morphology variants and that the acute infection event synchronizes colony morphology as well as metabolism .
All the animal experiments described in the present study were conducted in strict accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health . All animal studies were conducted under a protocol approved by the Landesamt für Landwirtschaft , Lebensmittelsicherheit und Fischerei Mecklenburg-Vorpommern ( LALLF M-V; 7221 . 3–1 . 1-020/11 ) . All efforts were made to minimize suffering and ensure the highest ethical and human standards . All experiments with B . pseudomallei were carried out in biosafety level 3 ( BSL3 ) laboratories . Two different B . pseudomallei strains were used in this study . The sequenced B . pseudomallei strain K96243 was originally isolated in 1996 from a 34-year-old female diabetic patient in Thailand [19] and has been used as a reference strain in many studies [20–22] . The second strain is the yet unsequenced B . pseudomallei strain E8 , an environmental isolate from Ubon Ratchathani , Thailand [23 , 24] . Both strains were cultivated at 37°C and 140 rpm in Tryptone Soja Broth ( TSB ) to an optical density of 0 . 8 ( OD 650nm ) and cells were stored in Luria broth ( LB ) media with additional 20% glycerol at– 80°C until usage . Additionally , cells were plated on LB agar plates and incubated at 37°C for 4 days , but no different morphotypes could be detected . To generate strain specific morphotypes , different nutritional limitation conditions were used as described previously [7] with the following modifications . In brief , 10 μl of the strain K96243 and strain E8 glycerol stocks were used for the inoculation of i ) 3 ml LB media , ii ) 3 ml LB media with 5% glycerol , iii ) 3 ml Dulbecco’s Modified Eagle’s Medium ( DMEM ) , iv ) 3 ml DMEM media with 5% glycerol , v ) 3 ml RPMI-1640 ( supplemented with 15 mM citrate ) media and vi ) 3 ml RPMI-1640 ( supplemented with 15 mM citrate ) media with 5% glycerol . The cells were aerobically cultured at 37°C under static conditions for 24 days . Afterwards , serial dilutions were plated onto Ashdown agar , incubated at 37°C in air for 4 days and the different colony morphotypes were photographically documented and used for the generation of further morphotype stocks and stored at -80°C . To confirm their stability after storage at -80°C , all morphotypes were plated again on Ashdown agar and incubated for 4 days at 37°C in air , but no further morphotype switching was observed . Three colonies of every morphotype were picked and used as biological replicates in further growth experiments . For in vivo experiments , female 8 to 12-week-old BALB/c mice were purchased from Charles River ( Germany ) . BALB/c mice were housed under specific-pathogen-free conditions . All mice received 100 colony forming units ( CFU ) of B . pseudomallei strain K96243 colony morphotypes intranasally and were monitored daily after infection . After 72 hours , mice were sacrificed and the lungs were homogenized and plated onto Ashdown agar plates in appropriate dilutions . Occurring morphotypes were photographically documented and their colonies were stocked in LB with 20% glycerol at -80°C until usage . Also their stability after storage was confirmed as described above . For in vitro experiments , we used the murine macrophage cell line RAW 264 . 7 purchased from the American Type Culture Collection ( ATCC ) ( Rockville , MD ) . Briefly , RAW 264 . 7 cells were seeded in six-well plates ( 5 x 105 cells/well ) and grown for 24 hours in DMEM + 10% fetal calf serum ( FCS ) before starting the infection experiments . Cells were infected at a multiplicity of infection of ~5 of all six isolated B . pseudomallei morphotypes from strain K96243 for 30 min , washed twice with phosphate buffered saline ( PBS ) , then DMEM medium + 10% FCS containing 250 μg/ml kanamycin ( Km ) was added for 6 hours and finally the cells were further incubated for 18 h with DMEM medium + 10% FCS and 125 μg/ml Km . After a total of 24 hours , cells were washed three times with PBS before lysis with 1 ml Aqua destillata at 37°C for 15 minutes . Serial dilutions were spread plated onto Ashdown agar , incubated at 37°C in air for 4 days and occurring colony morphotypes were photographically documented and used for the generation of further in vitro morphotype stocks and stored at -80°C . Their stability after storage was confirmed as described above . All morphotypes ( obtained from nutrients starvation , in vivo and in vitro experiments ) were spread on Ashdown agar and cultivated at 37°C for 4 days . Then , relevant colonies were scraped from the plates and diluted in 700 μl PBS to an OD 650nm of about 10 . These dilutions were used for the inoculation of 50 ml RPMI-1640 media containing 15 mM citrate to an OD650 of 0 . 05 . All samples were cultured at 37°C and with vigorous agitation for 60 hours . During and after the cultivation an aliquot of the culture was plated on Ashdown agar as described above to confirm the colony morphotype . Cultivations of every morphotype and every condition were carried out as triplicates . 2 ml of cell free culture medium were taken at 5 . 5 , 8 . 25 , 11 . 5 , 16 . 5 , 30 . 5 and 60 hours by sterile filtration and directly frozen until measurement . 1H-NMR analysis was carried out as described previously [25] . In brief , 400 μl of the sample was mixed with 200 μl of a sodium hydrogen phosphate buffer ( 0 . 2 M , pH 7 . 0 ) to avoid chemical shifts due to pH , which was made up with 50% D2O . The buffer also contained 1 mM trimethylsilyl propanoic acid ( TSP ) which was used for quantification and also as a reference signal at 0 . 0 ppm . To obtain NMR spectra a 1D-NOESY pulse sequence was used with 64 FID scans with 600 . 27 MHz at a temperature of 310 K using a Bruker AVANCE-II 600 NMR spectrometer operated by TOPSPIN 3 . 1 software ( both from Bruker Biospin ) . For qualitative and quantitative data analysis we used AMIX ( Bruker Biospin , version 3 . 9 . 14 ) . We used the AMIX Underground Removal Tool on obtained NMR-spectra to correct the baseline . Thereby we used the following parameters: left border region 20 ppm and right border region -20 ppm and a filter width of 10 Hz . The region of noise , used for final baseline correction was between 5 . 5 ppm and 5 . 6 ppm . In some cases noise or unknown signals appeared in regions of integrated metabolites after these were consumed completely during cultivation . If these signals were different from the signals in our database they were regarded as false positive . These false positive signals were replaced with an adequate integral of the noise region mentioned . Absolute quantification was performed as previously described [25] . In brief , a signal of the metabolite , either a complete signal or a proportion , was chosen manually and integrated . The area was further normalized on the area of the internal standard TSP and on the corresponding amount of protons and the sample volume . Microsoft Excel 2007 was used for all final calculations and for generation of tables . The software PAST was used for the generation of principle component analysis ( PCA ) [26] . For that matter the calculated concentrations of amino acids were mean centered and autoscaled before being applied to the PCA [27] . Single values from all K96243 morphotypes and ex vivo isolates were grouped by the time point of sampling . For statistical comparison of extracellular metabolite data and growth we performed the two-way ANOVA provided by Prism ( version 6 . 01; GraphPad Software ) . Multiple comparisons were corrected by applying the Holm-Šidák approach and the level on confidence ( alpha ) was set to 0 . 01 . Bar-charts and XY-plots were also done using PRISM software . Heatmaps of extracellular amino acids were created using MeV v4 . 8 . 1 [28] with the following settings for hierarchical clustering: optimized gene leaf order , euclidean distance metric and average linkage method .
We were able to identify 6 morphotypes of the K96243 strain and 8 morphotypes of the E8 strain by plating on Ashdown agar after 24 days of starvation in 6 different media ( Fig 1A ( iii ) ) . Both strains were able to form rough and smooth colonies with variations in color and inner and outer shape ( Fig 1B ) . We found very similar morphotypes in both strains ( E8 MT01 and K96243 MT03; and E8 MT08 and K96243 MT10 ) but also unique morphologies ( E8 MT07 , E8 MT12 and K96243 MT20 ) . We used these 14 morphotypes to elucidate differences in uptake and secretion of diverse nutrients and metabolites ( Fig 1A ( vi ) ) . When we cultivated the morphotypes in modified RPMI for 60 h , no lag phase was detected ( Fig 2 ) . After 5 hours the midexponential growth phase was reached and the optical densities of morphotypes differed significantly . After 8 . 25 hours cells entered the transition phase and subsequently after 30 hours the stationary phase . K9 morphotypes reached maximum optical densities between 1 . 62 ( K96243 MT01 ) and 3 . 64 ( K96243 MT03 ) whereas E8 derived morphotypes showed a higher variation between 1 . 08 ( E8 MT12 ) and 3 . 73 ( E8 MT01 ) . After 60 hours of cultivation we observed cell aggregation in some morphotypes cultures and in accordance to that the optical density , especially in the culture of E8 MT07 and K96243 MT20 , dropped slightly . Beside glucose , citrate and myo-inositol are available in the modified RPMI medium and serve as carbon sources . Notably , citrate was present after supplementation in a concentration of 15 . 65 mM whereas myo-inositol was rather low concentrated at 0 . 29 mM . All morphotypes showed citrate uptake with some variation over time ( Fig 3A/3B ) . Unlike glucose or amino acids , citrate was not completely depleted after 60 h of cultivation but the uptake was strongly reduced after 30 h for most morphotypes . The decrease in concentration of citrate within the first 30 h of cultivation was measured between 4 . 91 mM and 9 . 55 mM and only between 0 . 65 mM and 3 . 49 mM during the second 30 hours . A significant uptake of myo-inositol was first measured 16 h or 30 h after inoculation , depending on the morphotype ( Fig 3A/3B ) . K96243 MT01 and K96243 MT10 and E8 MT08 and E8 MT12 showed the fastest myo-inositol uptake . E8 MT11 however showed only very little uptake of myo-inositol . Based on the decline in concentration , amino acids were classified in two sets ( Fig 3C ) . Set A consists of the amino acids glutamine ( gln ) , glutamate ( glu ) , aspartate ( asp ) , asparagine ( asn ) , serine ( ser ) , glycine ( gly ) and proline ( pro ) and showed a decline in concentration below detection limit within the exponential growth phase . In the culture of the morphotypes K9 MT01 and E8 MT08 the uptake of most set A amino acids was completed even before reaching midexponential phase . Set B includes aromatic amino acids ( tyrosine ( tyr ) , phenylalanine ( phe ) and histidine ( his ) ) , branched chain amino acids ( isoleucine ( ile ) , leucine ( leu ) and valine ( val ) ) , positive charged amino acids ( lysine ( lys ) and arginine ( arg ) ) , non-proteinogenic ( 4-hydroxyproline , 5-oxoproline ) and threonine ( thr ) . Amino acids of set B were consumed slower during growth and for some morphotypes almost no uptake was observed until midexponential phase ( 5 hours ) . This applies especially to thr , of which more than 96% of the initial amount was still being available in all morphotypes cultures after 5 h . Also only little amounts of arg , tyr , 4-hydroxyproline and 5-oxoproline are taken up in that time . In the late transition phase the diversity between the uptake profiles was relatively high , since some morphotypes took up amino acids of set B more efficiently than others . Interestingly , E8 MT11 showed the fastest uptake of the non-proteinogenic amino acid 5-oxoproline , whereas being rather slow in the uptake of other amino acids of set B . After 30 hours only two morphotype cultures still contained amino acids in measurable amounts ( ile; val; leu in the culture of E8 MT11 and arg in the culture of K96243 MT20 ) . However , at the end of cultivation a complete uptake of amino acids was observed for all morphotypes . The first step in branched chain amino acid ( BCAA ) degradation is the deamination via IlvE ( branched chain amino acid aminotransferase ) . We found the corresponding α-keto-acids ( leu→ 4-methyl-2-oxovalerate; ile→ 3-methyl-2-oxovalerate; val→ 2-oxoisovalerate ) to be secreted into the medium by several morphotypes in minor but distinct amounts ( Fig 4 ) . Overall , K96243 MT20 , E8 MT11 and E8 MT15 secreted the highest amounts . Notably for E8 MT11 no reuptake of these compounds was observed , whereas the other morphotypes consumed the secreted metabolites again . The reuptake for K96243 MT20 and E8 MT15 started after 30 hours of growth , when branched chain amino acids were mostly depleted , yet for E8 MT11 the uptake of branched chain amino acids was still in progress . Therefore the reuptake of α-keto-acids may be connected to the depletion of corresponding amino acids . Isovalerate , which might be a degradation product of leucine , was solely found in the culture medium of two E8 morphotypes ( MT03 and MT15 ) ( S2 Fig ) . Secretion of isovalerate started in the transition phase and no reuptake for this substance was observed in both cultures . Several other signals were appearing in the spectroscopic data and point to the secretion of other organic compounds into the medium especially , when cells enter the stationary phase . Unfortunately we were not able to identify most of these metabolites ( detailed signal pattern are summarized in S1 File , Table A ) . However , even if the signals are unknown , they can be used for a discrimination of morphotypes by their secretion pattern ( S2 Fig ) . To elucidate the effect of infection on morphotype stability and metabolic activity we performed infection experiments in mice ( Fig 1iv/2 ) and murine macrophage cell cultures ( Fig 1iv/3 ) and isolated bacteria out of the lung and the intracellular compartment of macrophages . Post in vivo and in vitro infection ( p . i . ) , the isolated K96243 morphotypes showed a homogenous rough colony , which was similar to the K9 MT03/MT02 morphotype prior to infection ( a . i . ) ( Fig 1B ) . These isolates were cultivated again in liquid modified RPMI medium for 60 hours and samples to investigate the metabolic footprint were taken . We found that , in contrast to a stationary phase , a second lower growth rate was established after the exponential phase and remained until the end of the experiment ( Fig 2 ) . Most isolates of in vitro infections were growing within 60 hours to similar optical densities between 1 . 97 and 2 . 24 , whereas MT20 grew up to 3 . 21 and MT02 only reached 0 . 79 . A slightly higher growth was determined for isolates after in vivo infection which grew to optical densities between 2 . 57 and 3 . 22 with MT02 being an exception with a maximum of only 0 . 96 . After in vitro and in vivo infection the isolates were grown again in modified RPMI medium for 60 h . The glucose concentration decreased significantly within the first 8 . 5 h ( exponential phase ) in all isolate cultures but slower compared to the a . i . cultivation ( Fig 5 ) . Whereas after cultivation for 16 hours pre infection glucose was depleted in the culture media , the lowest glucose concentration was measured in the K96243 MT02 ( ex vitro ) culture with still 4 . 97 mM . Other morphotypes were found with even higher extracellular glucose concentrations of up to 8 . 58 mM . Consequently , we detected gluconate in morphotype cultivations p . i . in slower rising concentrations ( Fig 5 ) . The maximum extracellular gluconate concentration was measured for most morphotypes p . i . after 30 h of cultivation and reached up to 9 . 00 mM . At 60 h of cultivation the concentration of gluconate decreased in all morphotype cultures p . i . except for MT18 ex vitro and MT20 ex vivo . A significant reduced combined concentration of glucose and gluconate was first measured 30 . 5 h after inoculation for all isolates except MT02 in vitro and MT18 in vitro ( lower but not significant amount ) ( Fig 5 ) . Whereas prior to infection most of the glucose and gluconate was taken up during the transition phase , in the cultivation p . i . the majority was consumed in the slow growth phase between 30 h and 60 h . In this time period the variation in glucose and gluconate consumption among the morphotypes increased significantly . Other carbon sources were used less after infection . The total amount of consumed citrate was reduced from 8 . 58 ± 0 . 74 mM in average pre infection to 3 . 52 ± 0 . 42 mM and 3 . 90 ± 0 . 65 mM post in vivo and in vitro infection respectively ( S3 Fig ) . Notably the vast majority of citrate was taken up after 16 h of growth . The highest variation in extracellular citrate concentrations was observed at the very end of the experiment . Myo-inositol remained at the initial level for most morphotypes and no uptake was observed ( S3 Fig ) . The amino acid uptake profile of cultivated morphotypes p . i . showed similarities to the morphotypes a . i . with regard to the classification of amino acids . We found a set of amino acids with a fast uptake profile ( set A: gly , ser , pro , gln , asp , asn ) and a slow uptake ( set B: arg , leu , ile , val , his , lys , thr , tyr , phe , 5-oxoproline and 4-hydroxyproline ) ( Fig 6A ) . Additionally , glutamate showed rising concentrations in the extracellular space of some morphotypes indicating a secretion of this amino acid . Subsequently , it was consumed in a fast manner . In all morphotype cultures p . i . ( in vitro and in vivo ) amino acids of set A are still present 5 h after inoculation , which was not the case before infection . Some amino acids of set B , especially branched chain amino acids , were taken up to a much lesser extend and were not depleted after 60 h of cultivation . For threonine no significant uptake could be measured until 30 h of cultivation for any morphotype . For isoleucine and arginine no uptake between mid-exponential phase and 16 h , 30 h or 60 h depending on the morphotype was observed . In general the data show , that usage of both sets was slower after infection , compared to the cultivation prior to infection ( Fig 6B ) . All K96243 morphotypes showed p . i . ( in vivo and in vitro ) a more focused use of set A amino acids during the exponential phase . After 16 . 5 hours approximately 25% of set B amino acids were used in comparison to about 75% prior to infection in the same time ( Fig 6B ) . In the later growth phase , the content of set B amino acids was decreased to 60% in the p . i . cultivation but completely consumed in the a . i . cultivation . Furthermore , the morphotypes p . i . showed very small variation in the usage of amino acids . This can be also shown when we apply the amino acid concentrations to principle component analysis ( Fig 6C ) . The PCA plot of PC1 versus PC3 shows that K96243 morphotypes a . i . form wide spread and overlapping groups which correlate with diverse amino acid uptake profiles compared to morphotypes p . i . , which group in smaller clusters and more distinct . The PCA plot of PC1 versus PC2 shows a orthogonal structure caused by the successive uptake of the two amino acid sets as indicated by the loading plot of PC2 ( S4 Fig ) . Even if the morphotypes p . i . are very similar , especially MT02 showed differences compared to the other morphotypes . After in vivo and in vitro infection , the uptake of glu , gln , leu , ile and arg was faster in MT02 than in other isolates . In fact MT02 was the only isolate , which consumed arg completely within 60 h . Additionally a faster uptake was also seen for his , asp , asn , and thr only after in vitro infections . This is indeed surprising since MT02 showed reduced growth compared to the other morphotypes with regard to the optical density . Prior to infection K96243 MT20 was found to produce the majority of α-keto-acids of all K96243 morphotypes and an additional reuptake of these compounds was observed over time . However , after infection these metabolites were minimally secreted in any morphotype culture within the first 30 h of cultivation . Isolates of K96243 MT02 were found to produce the majority of α-keto-acids starting at 11 . 5 h or 16 h respectively ( Fig 7 ) . Also no reuptake of secreted metabolites was measured . Similar to growth , amino acid uptake and carbon assimilation isolates of in vivo and in vitro infections show rather similar secretion patterns during cultivation ( S4 Fig ) .
In our study , we observed major qualitative similarities in metabolic behaviour of the investigated B . pseudomallei morphotypes like gluconate production and amino acid uptake . However , we also observed various differences in growth , uptake rates and metabolite secretion patterns supporting the idea , that beside the similarity , morphotypes exhibit different regulations on a metabolic level . Interestingly , these variations were strongly reduced , after the morphotypes were isolated from the lung of infected mice or from the intracellular environment of murine macrophages . Extracellular gluconate production was described before for other proteobacteria like Pseudomonas fluorescence , Pseudomonas cepacia , and B . cepacia and is thought to be connected to the mineral phosphate solubilizing ability of gluconate [29–31] . In a metabolic profiling study , it was shown , that clinical P . aeroginosa isolates from various cystic fibrosis patients do secrete gluconate , and that a small non-coding RNA ( CrcZ ) regulates gluconate production [32] . Extracellular gluconate is produced by a periplasmatic glucose dehydrogenase ( Gdh ) and can be further converted by a gluconate dehydrogenase ( Gad ) into 2-keto-gluconate . Gluconate or 2-keto-gluconate can then be taken up by the cells and further metabolized by either the Entner-Doudoroff-pathway ( ED-pathway ) or by the pentose-phosphate-pathway ( PP-pathway ) . Consistently , a study investigating intracellular carbon fluxes for various bacteria ( among these P . fluorescence and P . putida ) showed , that the ED-pathway is the main pathway for carbon catabolism in many bacteria [33] . Surprisingly , neither gdh , nor gad is present in the genome of B . pseudomallei and moreover , we did not detect 2-keto-gluconate in the extracellular space . Therefore , it is still unknown whether gluconate is produced by an unassigned glucose dehydrogenase in the periplasm or by an intracellular glucose dehydrogenase followed by a subsequent secretion into the medium . In the case of P . fluorescence it was described , that gluconate is taken up when glucose is depleted [33] . We also observed gluconate uptake when the extracellular glucose level reached a minimum . Yet with regard to the extracellular gluconate concentration we found two variations . Some morphotypes reached high extracellular gluconate concentrations ( 71%-88% of the provided glucose ) and some cultures contained less gluconate ( 31%-45% ) . Either gluconate production is lower in some morphotypes and glucose is taken up directly or the gluconate uptake rate is higher in these morphotypes . If direct glucose uptake is enhanced in these morphotypes they do not show a significant difference in glucose depletion compared to the other morphotypes . Therefore , we would rather suggest that gluconate uptake is elevated in these morphotypes . We found , that low extracellular gluconate concentrations correlated with higher final optical densities , suggesting a faster gluconate uptake favors growth . In general , these data imply that typical gluconate catabolizing pathways as the ED-pathway or the PP-pathway are the preferred pathways for carbohydrate catabolism instead of the Embden-Meyerhof-Parnas-pathway in B . pseudomallei . After being isolated from the intracellular environment , we found that the uptake of glucose and gluconate was strongly delayed and very similar among all isolates , which consumed mainly certain amino acids during exponential growth . For many bacteria a regulative process called carbon catabolite repression ( CCR ) is known , which regulates the hierarchy of carbon assimilation and might also play a role in B . pseudomallei . As mentioned above there are similarities between the genus Burkholderia and the close related bacterial family of Pseudomonads . Pseudomonas species exhibit a CCR-system that is different from model organisms like Escherichia coli and Bacillus subtilis . Contrary to these species , Pseudomonads prefer organic acids and amino acids as carbon sources rather than glucose or other sugar derivatives including gluconate [34 , 35] . Not only the preference for the carbon source is different , but rather the whole CCR-system . Whereas a phosphotransferase-system senses the availability of preferred carbon sources in E . coli and B . subtilis , in Pseudomonas sp . a protein called Crc acts as a global regulator controlled by CrbA , CbrB and the above mentioned small non coding RNA CrcZ [34 , 36] . Crc does not only distinguish between amino acids and sugars it is also able to establish a hierarchy of amino acid uptake [37] . Unfortunately , very little is known about the process of carbon catabolite repression in Burkholderia species . For B . cepacia the expression of a gene ( alkB ) involved in alkane degradation as a carbon source seems to be also regulated by catabolite repression [38] . A genome wide search in B . pseudomallei for CrcZ , CbrA and CbrB was without result , but we found a homolog to Crc from P . aeruginosa with 41 . 9% identity to BPSL0191 in B . pseudomallei . Therefore a similar CCR as described for P . aeruginosa is possible but still speculative in Burkholderia species . Even though the molecular mechanisms of carbon catabolite repression in Burkholderia sp . are unclear , our data indicate that carbon source uptake might be regulated in B . pseudomallei and that certain amino acids ( gln , glu , asn , asp , ser , gly , pro ) are the preferred carbon sources during exponential growth , especially after infection . Some amino acids of set B ( thr , arg , and ile ) showed not only a slower uptake over time but rather a delayed uptake , which also points to a hierarchical uptake of amino acids . Another indication towards carbon catabolite repression is the uptake of myo-inositol . It only occurred when glucose and gluconate were depleted , in opposite to citrate which was consumed mostly with other nutrients still being present . A possible explanation for the distinct regulation of nutrition uptake after infection might be an adaptation to the environment during infection . After in vivo infection , bacteria were isolated from the lung and not differentiated between intra- and extracellular bacteria , whereas in our in vitro approach , bacteria were isolated solely from the intracellular compartment . Interestingly , there was almost no difference in the metabolic behavior between ex vivo and ex vitro isolates suggesting , that the trigger for morphotype switches and changes in metabolic behavior is present in both infection models . We therefore favor the idea , that the trigger that induces morphotype switching is an intracellular signal . Especially intracellular pathogens must have the ability to use nutrients provided by the host cell to satisfy their own needs for replication , since the host-cell cytosol is not a suitable growth medium for bacteria in general [39] . Therefore intracellular pathogens like Francisella tularensis and Legionella pneumophila have developed strategies to manipulate cellular host processes like autophagy or proteasomal degradation to elevate the amount of free amino acids [40 , 41] . So far , such mechanisms are not known for B . pseudomallei , but like the two former mentioned bacteria , it also shows defects in intracellular replication in a mouse model when the stringent response , an amino acid sensing system , is defect [20] . The slower uptake of certain amino acids after infection could be explained , if , due to the low intracellular abundance in the host , these metabolites have to be synthesized and therefore uptake systems for these metabolites might be downregulated . Biosynthesis of BCAAs is essential for B . pseudomallei to establish long time persistence and the demand cannot be satisfied by host cell derived BCAAs [42] . Indeed , our data suggest that B . pseudomallei is rather used to low amounts of BCAAs since we found that some morphotypes secreted 3-methyl-2-oxovalerate , 4-methyl-2-oxovalerate and 2-oxoisovalerate during cultivation . Interestingly , the secretion of isovalerate by only two E8 morphotypes indicates morphotype specific metabolic aspects of leucine degradation , whereby the exact pathway remains unclear [43] . A secretion of degradation products of branched chain amino acids has been described previously for two S . aureus strains that were grown in RPMI medium [25] . These metabolites are either substrates of branched-chain amino acid aminotransferase ( IlvE ) during leucine , isoleucine and valine synthesis , respectively , or products of IlvE , when BCAAs are deaminated during degradation . We also could observe a reuptake of these metabolites in stationary phase , indicating that during exponential growth BCAAs are consumed in excess and cannot be used for biosynthesis or catabolism . Similar to auxothrophic mutants for BCAA-biosynthesis mutants in other pathways like purine synthesis , histidine synthesis and p-aminobenzoic acid synthesis showed attenuated virulence and growth defects in a mouse model [44] . This confirms the lack of specific nutrients inside the host . Such mutants in biosynthesis pathways of essential metabolites are of clinical importance because of their display of potential candidates for vaccination [45] . However , the metabolic adaption and morphotype switch we present in our study is caused by acute infection in a mouse pneumonia model or in a murine macrophage cell line and might only represent a short episode of adaptation . Obligate intracellular bacteria like Buchnera , Wigglesworthia and Blochmannia , which are highly adapted to the host , show extensive gene reductions of housekeeping genes and thereby strong dependences of host nutrient supply [16] . And indeed , a recent study shows that after long-term ( 12 years ) persistence biosynthetic pathways for amino acids are lost due to genome reduction in B . pseudomallei which indicates usage of host provided amino acids or redundant pathways [46] . Further research is needed to investigate i ) the biosynthesis pathways , which are required for B . pseudomallei to establish an infection and ii ) the intracellular conditions in the host cell , which allow microbial replication . Additionally to metabolic footprints , the intracellular metabolome , the fingerprint , should be examined to identify metabolic differences between B . pseudomallei morphotypes and particularly to uncover the metabolic state of the morphotype after infection . We would favor a combination of biosynthesis pathway mutants and metabolome approaches to uncover the dependence of the microbial metabolism on host derived metabolites . Overall our metabolic footprint study provides for the first time insights into the so far unknown metabolic characteristics of B . pseudomallei morphology variants . The finding of gluconate production points out , that metabolome approaches are needed to describe the metabolism of an organism , despite the availability of genomic data , since no gluconate production enzymes were assigned so far for B . pseudomallei . We identified a synchronization effect in colony morphology and metabolism due to acute infection that might play an important role in the pathogenicity of B . pseudomallei . Our metabolomic study therefore contributes to the necessary knowledge about a hazardous pathogen and its adaption to the host in the acute phase of melioidosis . | Melioidosis is a common disease in Northern Australia and East Asia , with regional mortality rates of up to 40% . Clinical manifestations range from soft tissue infections to severe sepsis . It is caused by the Gram negative saprophytic water and soil bacterium Burkholderia pseudomallei , which forms a variety of colony morphologies on solid agar . Various morphotypes appear after the bacterium is exposed to physiological stress conditions or underwent the process of infection , yet the physiological function is unclear . Metabolism is closely linked to virulence in many pathogens , and since metabolic data are not available so far for this bacterium , we monitored the nutrition uptake and metabolite secretion of B . pseudomallei morphotypes . Interestingly , despite typical genes responsible for gluconate production are missing in the B . pseudomallei genome , we observed high amounts of gluconate in the extracellular space . Furthermore , we were able to investigate metabolic differences among the morphotypes and identified synchronization in morphology and metabolism after infection as an adaptation to the host environment . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [
"carbohydrate",
"metabolism",
"medicine",
"and",
"health",
"sciences",
"chemical",
"compounds",
"pathology",
"and",
"laboratory",
"medicine",
"intracellular",
"pathogens",
"burkholderia",
"pseudomallei",
"pathogens",
"microbiology",
"carbohydrates",
"glucose",
"metabolism",
... | 2016 | Burkholderia pseudomallei Colony Morphotypes Show a Synchronized Metabolic Pattern after Acute Infection |
Hepatitis A virus ( HAV ) , the prototype of genus Hepatovirus , has several unique biological characteristics that distinguish it from other members of the Picornaviridae family . Among these , the need for an intact eIF4G factor for the initiation of translation results in an inability to shut down host protein synthesis by a mechanism similar to that of other picornaviruses . Consequently , HAV must inefficiently compete for the cellular translational machinery and this may explain its poor growth in cell culture . In this context of virus/cell competition , HAV has strategically adopted a naturally highly deoptimized codon usage with respect to that of its cellular host . With the aim to optimize its codon usage the virus was adapted to propagate in cells with impaired protein synthesis , in order to make tRNA pools more available for the virus . A significant loss of fitness was the immediate response to the adaptation process that was , however , later on recovered and more associated to a re-deoptimization rather than to an optimization of the codon usage specifically in the capsid coding region . These results exclude translation selection and instead suggest fine-tuning translation kinetics selection as the underlying mechanism of the codon usage bias in this specific genome region . Additionally , the results provide clear evidence of the Red Queen dynamics of evolution since the virus has very much evolved to re-adapt its codon usage to the environmental cellular changing conditions in order to recover the original fitness .
Non-random usage of synonymous codons is a widespread phenomenon observed in genomes from many species in all domains of life and it has been proposed that each genome has a specific codon usage signature that reflects particular evolutionary forces acting within that genome [1] , [2] . Such codon usage biases may result from mutational biases , from natural selection acting on silent changes in the genomes or both . Selection on codon usage due to codon adaptation to the tRNA pool was clearly demonstrated in Escherichia coli [3] . Later on , this premise was also confirmed in eukaryotic organisms including vertebrates [4]–[8] , supporting the hypothesis of translational selection as a driving evolutionary force . However , this hypothesis is not extensive to all organisms and in humans there is no clear evidence of translation selection as the single driving force of codon bias [9] . In this latter case many different reasons have been identified to explain codon bias: isochoric structure of GC content [10] , mRNA secondary structure selection [11] , exonic splicing enhancer selection [12] , [13] , and last but not least translation selection [14] , [15] . Additionally , there is also evidence of pressure on codon usage for the control of translation kinetics rather than for the efficiency and accuracy of translation [16] , [17] . Translation kinetics control is exerted through the use of common and rare codons , which affect the rate of ribosome traffic on the mRNA due to the longer time required for incorporating those tRNAs pairing with rare codons into the ribosome A-site . The right combination of codons allows a regulated ribosome traffic rate that temporally separates protein folding events , ensuring “beneficial” and avoiding “unwanted” interactions within the growing peptide [18] . This kind of selection , known as fine-tuning translation selection , differs from translation selection in that preferred codons are not always advantageous if the optimal folding requires a slow translation . A statistical model for translational selection measure was developed based on fully sequenced genomes from archaea , bacteria and eukarya [19] . This model shows that small genome size allows higher action of translation selection , whilst lack of tRNA gene redundancy accounts for the absence of translationally selected codons . Bearing in mind this model , it seems reasonable that translation selection might play a key role on eukaryotic viruses , which present tiny genomes and have huge numbers of tRNA genes available from their hosts . Certainly codon bias has been observed in several viruses , however , its driving force has only been studied in particular cases , such as Epstein-Barr Virus [20] and papillomavirus [21] among DNA viruses , and poliovirus ( PV ) [22] and hepatitis A virus ( HAV ) [23] among RNA viruses . While translation selection seems to be the underlying mechanism of the codon bias of those genes expressed during the productive phase in the DNA viruses , the mechanism in the RNA viruses is more variable . The codon usage bias in the capsid of PV , which presents a highly optimized codon usage with that of its host , has been proposed to be mostly the result of an additional step on translation selection , i . e . , the effect of certain codon-pair combinations on the rate of translation [24] rather than only the tRNA availability . This may be due to the fact that translational step times are influenced by the compatibilities of adjacent tRNA isoacceptor molecules on the surface of a translating ribosome [25] . Alternatively , it has also been proposed the GC dinucleotide content as the cause of codon bias in the capsid of PV [22] . However , HAV is clearly the exception to the rule . HAV presents a highly biased codon usage and mostly opposed to that of the host cell [23] . Despite lacking mechanisms of inducing cellular shutoff [26] , [27] and having a very inefficient IRES [28] , HAV is able to synthesize its proteins by adapting their codon usage to those less commonly used cellular tRNAs , resulting in a low replication rate [29] . With that naturally deoptimized codon usage the role of translation selection in shaping the HAV codon bias does not seem very obvious . HAV is the type species of the genus Hepatovirus within the family Picornaviridae , existing as a single serotype . The occurrence of highly conserved clusters of rare codons in the HAV capsid-coding region has been related to the low antigenic variability [30] , since mutations in these clusters are negatively selected even in the presence of immune pressure . Thus , a certain beneficial role of these rare codons is envisaged . Altogether , it can be concluded that codon usage plays a key role in HAV replication and evolution . To elucidate the underlying mechanisms of the naturally deoptimized codon usage of HAV , a system to cultivate the virus in a cellular environment with a modified tRNA pool was developed . This was achieved using actinomycin D ( AMD ) , which specifically inhibits the DNA-dependent RNA polymerases with no effect on the RNA-dependent RNA polymerases [31] . In these conditions of cellular transcription inhibition and hence cellular protein synthesis shut-off , the tRNA pool available for the virus is expected to increase and consequently the virus codon usage may readapt to the new conditions . HAV , being an RNA virus , replicates as a complex and dynamic mutant spectrum or swarm of non identical but very closely related individuals , called viral quasispecies [32] , [33] . The population landscape of these molecular spectra allows a more broad analysis of all ongoing mutations necessary for the study of codon usage adaptation . The response regarding the viral replicative fitness and codon usage assessed during the adaptation to AMD exclude translation selection and instead suggest fine-tuning translation kinetics selection as the underlying mechanism of the codon usage bias in the capsid coding region .
Actinomycin D ( AMD ) treatment induced a clear dose-dependent inhibition of gene expression of FRhk-4 cells with a total cytoplasmic RNA reduction of around 40% and over 80% at concentrations of 0 . 05 µg/ml and 0 . 2 µg/ml , respectively ( Fig . 1 ) . Additionally , the expression of the housekeeping genes HPRT-I and GAPDH was also significantly reduced by 62% and 80% , respectively , at 0 . 05 µg/ml of AMD and 97% and 94% , respectively , at 0 . 2 µg/ml of AMD . In contrast , neither the cellular RNA abundance nor the housekeeping gene expression was significantly affected as a consequence of HAV replication . The AMD-induced cellular shut-off resulted in a cell viability of 75% and 0 . 25% with 0 . 05 µg/ml and 0 . 2 µg/ml , respectively , at 7 days post-treatment; of around 100% and 10% , respectively , at 4 days post-treatment , and of 100% and 60% , respectively , at 2 days post-treatment . The effect of AMD-induced cellular shut-off on viral fitness was analyzed in an attempt to reproduce what occurs in other picornaviruses such as poliovirus . In the presence of 0 . 05 µg/ml of AMD , and during the early passages , HAV showed a loss in fitness ( figured as virus production per cell ) , and at 0 . 2 µg/ml of AMD concentration the virus population was completely extinct ( Fig . 2 ) . In contrast PV , which induces cellular shut-off by itself and that has an optimized codon usage , readily showed a fitness gain in the presence of 0 . 2 µg/ml of AMD ( Fig . 3 ) . To find out whether HAV could recover its fitness through a change of its codon usage , further passages in the presence of the drug were performed ( Fig . 4 ) . Replication in the presence of 0 . 05 µg/ml of AMD ( lineage 1 ) was examined for over 150 passages ( Fig . 4B ) using as baseline control the viral replication in the absence of AMD ( lineage 0 ) ( Fig . 4A ) . As aforementioned , after few passages in the presence of the drug , viral production was severely decreased , with peaks of less than 1 TCID50 per cell ( Fig . 4B ) . However , thereafter the viral population progressively adapted to AMD , giving rise again to viral progenies quantitatively equivalent to those of the population growing in absence of the drug . Additionally , after 65 passages the AMD-pre-adapted population was submitted to an increased concentration of the drug ( 0 . 2 µg/ml ) ( lineage 2 ) , and again a decrease followed by a clear recovery in the viral production was observed ( Fig . 4C ) . To further test the fitness of the populations adapted to the different AMD concentrations , viral populations were again submitted to the original conditions . For instance , once adapted to 0 . 05 µg/ml ( lineage 1 ) , the virus population was brought back to grow in the absence of the drug ( lineage 3 ) ( Fig . 4D ) . The immediate response was again an abrupt decrease in the viral progeny that , however , was very rapidly recovered . Similarly , when the 0 . 2 µg/ml AMD-adapted population ( lineage 2 ) was brought back to the original 0 . 05 µg/ml concentration ( lineage 4 ) , the same pattern of behavior of an initial loss in fitness followed by a fitness recovery was observed ( Fig . 4E ) . Nevertheless , the analysis of the viral production of the first 20 passages of the different adaptive processes showed that the kinetics of adaptation was faster during the re-adaptation to the original conditions than during the first adaptation to the different AMD concentrations , with significantly ( p<0 . 05 ) different slopes of the regression lines ( Fig . S1 ) . The Relative Codon Deoptimization Index ( RCDI ) [34] measures the adaptation of a virus codon usage to that of its host . An RCDI value of 1 indicates the virus follows the cell host codon usage , while the higher the value the higher the deviation from the host . Thus , a value of around 1 . 70 denotes that HAV posses a highly deoptimized codon usage compared to other picornaviruses whose RCDI range from 1 . 14 to 1 . 39 ( Table S1 ) . Since HAV is not able to induce cellular shut-off [26] , [27] ( Fig . 1 ) , the optimization of the viral codon usage to that of the host cell could lead to an unfair competition for tRNA resources . In contrast replication in the presence of AMD , which clearly inhibits cellular expression , may represent an environment with increased available tRNA pools . Hence , an analysis to test the adaptation to tRNA pools was performed by studying the mutant spectrum of each adapting lineage at different passages and determining the codon usage in each of these spectra . Particularly passages P4 , P5 , P20 , P36 , P38 , P41 , P44 , P65 and P85 of lineage 1 and P20 and P38 of lineage 2 ( P20' and P38' in Fig . 5 ) were analyzed . Additionally , the mutant spectrum of lineage 3 was also analyzed at P21 . In each case the quasispecies distribution of lineage 0 was used as baseline control of the molecular evolution . Three genomic regions were analyzed: two fragments from the structural polyprotein coding region ( a VP3 fragment and a VP1 fragment ) and a fragment from the polymerase coding region . Any mutation induces a codon change and the newly generated codons were classified as being similarly frequent ( within a 10% range ) , less frequent ( below 10% ) or more frequent ( above 10% ) than the original ones with respect to the cell host codon usage , as an indication of the adaptation to the cellular tRNA pool . A different pattern of molecular evolution was observed depending on the genomic region analyzed . The structural polyprotein coding regions of the AMD-adapting populations showed a tendency to progressively accumulate mutations that induced the use of codons less common than the original ones ( Fig . 5B ) , with an average of 75% of mutations in this direction in both the last passages of lineage 1 and all passages of lineage 2 . On the contrary , the new generated codons in lineage 0 were mostly similar to the original ones ( Fig . 5A ) . Particularly , 54% of the mutations gave rise to codons of the same frequency , 31% to less frequent codons and 15% to more common codons . Quite the opposite , in the polymerase region a complete dominance of mutations inducing changes in the level of frequency ( almost equally in both directions less and more frequent ) of the new codons was observed in all lineages ( Fig . 5C and 5D ) . Assuming a parallel behavior between the particular capsid coding regions analyzed and the complete capsid coding region , and applying the Poisson distribution model , it may be postulated that while at P5 , during the adaptation to 0 . 05 µg/ml of AMD , around 50% of the genomes would present only 1 mutation inducing the generation of a less common codon , from P38 and further 50% of the genomes would harbor at least 5 of such mutations . In other words , the percentage of genomes with zero mutations of this kind would progress from 20% at P5 to 0 . 20% at P85 . Since competition is established for the tRNA resources , a refined analysis involving the study of anticodon usage was performed by inferring theoretical anticodon usage tables from the actual codon usage tables which were built from the analysis of 50 molecular clones representative of a mutant spectrum ( Table S4 , A and S5 , A ) , but introducing the codon-anticodon multiple pairing effects corrected by the codon-anticodon coupling efficiency [19] . Two codon usage tables were made for each viral population at each analyzed passage , one using the mutant spectra in the VP1- and VP3-region analyzed ( Table S4 , A ) and a second one using the mutant spectrum of the 3D-region analyzed ( Table S5 , A ) . Likewise , two anticodon usage tables were built: one for the capsid region ( Table S4 , B ) and another for the polymerase region ( Table S5 , B ) . The relative anticodon usage for each amino acid family was also figured ( Tables S4 , C and S5 , C ) . Additionally , an anticodon usage table for the host cell was also built and a relative anticodon usage determined by arbitrarily giving to the most abundant anticodon in each amino acid family a value of 100% and the remaining anticodons were percentualy referred to this most abundant one . Anticodons were then sorted in three groups following this cellular anticodon usage table: anticodons used by the cell at a proportion below 20% ( <20% ) , anticodons used by the cell at a proportion between 20% and 60% ( 20%–60% ) and anticodons used by the cell at a proportion above 60% ( >60% ) . The viral relative anticodon usage variation of each population in each passage , with respect to the initial passage in the absence of the drug , was calculated ( Table S4 , D and Table S5 , D ) . Increases or decreases ( in percentage ) of use of the anticodons belonging to the above mentioned groups were analyzed and the mean variation figured ( Fig . 6 ) . Lineage 0 showed no major statistical ( p<0 . 05 ) variation of the anticodon use , with the exception of the anticodon group <20% in the capsid region at P41 , P44 and P65 which decreased ( Fig . 6 ) . In contrast , lineage 1 showed a significant ( p<0 . 05 ) and consistent tendency , in the capsid region , to decrease the anticodon group >60% and increase the anticodon group 20%–60% ( Fig . 6 ) . This tendency was further confirmed with lineage 2 at P20 and P38 ( Fig . 7 ) . Such a tendency was not observed in the polymerase region ( Fig . 6 ) . Five anticodons ( Ile: uag , uai; Thr: ugg; Val: cau , cac ) among the >60% group were responsible for the decrease of the whole group , being their decrease at P65 of lineage 1 of −5 . 6%±2 . 5% . Further on this group decreased from P20 to P38 of lineage 2 from −7 . 4%±3 . 2% to −9 . 5%±7 . 3% ( Fig . 7 ) . All these variations were significantly ( p<0 . 05 ) different from the values shown by the same anticodons in lineage 0 . On the other sense , the variation of the six anticodons ( Arg: ucc; Ile: uaa , uau; Val: cag , cai , caa ) of the 20%–60% group responsible of the increase of the whole group was also significantly ( p<0 . 05 ) different from that of lineage 0 and of 5 . 4%±4 . 0% ( P65 of lineage 1 ) , 8 . 6%±5 . 5% ( P20 of lineage 2 ) and 9 . 6%±4 . 5% ( P38 of lineage 2 ) ( Fig . 7 ) . Interestingly , when lineage 1 was returned to the original growing conditions of absence of AMD ( lineage 3 ) for 21 passages ( P21R ) , the anticodon groups >60% and 20%–60% increased and decreased , respectively . However , although the five specific anticodons of the >60% group and the six specific anticodons of the 20%–60% group increased from −6 . 3%±3 . 5% to −1 . 1%±4 . 8% and decreased from 6 . 9%±3 . 9% to 3 . 5%±2 . 9% , respectively , these variations were not statistically significant . However , six additional anticodons belonging to the >60% group ( Ala:cgg , cgi; Glu:cuc; Gly:ccc; His:gug; Pro:ggu ) did significantly ( p<0 . 05 ) increase an average of 7 . 8%±6 . 5% in comparison with the original lineage 1 ( P0R ) and four more from the 20%–60% group ( Ala:cgc; Gly:cca; His:guc; Ser:ucg ) significantly ( p<0 . 05 ) decreased an average of −6 . 30%±1 . 7% ( Fig . S2 ) . To confirm that the observations made with the molecular quasispecies analysis of the three specific regions may be inferred to the whole genome , consensus sequences at P127 of lineages 0 ( absence of AMD ) and 1 ( presence of 0 . 05 µg/ml of AMD ) , and at P62 of lineage 2 ( presence of 0 . 2 µg/ml of AMD ) were obtained . Nine mutations characterized lineage 1 . Three of them were located in the capsid region ( µ = 1 . 3×10−3 ) and 6 at the non-structural proteins region ( NSP ) ( µ = 1 . 4×10−3 ) . Lineage 2 showed 16 mutations , one at the 5′ non coding region ( µ = 1 . 4×10−3 ) , 8 in the capsid region ( µ = 3 . 4×10−3 ) and 7 at the NSP region ( µ = 1 . 6×10−3 ) . Most of the mutations occurring in the capsid region ( 63% ) induced a change to a less frequent codon , as also occurred in the mutant spectra ( 75% ) , while at the NSP region they mainly induced the change to a more frequent or similar one ( 57% and 29% , respectively ) . Furthermore , the anticodon analysis was also performed and it showed the same trend . A decrease of the anticodon group of >60% was observed and the average variation of the five main anticodons ( Ile: uag; Leu: gau; Phe: aag; Val: cau , cac ) responsible for the change of the whole group in lineage 1 was −2 . 08% and the variation of the eight main anticodons ( Ile: uag; Leu: gau; Phe: aag; Pro: ggg , ggi; Tyr: aug; Val: cau , cac ) responsible for the change in lineage 2 was −3 . 03% . Also an increase in the 20%–60% group was detected , with variations of the six main responsible anticodons ( Ile: uaa , uau; Leu: aac; Phe: aaa; Val: cag , cai ) in lineage 1 of 4 . 92% and of the ten main responsible anticodons ( Ile: uaa , uau; Leu: aac; Phe: aaa; Ser: aga , uca; Tyr: aua; Val: cag , cai , caa ) in lineage 2 of 4 . 25% . Although the population landscapes obtained with consensus sequences are less representative than those obtained with the molecular spectra , this is in some way compensated by the analysis of a wider length and the results observed support the conclusions from the molecular quasispecies analysis . Moreover , during the process of adaptation to AMD , the RCDI of the capsid coding region significantly ( p<0 . 05 ) increased ( Fig . S3 and Table S1 ) , indicating a re-deoptimization of the virus codon usage under the new growing conditions . To investigate whether codon usage plays a significant role on viral fitness in conditions of depleted and abundant tRNA pools , competition experiments between viral populations with codon usages adapted to replicate in the absence or presence of AMD were carried out . With this aim mixed populations at different quantitative ratios were grown under different conditions ( Fig . 8 ) . These experiments clearly demonstrated that lineage 1 ( adapted to grow in 0 . 05 µg/ml of AMD ) was the fittest in the presence of 0 . 05 µg/ml of AMD and as early as after 4 passages completely out-competed lineage 0 ( adapted to grow in the absence of AMD ) when mixed at equal concentrations ( Fig . 8A ) . In the most quantitatively unfavorable condition , 12 passages were required to out-compete lineage 0 ( Fig . 8B ) . On the contrary , this latter population was the fittest in the absence of the drug and clearly out-competed the drug-adapted population after 6 and 12 passages , when mixed at equal concentrations ( Fig . 8A ) and in quantitatively unfavorable condition ( Fig . 8B ) , respectively . In the presence of 0 . 2 µg/ml of AMD , lineage 2 ( adapted to grow in 0 . 2 µg/ml of AMD ) clearly showed a better fitness and rapidly out-competed lineage 1 ( Fig . 8C ) even in the most unfavorable condition that required only 9 passages ( Fig . 8D ) . However , in the presence of 0 . 05 µg/ml of AMD , lineage 1 was never able to out-compete lineage 2 and this latter lineage , although showing a relative better fitness in this condition , was unable to totally out-compete lineage 1 when mixed at equal or unfavorable ratio ( Fig . 8C and 8D ) . This relative better fitness is also evidenced by the incapacity of lineage 1 to affect lineage 2 in 0 . 05 µg/ml of AMD when the starting concentration of lineage 2 was highly favorable ( Fig . 8D ) . The effect of the Hsp90 chaperone inhibitor geldanamycin on HAV production was investigated . HAV titers were not affected by the presence of increasing concentrations of the drug ranging from 0 to 1 µM . The estimated geldanamycin concentration inducing a 50% virus titer reduction ( IC50 ) was 5 . 370 µM ( Fig . 9 ) . In contrast , PV titers were severely affected by the presence of the drug , with an estimated IC50 of 0 . 275 µM . Since in the particular case of PV , the decrease in titer is thought to be the result of an impairment of capsid folding , which is dependent on the activity of the Hsp90 [35] , it can be assumed that HAV capsid folding is not dependent on the activity of this specific chaperone and that it should depend on other factors as might be the codon usage .
Translation selection drives the optimal co-adaptation of the codon usage and tRNA concentration , i . e . the most abundant codons pair with the most abundant tRNAs , in order to get a quantitatively highly efficient and accurate rate of translation [18] . In contrast , fine-tuning translation kinetics selection also presses for a co-adaptation between codon usage and tRNA concentration but in a different sense , i . e . the use of many different rare codons pairing with non-abundant tRNAs , in order to get a locally slow ribosome traffic rate to allow the proper protein folding [18] . The codon usage of HAV is indeed highly biased and highly deoptimized [23] with the highest RCDI value among picornaviruses , and thus translation selection does not seem to be the evolutionary driving force of its codon bias . In such a situation the viral translation rate is expected to be very slow , in agreement with a highly inefficient IRES [28] , since the rate of translation is proportional to the concentration of charged tRNAs . We attempted to adapt HAV to grow in an environment with a higher tRNA availability through the specific inhibition of the cellular protein synthesis with AMD , and to study the codon usage re-adaptation , if any , to these new conditions . Although our initial hypothesis was that an environment of increased tRNA availability would result in an improvement of HAV viral translation rate , what we found during the first passages of the virus in the presence of AMD was a significant decrease of the infectious viral production per cell ( Fig . 2 ) . Nevertheless , the most striking finding was that further on ( over 40 passages ) a fitness recovery was observed ( Fig . 4 ) , in both the population adapting from 0 to 0 . 05 µg/ml ( lineage 1 ) and the population adapting from 0 . 05 to 0 . 2 µg/ml of AMD ( lineage 2 ) . In contrast , PV did not suffer a decrease in fitness during the process of adaptation to AMD and rather experienced a significant increase of virus production per cell when growing in the presence of 0 . 2 µg/ml of AMD ( Fig . 3 ) . However , PV follows a completely different strategy with a highly optimized codon usage to that of the cell and thus confirming that viral production is at its best when there is a good match between codon usage ( demand ) and tRNA availability ( supply ) . The analysis of the HAV codon usage adaptation to AMD revealed an interesting adjustment in the capsid region . If translation selection is the driving force of the codon usage , a tendency to optimize the codon usage should be expected . Instead what was detected was a re-deoptimization getting to an increased use of those uncommon tRNAs . This re-deoptimization was associated with a fitness recovery in terms of infectious virus production , suggesting that the loss of efficiency in translation would be compensated by a different capsid folding , affecting stability and/or exposure of the receptor binding site . Preliminary data point to a change in the antigenic structure and thermal stability of the viral capsid during the adaptation to AMD ( data not shown ) . Further proteomic analyses to confirm these hypotheses are in progress , but the genomic studies provide evidence suggesting that fine-tuning translation selection is actually contributing to the codon usage bias of HAV . Additional selective pressures may derive from the decrease of some cellular factors required for HAV replication and translation in conditions of cellular shut-off . Virus adjustment to this new situation may be mediated by mutations inducing changes in the RNA structure . However , it is unlikely that these mutations result in a change in RNA structure concomitant with a change in codon usage . Other alternative interpretations include AMD-associated alterations of 3D , 3C or 3CD activities although fitness recovery was not associated with mutations in these regions . Fine-tuning translation of the capsid is graphically evidenced during the back adaptation process of viral lineage 3 , from 0 . 05 µg/ml to 0 . 0 µg/ml of AMD ( Fig . S2 ) , where the population seeks a kind of dynamic equilibrium regarding the variation in the anticodon usage . Most rare codons in the capsid coding region are rare codons pairing with abundant tRNAs , while only a few of them are rare codons pairing with rare tRNAs . Many of these rare codons pairing with abundant tRNAs ( 62% ) were replaced during the process of adaptation to AMD , all of them to codons pairing with less abundant tRNAs , while those more common codons pairing with non-abundant tRNAs were replaced at a lower frequency ( 34% ) , and most of them to codons pairing with even less common tRNAs ( Table S2 ) . Most HAV capsid residues coded by rare codons are strategically located in the carboxy terminal regions of the putative highly structured elements [23] . A higher tendency of replacement of these strategically located rare codons in comparison with those located apart from these regions was observed ( Table S3 ) , indicating the potential relevance of the translation kinetics in providing locally slow ribosome traffic rates and thus contributing to the proper capsid protein folding . Actually , it is interesting that whereas the capsid folding of many picornaviruses is dependent on the activity of the heat-shock protein 90 ( Hsp90 ) chaperone [35] , that of HAV is not ( Fig . 9 ) . Additionally , the fact that substitutions detected in the 3D region during the adaptation to AMD do not tend to re-deoptimize the codon usage as occurs in the capsid region ( Fig . 6 ) together with a significantly lower mutation rate in the 5′ NCR of the population adapting to the drug ( data not shown ) , which are not under the pressure of the translational machinery , re-enforces the critical role of translation kinetics in the capsid region . Selection for fine-tuning translation kinetics in the HAV capsid acts on the whole virus population and the flattest population ( mutant spectrum ) rather than the fittest individual is selected . In fact , a blend of mutations occurring around the swarm of genomes affecting the overall codon usage was associated with fitness recovery . The critical role of the mutant spectra is also observed in the faster adaptation of the populations during the back processes from higher to lower and from presence to absence of AMD than during the forward processes , pointing to the existence of molecular memory in the quasispecies as described elsewhere [36] . Codon usage adaptation to tRNA availability must find a critical balance between the rate of translation and the proper protein folding to reach the highest fitness . While , generally , the viral populations adapted to a given tRNA pool , out-competed the non-adapted populations under those specific conditions ( Fig . 8 ) , the exception to the rule was the particular case of lineage 2 ( adapted to grow in 0 . 2 µg/ml of AMD ) in competition experiments with lineage 1 ( adapted to grow in 0 . 05 µg/ml of AMD ) in the presence of 0 . 05 µg/ml of AMD . Under these conditions , the populations were unable to out-compete each other , suggesting some kind of cooperation rather than competition ( Fig . 8 C and D ) . Although difficult to predict , it may be hypothesized that the expected slower translation rate of lineage 2 might be compensated with a higher quality capsid-folding and that the faster translation rate of lineage 1 might provide a higher level of the viral enzymes required for RNA replication and capsid maturation . Experiments are , presently , in progress to assess this point . HAV behavior in terms of mutation-selection for a fine-tuning translation kinetics allowed for fitness recovery but not fitness gain during the different processes of adaptation to the cellular tRNA changing conditions and thus it may represent an additional view of the Red Queen dynamics of protein translation [37] . At least from the viral side it is clear that in a hostile environment “it takes all the running you can do to keep in the same place” . Evolutionary adaptive changes are required to maintain fitness and cessation of change may result in extinction .
The cytopathogenic pHM175 43c strain of HAV was used for the study of HAV replication and evolution in the presence of actinomycin D ( AMD , Sigma ) . Serial passages in 0 . 0 µg/ml ( lineage 0 ) , 0 . 05 µg/ml ( lineage 1 ) and 0 . 2 µg/ml ( lineage 2 ) of AMD were carried out with a multiplicity of infection ( m . o . i . ) of 1 , at a 7-day interval . Additionally , pre-adapted populations were returned to the original conditions from 0 . 05 µg/ml to 0 . 0 µg/ml of AMD ( lineage 3 ) and from 0 . 2 µg/ml to 0 . 05 µg/ml of AMD ( lineage 4 ) . The LSc 2ab strain of poliovirus was also grown in the absence or presence of AMD and passaged every 2 days at a m . o . i . of 1 . The infectious virus titer ( TCID50 ) was obtained for both viruses in FRhK-4 cell monolayers . Virus yield per cell was figured taking in consideration the average cell viability under each condition ( between days 4 and 7 , and at day 2 , for HAV and PV , respectively ) . HAV and PV production in the presence of geldadamycin concentrations from 0 . 062 to 1 µM was evaluated in the same way . AMD-associated cytotoxicity was measured by counting viable cells using the trypan blue exclusion method . Total cytoplasmic RNA abundance from 106 cells treated with 0 . 2 µg/ml , 0 . 05 µg/ml or 0 . 0 µg/ml of AMD , and from untreated cells infected with HAV was quantified using the NanoDrop® ND-1000 spectrophotometer , as a measure of the cellular genome expression . Additionally the expression level of two cellular genes , HPRT-I ( hypoxanthine phosphoribosyl-transferase I ) and GAPDH ( glyceraldehide-3-phosphate dehydrogenase ) [38] , [39] , was also monitored by end-point dilution RT-PCR using previously described primers [40] , [41] for all aforementioned conditions . Two genomic regions of the capsid coding region were analyzed: a fragment of the VP3-coding region , corresponding to amino acids 1–123 , and a fragment within the VP1-coding region , corresponding to amino acids 85–245 . Besides , another fragment corresponding to amino acids 1–253 of the nonstructural protein 3D ( polymerase ) , was also analyzed . RT-PCR amplification of the specified RNA fragments was performed as described elsewhere [33] . Previously described primers [33] were used for the amplification of the VP3 and VP1 fragments , while for the 3D- fragment primers 3D- ( 5′ATGATTCTACCTGCTTCTCT3′ ) and 3CD ( 5′ATTGGGATCCAAGAAAATTGAAAGTCA3′ ) were designed . PCR products were cloned and the sequence from 50 molecular clones obtained as previously described [33] . Viral codon usage tables were obtained for each viral population at several passages through the analysis of the sequences of 50 molecular clones . Two codon usage tables were made , one using the sequences of the VP1- and VP3-fragments as a model for the structural proteins coding region and another using the sequences of the 3D-region as a model for the non-structural proteins coding region . Anticodon usage tables were inferred from these codon usage tables by assuming a model based on the frequency of the codons , the anticodon degeneracy and the codon:anticodon match pairing preferences [19] ( Tables S4 and S5 ) . Additionally , an anticodon usage table was also drawn for the host cell ( Tables S4 and S5 ) and anticodons sorted in those used less than 20% , those used between 20 and 60% and those used more than 60% . The relative variation of usage of each anticodon at each viral passage compared to the initial passage was calculated . The mean variation of the viral anticodons belonging to each of the previously defined groups was calculated and significant differences between the mean variations in the populations growing in the absence or presence of the drug analyzed by a T-student test . Growth competition experiments between viral lineages 0 and 1 were performed after mixing the populations at ratios of 1∶1 , 100∶1 and 1∶100 and grown in the absence or presence of 0 . 05 µg/ml of AMD . Additionally , competition experiments between lineages 1 and 2 at ratios 1∶1 , 100∶1 and 1∶100 , and grown in the presence of 0 . 05 µg/ml and 0 . 2 µg/ml of AMD were also performed . A m . o . i . of 1 was used , with the exception of those experiments in which the mixing ratios were 1∶1 in which the m . o . i . was 2 . Viral progeny of each competition experiment was passaged several times and consensus sequences obtained . To follow up the proportion of each population several genetic markers were used: two mutations that were present in the consensus sequence of lineage 1 and absent in lineage 0 ( a→g at nucleotide 2459 and a→g at nucleotide 2643 ) , and two additional mutations only present in lineage 2 ( c→u at position 1282 and c→u at nucleotide 1393 ) . These latter mutations occurred in the VP0 coding region which was sequenced using previously described primers [42] . These specific genetic markers allowed a semiquantitative monitoring of the populations through determination of the proportional height of the two peaks at each nucleotide position inferred from the chromatogram of the consensus sequences . | Each organism has a specific codon usage signature . Translational selection i . e . , selection for the codon adaptation to the tRNA pools , is one of the driving forces of codon bias . In the virus world , this implies an adjustment of the virus codon usage to that of the host cell . Hepatitis A virus appears as an exception to the rule , with a highly deoptimized codon usage , suggesting that translational selection is not the underlying mechanism of its codon bias . However , since the virus lacks a mechanism of cellular protein synthesis inhibition , the deoptimized codon usage may be envisaged as a hawk ( cell ) and dove ( hepatitis A virus ) competition strategy for tRNAs and translational selection as well . To confirm this possibility , we artificially induced cell protein synthesis shut-off , thus increasing the tRNA pool availability for the virus , and we took advantage of the quasispecies dynamics to elucidate changes in its codon usage . Virus adaptation to the drug results in a re-deoptimization of codon usage in the capsid region , suggesting a requirement of a slow translation rate , i . e . , a translation kinetic selection , instead of a translational selection associated with an optimization of the codon usage . Translation kinetics control is based on the right combination of codons ( common and rare ) that allows a regulated ribosome traffic rate ensuring the proper protein folding . Capsid folding is critical for a virus transmitted through the fecal-oral route with long extracorporeal periods . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"virology/virus",
"evolution",
"and",
"symbiosis",
"virology"
] | 2010 | Fine-Tuning Translation Kinetics Selection as the Driving Force of Codon Usage Bias in the Hepatitis A Virus Capsid |
Learning causal networks from large-scale genomic data remains challenging in absence of time series or controlled perturbation experiments . We report an information- theoretic method which learns a large class of causal or non-causal graphical models from purely observational data , while including the effects of unobserved latent variables , commonly found in many genomic datasets . Starting from a complete graph , the method iteratively removes dispensable edges , by uncovering significant information contributions from indirect paths , and assesses edge-specific confidences from randomization of available data . The remaining edges are then oriented based on the signature of causality in observational data . The approach and associated algorithm , miic , outperform earlier methods on a broad range of benchmark networks . Causal network reconstructions are presented at different biological size and time scales , from gene regulation in single cells to whole genome duplication in tumor development as well as long term evolution of vertebrates . Miic is publicly available at https://github . com/miicTeam/MIIC .
Network reconstruction methods have become ubiquitous to analyze large-scale information-rich data from the latest genomic technologies . Recently , methodological advances in the field have been seeking to learn causal relationships using time series or controlled perturbation experiments [1 , 2] . However , such strategies can be technically impracticable or costly , if not unethical , in many biological contexts . Alternatively , graphical models can be learned by simply observing enough random variations in unperturbed data , as for the reconstruction of gene regulatory networks from single-cell gene expression data . However , most methods based on this principle , such as Bayesian search-and-score [3] , sparse inverse covariance estimation [4] , maximum entropy [5] or diffusion map [6] methods , assume as underlying models either causal networks with only directed edges or non-causal networks with only undirected edges . Thus , they cannot uncover nor rule out causality in observational data . By contrast , constraint-based methods [7–10] , which identify structural constraints corresponding to all dispensable edges in a graph , can in principle uncover causality from purely observational data . Advanced constraint-based methods [9 , 10] reconstruct Markov equivalent models of a broad class of “ancestral graphs” [11] , that include undirected ( − ) , directed ( → ) and possibly bidirected ( ↔ ) edges originating from latent common causes , L , unobserved in the available data ( i . e . ⇠ L ⇢ ) . However , constraint-based methods are often not robust on small datasets and have algorithmic complexity issues when including unobserved latent variables [9–12] . Yet , latent variables are commonly found in many real applications , as in the case of an unobserved transcription factor TF co-regulating two co-expressed genes , i . e . G1 ⇠ TF ⇢ G2 ( see example of single cell transcriptomics in the Results section ) . These unobserved variables should not be ignored in practice , as they actually impact the causal relationships between observed variables , leading to spurious causal association between co-regulated genes G1 and G2 in the previous example . While the algorithmic difficulties of constraint-based methods have so far limited their applicability in practice , understanding cause-effect relationships [13] remains of primary interest to model complex biological systems and anticipate their response to environmental changes or genetic alterations . We report here an information-theoretic method , that simultaneously circumvents the complexity and robustness issues of constraint-based approaches , and demonstrate its applicability to real biological data . The method builds on an earlier information- theoretic approach [14] , in order to i ) include latent variables , a notorious conceptual and algorithmic difficulty in causal network reconstruction [9–13] , and ii ) provide an edge specific confidence assessment of retained edges , which lacks in traditional constraint-based methods . Both aspects are important in practice to reconstruct robust networks from actual biological data . The approach is applied to reconstruct causal networks from a variety of genomic datasets at different biological size and time scales , from single cells to organisms and entire phyla .
Our information-theoretic method for network reconstruction is based on the analysis of multivariate information [14–19] , I ( X; Y; Z; ⋯ ) , which extends the concept of mutual information [20] beyond two variables , I ( X; Y ) = ∑x , y p ( x , y ) log ( p ( x , y ) /p ( x ) p ( y ) ) , where p ( x ) , p ( y ) and p ( x , y ) are the measured probability distributions of single or joint variables X and Y from the available data D ( see Materials and methods ) . Most importantly , unlike two-point mutual information , I ( X; Y ) , which cannot distinguish causal from non-causal relations between variables X and Y , multivariate information involving more than two points , I ( X; Y; Z; ⋯ ) , may imply cause-effect relationships between the underlying variables , S1 File . In fact , the signature of causality in purely observational data is associated to a unique correlation pattern involving at least three variables [13 , 21]: it concerns two mutually ( or conditionally ) independent variables , I ( X; Y ) = 0 , which are therefore not connected to each other , yet both connected to a third variable Z , Fig 1A . This situation entails the orientations of a ‘v-structure’ or ‘unshielded’ collider , X → Z ← Y , because the edges XZ and YZ cannot be undirected , nor Z be a cause of X or Y , as these alternative graphical models imply correlations that would contradict independence between X and Y . V-structures are the hallmark of causality in observational data: networks with v-structures are necessary causal , while causal models without v-structures can be shown to be equivalent to their undirected counterparts from the viewpoint of observational data . Beyond v-structures , colliders may also be found in series along a collider path , as in X → Z ↔ Y ← W , Fig 1B & 1C , where the bidirected edge , Z ↔ Y , indicates that Z is not a cause of Y nor Y a cause of Z . It implies that the correlation between Z and Y is due to at least one latent common cause , L , unobserved in the available dataset , Z ⇠ L ⇢ Y , as outlined above . Hence , statistical dependencies and independencies in purely observational data can , in principle , provide structural constraints for network reconstruction as well as information on causal relationships across observed and possibly unobserved latent variables . These results underline the wealth of information which cannot be captured from two-point correlations only . The signature of causality and unobserved latent variables in multi-point correlation statistics enables to rephrase constraint-based methods [7–10] within an information-theoretic framework . Constraint-based approaches , sketched in Fig 1D , start from a fully connected network and proceed by iteratively removing dispensable edges between variables X and Y for which a conditional independence can be found , i . e . I ( X; Y|{Ai} ) = 0 ( Fig 1D , step 1 ) . This rationale of constraint-based methods can be interpreted from an information perspective [22] , using the generic decomposition of mutual information , I ( X; Y ) , relative to the set of variables {Ai} , I ( X ; Y ) = I ( X ; Y ; { A i } ) + I ( X ; Y | { A i } ) , ( 1 ) where I ( X; Y;{Ai} ) can be seen as the global indirect contribution of {Ai} to I ( X; Y ) and I ( X; Y|{Ai} ) as the remaining ( direct ) contribution ( see Eq 8 in Materials and methods ) . Conditional independence , I ( X; Y|{Ai} ) = 0 , then implies that {Ai} is a ‘separation set’ which intercepts all indirect paths contributing to the total mutual information , i . e . I ( X; Y ) = I ( X; Y; {Ai} ) . In practice , however , conditional mutual information cannot be exactly zero for finite datasets but the probability that the XY edge should be removed can be estimated from the available data as , PXY ∼ exp ( −NI ( X; Y|{Ai} ) ) , up to some normalization constant , where N is the number of independent samples ( S1 File ) . The undirected network ‘skeleton’ , resulting from the removal of all dispensable edges , is then partially directed by orienting all v-structures ( Fig 1D , step 2 ) , based on the signature of causality , outlined above , and propagating these orientations on downstream edges ( Fig 1D , step 3 ) , based on specific propagation rules consistent with ancestral graphs [23] . The main computational complexity of constraint-based methods is to uncover a valid combination of contributing nodes {Ai} for each dispensable edge XY . In absence of latent variables , the combinatorial search can be restricted to the sole neighbors of X or Y , which are sufficient to intercept all information contributions from indirect paths [7 , 8] . However , this efficient algorithm cannot be used in the presence of latent variables , as collider paths may require to extend the combinatorial search for conditioning set {Ai} to non-adjacent variables of X and Y [9] , as illustrated in Fig 1C . In practice , this intrinsic difficulty stemming from latent variables has been addressed through more complex algorithmic approaches , such as the FCI algorithm [9] and its more recent approximate variant , RFCI [10] . Beyond algorithmic complexity issues , traditional constraint-based methods are also known to be highly sensitive to sampling noise inherent to finite datasets and are not robust on typical datasets of interest ( e . g . expression data of 30 to 40 genes measured in a few hundreds to thousands of single cells [24] , see application and Fig 2 below ) . The present algorithmic approach , miic ( multivariate information-based inductive causation ) , circumvents the complexity and robustness issues of standard constraint-based methods by avoiding to directly tackle the combinatorial search of complete separation sets . Instead , it progressively collects , one-by-one , their most likely contributors , {Ai}n = {A1 , A2 , ⋯ , An} , based on a quantitative score for each pair of variables XY ( S1 File ) . The global indirect contribution is then obtained iteratively as , I ( X ; Y ; { A i } n ) = I ( X ; Y ; { A i } n - 1 ) + I ( X ; Y ; A n | { A i } n - 1 ) , ( 2 ) where I ( X; Y; An|{Ai}n−1 ) > 0 , corresponds to the contribution of the most likely nth variable An after collecting the first n−1 most likely contributors , {Ai}n−1 ( see Eq 10 in Materials and methods ) . We demonstrate in the current study that this iterative framework , which proved to be robust to sampling noise in absence of latent variables [19] , can in fact be extended to include latent variables by collecting the contributors {Ai} within the whole set of observed variables , instead of amongst the sole neighbors of X and Y in absence of latent variables [14] . This simple approach to include latent variables circumvents the algorithmic complexity of standard constraint-based methods [9 , 10] , while improving ten to hundred folds their performance in both prediction accuracy and running time , as discussed in the next section . We have assessed the performance of miic on a broad range of causal and non-causal benchmark networks from real-life as well as simulated datasets from P ≃ 30 up to 500 variables and N = 10 up to 50 , 000 independent samples ( Materials and methods ) . The causal benchmark networks , which include an increasing fraction ( 0% to 20% ) of hidden latent variables , are derived using partially observed Bayesian networks , that is , considering some variables as hidden . These unobserved variables are usually present in many real applications and cannot be ignored in practice , as they actually impact the causal relationships between observed variables , as illustrated in Fig 1B–1D . The non-causal benchmark datasets have been obtained from Monte Carlo sampling of Ising-like interacting networks sharing approximately the same two-point direct correlations with real-life benchmark causal networks but lacking causality . Monte Carlo sampling leads , however , to significant correlations between successive samples , which needs to be taken into account through an effective number of independent samples ( Materials and methods ) . Reconstructed causal networks have been compared to partial ancestral graphs ( PAGs ) [23] , which are the representatives of the Markov equivalent class of all ancestral graphs consistent with the conditional independences in the available data . In practice , benchmark PAGs have been derived by hiding some variables in benchmark directed acyclic graphs ( DAG ) using the dag2pag function of the pcalg package with slight modifications [25 , 26] . The alternative inference methods used for comparison with miic are the FCI algorithm [9] and its recent approximate variant RFCI [10] implemented in the pcalg package [25 , 26] . The results obtained with FCI and RFCI are in fact very similar and we only present here comparisons with the more recent RFCI algorithm [10] . RFCI’s results are shown for an adjustable significance level α = 0 . 01 and using the stable implementation of the skeleton learning algorithm , as well as the majority rule for the orientation and propagation steps [27] , which give overall the best results . The results have been evaluated in terms of running time , as well as , Precision ( or positive predictive value ) , Recall or Sensitivity ( true positive rate ) , and F-score , which is the harmonic mean of Precision and Recall ( Materials and methods ) . Precision , Recall and F-score have been derived for the undirected skeleton of the networks ( dashed lines in Fig 1E ) or taking into account edge orientations ( solid lines in Fig 1E ) . The results on benchmark networks are presented in Fig 1E and 1F , as well as S1 , S2 , S3 , S4 , S5 , S6 and S7 Figs . Miic outperforms classical constraint-based approaches , including its advanced approximate variant RFCI , Fig 1E , especially on networks with many underlying parameters . It achieves significantly better or comparable results with much fewer samples ( Fig 1E , S1 , S2 and S3 Figs ) , and is typically ten to hundred times faster ( Fig 1F ) . In addition , miic’s ability to learn complex ancestral networks , which require conditioning on non-adjacent variables , can be directly demonstrated on the example of Fig 1C network , S4 Fig . The complexity of miic algorithm , while difficult to evaluate exactly , proves to be linear in terms of sample size ( Fig 1F ) and quadratic in terms of network size for sparse graphs irrespective of the inclusion of latent variables ( S5 Fig ) . By contrast , traditional constraint-based methods exhibit roughly quadratic complexity in terms of sample size ( Fig 1F ) and much steeper complexity scaling in terms of network size , especially when latent variables are included [12] . Furthermore , no causality is predicted by miic for non causal datasets , even from small effective numbers of independent samples ( Materials and methods and S6 and S7 Figs ) . This underlines miic accuracy to uncover true causality . This information-theoretic method and its algorithmic implementation ( S1 Software ) are very general and can be applied to a wide range of datasets , provided a sufficient number of independent samples is available . We report here the results obtained with genomic datasets spanning a broad range of biological size and time scales from single cells and tissues to organisms and entire phyla . In addition to including latent causal variables , we have also assessed the confidence of predicted edges with an edge specific confidence ratio C X Y = P X Y / 〈 P X Y rand 〉 , where PXY is the probability to remove the XY edge , introduced above , and 〈 P X Y rand 〉 the average of the same probability after randomizing the datasets for each variable ( see Materials and methods , and S1 File section 2 . 2 for details ) . Hence , the lower CXY , the higher the confidence on the XY edge , which can be used to retain only high confidence edges in the predicted networks . Interestingly , the effect of confidence filtering on the reconstruction of benchmark networks ( S8 & S9 Figs ) demonstrates that the filtering of individual edges improves the Precision of the reconstruction ( at the expense of its Sensitivity or Recall ) not only for the network skeleton , as expected , but also for the network orientations , while retaining overall similar F-scores . This demonstrates the interest and consistency of using such confidence filtering to obtain an enhanced and tunable precision of the reconstructed networks for real biological applications . Indeed , an enhanced precision might be desirable in many practical applications for which the correctness of predicted edges is more important than the occasional dismissal of less certain edges . All network reconstructions presented in Figs 2 , 3 & 4 have been obtained with an edge specific confidence CXY < 10−3 , while network skeletons obtained before edge filtering are displayed in S11 , S14 and S15 Figs . The general three-step reconstruction scheme of miic ( i . e . Step 1- graph skeleton , Step 2- edge filtering , Step 3- edge orientation ) is also sensitive to the fine tuning of other algorithmic parameters such as the complexity criterion introduced to estimate finite size effects . All results presented in this paper have been obtained with the decomposable Normalized Maximum Likelihood ( NML ) criterion introduced in [28 , 29] , which was shown to yield significantly better results than more traditional BIC/MDL criterion on benchmark networks , especially on small datasets , leading to simultaneous improvements in both recall and precision [19] . Choosing the BIC/MDL instead of NML criterion in the three genetic network applications , Figs 2 , 3 & 4 , leads to somewhat sparser reconstituted networks including 82% to 100% of initial edges , yet no additional edges ( i . e . consistent with a lower recall ) , and 66% to 75% conserved edge orientations ( i . e . identical ‒ , → , ← and ↔ edges ) , see S1 Table . At cellular level , we reconstructed regulatory networks from single cell expression data at the time of endothelial and hematopoietic differentiations from the primitive streak cells of the mouse early embryo , Fig 2A . This concerns the formation of primitive erythroid cells , a distinct and transient red blood cell lineage arising directly from mesodermal progenitors with restricted hematopoietic potential [32] , by contrast to the highly studied definitive erythroid cells which arise from multipotent hematopoietic stem cells . The dataset for this application is from Moignard et al [24] and includes the expression of 33 transcription factors ( TFs ) along with 13 non-TF genes ( markers ) in 3 , 934 single cells extracted at 4 different times of the mouse embryo development ( days E7 . 0 , E7 . 5 , E7 . 75 and E8 . 25 ) , Fig 2A–2C and S10 Fig . The cells extracted from E8 . 25 were also divided by the authors in two different pools: potential endothelial precursors and potential hematopoietic precursors based on the expression of the Runx1 hematopoietic marker . Gene expression was collected using single cell qRT-PCR and binarized by the authors , leading to two-state ( on / off ) expression levels in the available dataset . Pooling all cells together regardless of their developmental timing ( from day E7 . 0 to E8 . 25 ) , we first analyzed their population heterogeneity using principal component analysis ( PCA ) , Fig 2B , and K-means clustering , Fig 2C . Three main cell populations are identified and can be interpreted , based on gene functional classification ( Materials and methods ) , as progenitor , endothelial precursor and hematopoietic precursor populations , whose relative proportions vary from E7 . 0 to E8 . 25 , Fig 2C . The network predicted by miic , Fig 2D , includes 75 edges with CXY < 10−3 out of 82 edges in the unfiltered skeleton , S11 Fig . The differentiation bifurcation between endothelial and hematopoietic precursors , seen through principal component ( Fig 2B ) and clustering ( Fig 2C ) analyses , also clearly appears in the reconstructed regulatory network , Fig 2D , after labelling hematopoietic specific TFs ( in red ) , endothelial TFs ( in purple ) and common TFs expressed in both precursor lineages ( in blue ) , Materials and Methods . In fact , most predicted regulatory interactions across lineage specific TFs correspond to regulatory inhibitions ( in blue ) , which might originate either from direct regulatory repressions or possibly through indirect ‘ancestor’ regulations involving unobserved intermediary TFs . In addition , a number of known regulatory interactions are correctly predicted in the inferred network , Fig 2D , such as Ikaros → Gfi1b and Ikaros → Lyl1 [31] , Tal1 → Fli1 and Tal1 → Lmo2 [32] as well as HoxB4 → Erg ( with opposite orientation ) and Sox7 → Erg [24] . Yet , there are also many predicted regulations in miic network that have not been reported so far as well as a number of regulations documented in definitive erythroid cells [32] that appear to be missing in primitive erythroid cells ( e . g . Est1 → Tal1 , Sfpi1 → Tal1 and Sfpi1 → Myb ) . These results suggest a number of testable predictions , including five bidirected edges consistent with the absence of direct regulations reported between these genes . Indeed , bidirected edges imply the necessity to invoke unobserved latent co-regulators between such genes . In particular , the unmeasured Gata2 expression is possibly implicated in the co-regulation of Erg ↔ Lyl1 , based on an earlier study [33] . Hence , beyond the consistency with earlier reports as well as testable predictions , miic results may also help pinpoint possible latent regulators unobserved in Moignard et al’s study [24] , such as regulators specific to the initial progenitor cells , not yet committed to either hematopoietic or endothelial lineages and accounting for about 70% of analyzed cells at day E7 . 0 , Fig 2C . At tissue and organismal levels , we analyzed genomic alterations on breast tumors from the online Catalog of Somatic Mutations in Cancer ( COSMIC ) datase [34] , Fig 3A–3C . The dataset , which contains 807 samples without predisposing BRCA1/2 germline mutations , includes somatic mutations ( from whole exome sequencing ) and expression level information for 91 genes . These 91 genes have been selected based on earlier studies on mutation and/or expression alterations in breast cancer , Materials and Methods . Gene non-synonymous mutation status is binarized ( yes / no ) and gene expression status is categorized as under- , normal- or over-expressed by the COSMIC database . S12 Fig provides the distribution of altered expressions and S13 Fig the distribution of mutations for the 91 genes of interest . In addition to gene mutations and altered expression levels , we also integrated information on sample average ploidy , provided by the COSMIC database ( release v76 ) and discretized the clearly bimodal ploidy distribution ( Fig 3B ) with ploidy < 2 . 7 considered as diploid cells and ≥ 2 . 7 taken as tetraploid cells , in agreement with COSMIC convention [34] . Among the 807 samples , 401 correspond to diploid tumoral cells and 398 to tetraploid tumoral cells ( 8 samples have no ploidy information ) . As expected , TP53 , RB1 and PTEN tumor suppressors tend to be mutated , downregulated or lost , especially in tetraploid tumors , Fig 3B & 3C , which also exhibit significant normalized expression alterations , Fig 3C . The network predicted by miic is shown Fig 3D . We first note that , due to the limited numbers of samples ( N = 807 ) and recurrent gene mutants ( Fig 3C and S13 Fig ) , most gene mutations are not confidently linked to any altered expression levels ( compare Fig 3D with edge confidence CXY < 10−3 to the unfiltered skeleton , S14 Fig ) , with the notable exceptions of TP53 and RB1 mutations , which have a significant impact on gene expressions , Fig 3D . Interestingly , the overall effect of tetraploidization on normalized gene expression , Fig 3C , is predicted to be largely indirect and mediated by TP53 mutations which lead to dysregulation of mitosis controling genes , such as the under-expression of PPP2R2A [35] and over-expression of AURKA and CENPA genes . In addition , tetraploidy and TP53 mutations tend also to be concomitant with over-expression of metabolic ( GMPS ) and cell-growth modulating genes ( TSPYL5 , NDRG1 and FOXM1 ) [36] , favoring tumor progression and metastasis , as well as higher expression of APOBEC3B , which promotes mutational heterogeneity within tumors and , thereby , their drug resistance through subclonal selection [37] . Hence , miic results provide a direct link between the long-known incidence of TP53 mutations in ( breast ) cancer and the tetraploidization of tumor cells . These results , supported by a number of recent reports [35 , 37–40] , shed light on the poor prognosis associated with tetraploid tumors and their resistance to chemotherapy [40] . This presumably occurs as tetraploid cells can exploit their genome redundancy and heterogeneity to evolve resistance strategies under drug treatments , Fig 3A . Interestingly , this dynamics of tetraploid tumors in the course of cancer progression and treatment echoes the success of tetraploid species in the course of eukaryote evolution . Indeed , genome doubling events , possibly associated to environmental changes , have repeatedly led to successful evolutionary radiations of biodiverse subphyla , such as the vertebrates and the flowering plants [41] , although the underlying selection mechanism has remained a matter of debate [41–44] . We have investigated with miic this long term evolution following the two rounds of tetraploidization that occurred in early vertebrates some 500 million years ago , Fig 4A . While long lost species and subphyla cannot be directly studied , the genetic make up of extant vertebrates provides an information-rich data on the selection processes at work since these ancient genome duplications . In particular , we aimed at identifying the genomic properties potentially responsible for the biaised retention of ‘ohnolog’ gene duplicates [45] retained from these genome duplications in early vertebrates . We obtained 20 , 415 protein-coding genes in the human genome from Ensembl ( v70 ) and collected information on the retention of duplicates originating either from the two whole genome duplications at the onset of vertebrates ( ‘ohnolog’ ) or from subsequent small scale duplications ( ‘SSD’ ) as well as copy number variants ( ‘CNV’ ) , Fig 4B and S1 Data [45] . 5 , 504 ohnolog genes retained from the two rounds of whole genome duplications ( WGDs ) in the common vertebrate ancestor were obtained from the ‘Ohnologs’ server based on multi-species comparison of synteny [45] . All the small scale duplicates ( SSDs ) in the human genome were obtained from Ensembl Compara using BioMart [46] , and were restricted to the 4 , 506 genes duplicated after the WGDs . Genes with copy number variants ( CNVs ) were obtained from the Database of Genomic Variants [47] . A total of 5 , 185 genes were identified to be CNV genes as their entire coding sequence fell within one of the CNV regions in this database . We then collected information on the genomic properties of these 20 , 415 human genes , including their sequence conservation ( ‘Ka/Ks ratio’ ) , protein autoinhibitory folds and participation to protein complexes , their expression levels across tissues , association with dominant or recessive diseases and susceptibility to cancer mutations as well as their essentiality for development and reproduction , see Materials and methods . The resulting causal network , predicted by miic , relates the origin of duplicated genes in the human genome ( i . e . ‘ohnolog’ , SSD or CNV gene duplicates ) to their genomic properties and association to diseases , Fig 4C . The reconstructed network implies that the retention of ohnolog duplicates is more directly linked to their susceptibility to dominant mutations and protein autoinhibitory folds than other genomic properties such as dosage balance constraints in protein complexes [42] , gene essentiality or expression levels , which do not exhibit direct links to ohnolog retention , Fig 4C , even on the network skeleton obtained before edge confidence filtering , S15 Fig . Hence , miic analysis based on observational data provides an independent confirmation as well as significant extension of earlier reports based on correlations between two or three genomic properties [43] and on simple population genetic models [48] . All together , these results support an evolutionary retention of ohnologs by purifying selection through dominant diseases in tetraploid species ( consistent with the retention of ohnologs with low Ka/Ks ratio , Fig 4C , indicating sequence conservation ) while small scale duplicated genes have been retained through positive selection ( consistent with their higher Ka/Ks ratio , Fig 4C , indicative of underlying adaptation ) .
We report in this paper a novel information-theoretic method that learns a broad class of network models including latent causal effects from purely observational data , that is , in absence of time series or controlled intervention experiments , which can be technically impractical , costly or unethical to obtain in many biological contexts . The methodology of our approach is quite general and follows a three-step scheme: While resembling traditional constraint-based methods such as FCI , miic is in fact designed to be much faster and more robust to finite sample size through greedy algorithmic strategies based on quantitative information-theoretic scores at each algorithmic step , i . e . Step 1: iterative collection of most likely contributors based on an contributor ranking scheme , Step 2: filtering of weakly supported edges through an edge-specific confidence assessment , and Step 3: successive orientation of the remaining edges based on decreasing orientation probabilities . Unlike earlier robust methods for network reconstruction [3–6] , this general scheme circumvents the need to choose between causal and non-causal graphical models a priori , as the most appropriate class of models is directly learned from the available data . In addition , the approach can uncover the effect of unobserved latent variables , a notorious conceptual and algorithmic difficulty in causal network reconstruction [13] . Yet , latent variables are usually present in many real applications and cannot be ignored in practice , as they actually impact the causal relationships between observed variables . More specifically , miic relies on the analysis of multivariate information [14–19] , which extends the concept of mutual information to more than two variables . In practice , miic integration of constraint-based methods within an information-theoretic framework leads to greatly improved performances in both prediction accuracy ( Fig 1E ) and running time ( Fig 1F ) as well as favorable scalings in terms of sample size ( Fig 1F ) and network size ( S5 Fig ) . The likelihood ratio formalism also enables to derive an edge specific confidence index , CXY , which allows to filter predicted edges to obtain an enhanced and tunable precision of the reconstructed networks . This might be desirable in many applications for which the correctness of predicted edges is more important than the occasional dismissal of less certain edges . We have used miic to reconstruct causal networks from a variety of genomic datasets at different biological size and time scales , from gene regulation in single cells ( Fig 2 ) to whole genome duplication in tumor development ( Fig 3 ) as well as long term evolution of vertebrates ( Fig 4 ) . In all these applications , miic provides testable predictions and new biological insights summarized below: Beyond the three genomic network reconstructions presented in this paper ( Figs 2 , 3 and 4 ) , we anticipate that this information-theoretic approach may help uncover cause-effect relationships in other information-rich datasets from different fields of biological interest , such as developmental biology , neuroscience , clinical data analysis and epidemiology . The causal network learning tool , miic , is implemented in an R-package software with open source code and freely available under a General Public License ( S1 Software ) . | The reconstruction of causal networks from genomic data is an important but challenging problem . Predicting key regulatory interactions or genomic alterations at the origin of human diseases can guide experimental investigation and ultimately inspire innovative therapy . However , causal relationships are difficult to establish without the possibility to directly perturb the organisms’ genome for ethical or practical reasons . Besides , unmeasured ( latent ) variables may be hidden in many genomic datasets and lead to spurious causal relationships between observed variables . We propose in this paper an efficient computational approach , miic , that overcomes these limitations and learns causal networks from non-perturbative ( observational ) data in the presence of latent variables . In addition , we assess the confidence of each predicted interaction and demonstrate the enhanced robustness and accuracy of miic compared to alternative existing methods . This approach can be applied on a wide range of datasets and provide new biological insights on regulatory networks from single cell expression data or genomic alterations during tumor development . Miic is implemented in an R package freely available to the scientific community under a General Public License . | [
"Abstract",
"Introduction",
"Results",
"Discussion"
] | [
"cancer",
"genomics",
"medicine",
"and",
"health",
"sciences",
"genetic",
"networks",
"genome",
"evolution",
"applied",
"mathematics",
"basic",
"cancer",
"research",
"human",
"genomics",
"departures",
"from",
"diploidy",
"simulation",
"and",
"modeling",
"algorithms",
... | 2017 | Learning causal networks with latent variables from multivariate information in genomic data |
Staphylococcus aureus virulence has been associated with the production of phenol soluble modulins ( PSM ) . PSM are known to activate , attract and lyse neutrophils . However , the functional characterizations were generally performed in the absence of human serum . Here , we demonstrate that human serum can inhibit all the previously-described activities of PSM . We observed that serum can fully block both the cell lysis and FPR2 activation of neutrophils . We show a direct interaction between PSM and serum lipoproteins in human serum and whole blood . Subsequent analysis using purified high , low , and very low density lipoproteins ( HDL , LDL , and VLDL ) revealed that they indeed neutralize PSM . The lipoprotein HDL showed highest binding and antagonizing capacity for PSM . Furthermore , we show potential intracellular production of PSM by S . aureus upon phagocytosis by neutrophils , which opens a new area for exploration of the intracellular lytic capacity of PSM . Collectively , our data show that in a serum environment the function of PSM as important extracellular toxins should be reconsidered .
Staphylococcus aureus frequently colonizes human anterior nares and can cause many infectious diseases , ranging from mild superficial skin and wound infections to life-threatening disseminated infections [1] . The number of infections by this bacterium is increasing , especially infections caused by methicillin-resistant S . aureus ( MRSA ) strains . However , infections are still limited to a small percentage of colonized individuals . This suggests that the human innate immune system together with physical and humoral barriers can very effectively control invasive infections , even those caused by the invasive community-associated ( CA ) MRSA . Therefore , we hypothesize that virulence factors produced by S . aureus are likely generally counteracted by the innate immune system , and that a balance between the two determines the outcome of an infection . In order to survive within the host , S . aureus can make use of a variety of virulence factors [2] , including a repertoire of toxins [3] . The toxins induce host cell lysis and include superantigens , leukocidins and phenol soluble modulins ( PSM ) . In contrast to most other toxins , PSM are small core genome-encoded peptide toxins , except for PSM-mec , which is located on the methicillin resistance-encoding MGE staphylococcal cassette chromosome SCCmec [4] . The production of PSM is controlled by the quorum-sensing accessory gene regulator ( agr ) [5] , and the gene expression levels correlate with strain virulence . Especially CA-MRSA strains are associated with high productions of PSM , which is thought to account for the enhanced virulence , easier spreading and severity of infection of CA-MRSA strains compared to hospital-acquired MRSA strains ( HA-MRSA ) [6] , [7] . Thus far , all described PSM have a common amphipathic alpha helical region , which is thought to enable their cell lytic ability most likely by disrupting the cell membrane [7] . Despite having a similar structure , PSM are categorized in two groups , depending on their size . The smaller α-type PSM ( PSMα1 , PSMα2 , PSMα3 , PSMα4 , and δ-toxin ) , with a length of 20–30 amino acids , are regarded as the most toxic PSM [7] , whereas the larger β-type PSM ( PSMβ1 and PSMβ2 ) of approximately 44 amino acids seem to have additional functions . For instance , β-type PSM of S . epidermidis are described to play a role in biofilm dispersal [8] . Next to lysing neutrophils , PSM are described to activate and attract leukocytes . Neutrophils are the first leukocytes recruited to the site of infection and are crucial in controlling staphylococcal infections . They are attracted by both host factors and conserved microbial molecules also known as pathogen-associated molecular patterns ( PAMPs ) . Although many PAMPs are recognized by Toll-like receptors ( TLRs ) [9] , PSM are potent staphylococci-specific PAMPs which act mainly on the human formylated peptide receptor 2 ( FPR2 ) [10] . FPR2 is expressed on neutrophils , monocytes , macrophages , immature dendritic cells , and microglial cells , and its activation induces many neutrophil effector functions , including chemotaxis , exocytosis and superoxide generation [11] . While micromolar concentrations of PSM are needed for neutrophil lysis , nanomolar concentrations are enough for FPR2-mediated neutrophil stimulation . Although neutrophils sense PSM at nanomolar concentrations , S . aureus can subvert FPR2 signaling by producing the antagonists FPR2 inhibitory protein ( FLIPr ) [12] and its homologue FLIPr-like [13] . S . aureus typically resides in mucosal en epithelial surfaces and can invade beyond these physical barriers causing invasive infections . The switch from a colonizing phenotype to a virulent phenotype is regulated by agr [14] . This regulatory switch relies on the secretion of autoinducing peptide ( AIP ) sensed by the cell population and triggers the expression of virulence determinants such as proteases , haemolysins and toxins . To fight the infection , the human innate immune system has several effector mechanisms , both humoral and cellular , to clear the invading bacterium . Recently , Peterson et al [15] described a novel neutralizing mechanism; incorporating ApoB1 in VLDL and LDL lipoproteins in serum can sequester AIP and thereby disable staphylococcal quorum sensing . Mice lacking plasma ApoB1 are more susceptible to invasive staphylococcal infections , implicating that ApoB1 is an essential innate defense effector against S . aureus [15] . The current study shows that serum also provides a barrier against the pro-inflammatory and cytolytic activities of PSM , the highest agr-upregulated virulence factor of S . aureus . Our main finding is that PSM-induced activation and lysis of neutrophils are greatly inhibited by human serum . Lipoprotein particles were identified as the PSM-binding and -inhibiting components within serum , suggesting that they function as scavengers of PSM , thereby preventing host damage . We show that S . aureus potentially can produce PSM inside neutrophils after phagocytosis . This opens a new area of exploration of the intracellular toxic capacity of PSM .
Inflammatory PSM activities are generally studied in normal culture medium without the addition of serum [4] , [7] , [10] , [16] . We hypothesized that constituents of human serum could inhibit S . aureus virulence by interfering with PSM activity at sites of infection . First we studied the influence of serum on the FPR2-stimulatory capacity of PSM produced by S . aureus . Therefore , we used an FPR2-transfected HL-60 cell line ( HL-60/FPR2 ) , and examined the activating capacity of culture supernatants of wild type ( WT ) MW2 strain and an isogenic agr knockout ( MW2 agr KO ) strain , which is described not to produce PSM [10] ( Figure 1A ) . Without serum , the supernatant of WT MW2 activated HL-60/FPR2 cells very potently , in contrast to the supernatant of the MW2 agr KO strain . Addition of 1% normal human serum reduced the activity of the supernatant of WT MW2 to the level induced by the MW2 agr KO supernatant . The FPR2-activation by the MW2 agr KO supernatant was not affected by the addition of human serum , suggesting that human serum specifically inhibits the PSM-induced FPR2-activation . The inhibition of PSM-induced FPR2 activation by human serum was not species specific , as the addition of mouse , rabbit or bovine serum also inhibited the activation of HL-60/FPR2 cells by the WT MW2 supernatant ( data not shown ) . Next to the FPR2-activating capacity of S . aureus culture supernatants , the MW2 supernatant has also been described to very potently lyse isolated human neutrophils [7] . Therefore , we investigated the effect of serum on the culture supernatants of the same S . aureus strains in their capacity to lyse neutrophils . Indeed , the culture supernatant of WT MW2 very potently lysed human neutrophils , as measured by the release of LDH , in contrast to the supernatant of the MW2 agr KO ( Figure 1B ) . Also in this assay , 5% human serum completely abrogated the lysis of human neutrophils induced by the culture supernatant of WT MW2 , while leaving the effect of the MW2 agr KO supernatant uninfluenced . Agr not only regulates the production of PSM , but also controls the expression of other toxins , for instance the alpha toxin gene ( hla ) . Based on the known functions of PSM , especially their FPR2 activating capacity , we hypothesize that PSM are the main effectors causing the differences in cell activation and lysis between the WT MW2 and the MW2 agr KO supernatants . Therefore , we think that PSM are specifically inhibited by human serum . As described by others [7] , [10] , we also observed that the percentage S . aureus MW2 supernatant needed for neutrophil lysis is 1000 times higher than that needed for FPR2 activation . Not only S . aureus strain MW2 , but also strains USA300 and Newman are known to produce high levels of PSM . In contrast , strains N315 and COL have been described to produce low levels of PSM [7] , [10] . When we examined the activity of the supernats of these S . aureus strains , our results correlated with the described production of PSM . FPR2-induced cell activation and neutrophil lysis were only induced by the supernatants of the high PSM-producing S . aureus strains . Importantly , these activities were also potently inhibited by human serum ( Figure S1A and S1B ) . Our findings thus suggest that human serum inhibits the FPR2-activation and neutrophil lysis induced by S . aureus produced PSM . To demonstrate that human serum indeed targets PSM present in the supernatant of S . aureus , we also tested its activity on synthetic PSM . All S . aureus core genome-encoded PSM were tested , with the exception of PSMmec , for FPR2 stimulatory activity in the presence and absence of human serum . Pre-incubation of pure synthetic PSM with 0 . 1% human serum significantly inhibited the ability of all PSM to elicit calcium mobilization in neutrophils , whereas control stimuli , fMLP , IL-8 and C5a , were not inhibited ( Figure 2A ) . The supernatant of S . aureus strain MRSA252 containing PSMmec was also inhibited by human serum . Although not examined for synthetic PSMmec , there is no reason to believe that human serum acts differently to PSMmec than to other PSM . To investigate the kinetics of serum-induced PSM-inactivation , the incubation time of PSM with serum was varied from 0 to 1800 sec , before testing in a calcium mobilization assay using HL-60/FPR2 cells . 100 nM PSMα3 was inactivated by 0 . 1% serum within seconds , whereas inactivation of a 5 times higher PSM concentration was clearly delayed ( Figure 2B ) . Higher concentrations of PSM of up to 1 µM could be fully inhibited by 15 minutes pre-incubation with 1% serum ( Figure 2C ) . These results indicate a time- and dose-dependent inhibition of PSM-induced FPR2-activation by human serum . In order to investigate the ability of human serum to inhibit PSM-induced neutrophil lysis , we screened the synthetic PSM peptides preincubated with serum for neutrophil lysis . For this , we used a concentration range of 400 nM to 100 µM for each PSM . These concentrations are biologically relevant , as Wang et al [7] have described that the CA-MRSA MW2 and USA300 strains produce δ-toxin up to 30 µM in an overnight culture . Other α-type PSM are typically produced at somewhat lower concentrations , ranging from 5 µM to 15 µM . The cytolytic activity of PSMα1 , PSMα2 , PSMα3 , and δ-toxin towards neutrophils was completely abrogated in the presence of serum ( Figure 3 ) . In addition , the lysis of peripheral blood mononuclear cells by synthetic PSMα3 was inhibited by human serum ( Figure S2A ) . To investigate whether other staphylococcal toxins may also be inhibited by human serum , we tested the effect of serum on PVL-mediated lysis . The lysis of neutrophils induced by PVL toxin was not affected by human serum ( Figure S2B ) , implicating that serum does not generally inhibit all toxin-mediated cell lysis . As shown previously [7] , α-type PSM are very potent in cytolysis; however , in our hands neutrophil lysis by β-type PSM under physiologically-relevant concentrations was not observed . Collectively , we demonstrated that serum inhibits the FPR2-activating and neutrophil lysing capacity of all S . aureus PSM . To capture the serum components able to inactivate PSM , we coated CNBr-Sepharose beads with synthetic S . aureus PSMα1 or PSMα3 and incubated the generated beads with 20% heat-inactivated serum . Following extensive washing , the specifically-bound proteins were eluted , analyzed by non-reducing SDS-PAGE and visualized by Instant Blue staining ( Figure 4A ) . Both PSMα1 and PSMα3 bound to a protein of approximately 25 kDa , which was identified by mass spectrometry as ApoA1 . Interestingly , when we performed the same experiment and additionally washed the PSM-coated beads after serum incubation with a detergent , ApoA1 was not longer detected . We were unable to detect a direct interaction of immobilized PSM with recombinant ApoA1 using ELISA or Surface Plasmon Resonance ( data not shown ) . ApoA1 is however the major protein constituent of high density lipoprotein ( HDL ) , which could possibly bind PSM . Serum lipoproteins are complex particles with a neutral core containing triglycerides and cholesterol and covered by an amphipathic monolayer of phospholipids and unesterified cholesterol . The Apo-protein components bind to the surface of the particles and are either restricted to particular lipoproteins or freely exchangeable across lipoprotein categories . Therefore , the inability of PSM to bind to ApoA1 in the presence of a detergent most likely indicates that PSM bind to the lipid content of the HDL particle rather than a specific interaction with ApoA1 . To test this hypothesis , we performed serum size exclusion assays with FITC-labeled PSMα3 ( PSMα3-FITC ) to find the PSM-binding components in serum . The FITC-labeling did not affect the function of the PSMα3 molecule , as PSMα3-FITC was as potent as unlabeled PSMα3 in the FPR2-activation of neutrophils ( Figure S3 ) . At first , the retention volume of PSMα3-FITC was determined; PSMα3-FITC behaved as an aggregate of peptides in a physiological buffer , resulting in a broad peak of approximately 50 kDa ( Figure 4B ) . In contrast , when a similar size exclusion run was performed in the presence of a detergent , a peak of monomeric PSMα3-FITC of approximately 2 , 5 kDa was observed . This is in line with the first-described discovery of PSM , where similar poly-peptide complexes were observed when PSM were extracted from S . epidermidis culture supernatant [17] . Multimeric aggregations have also been reported for S . aureus δ-toxin with molecular weights of 5–200 kDa , with the lower molecular weights observed at extreme pH or in organic solvents , and the larger molecular weights observed in water [18] , [19] . Interestingly , when PSMα3-FITC was incubated with human serum prior to the size exclusion assay , the 50 kDa peak shifted towards two peaks , one of 150 kDa and one larger than 500 kDa , suggesting an association with HDL , known to run at 150 kDa , and low density ( LDL ) or very low density ( VLDL ) lipoproteins , known to run at higher than 500 kDa . To demonstrate that indeed PSMα3-FITC associated with lipoproteins in serum , we purified HDL and LDL from serum and repeated the size exclusion experiment using PSMα3-FITC preincubated with purified HDL or LDL . When PSMα3-FITC was preincubated with HDL , detection of the FITC signal shifted from a molecular weight of 50 kDa towards 150 kDa , whereas the signal of PSMα3-FITC preincubated with LDL , which , as VLDL , represents high molecular weight particles , shifted to the void volume of the gel filtration column . Importantly , the fluorescent 150 and >500 kDa peaks of the size exclusion chromatograms of PSMα3-FITC and serum correspond and are overlapping with the peaks of the size exclusion chromatograms of PSMα3-FITC with HDL and PSMα3-FITC with LDL or VLDL . As the interactions of PSMα3-FITC with HDL , LDL or VLDL do not shift the retention times of these lipoproteins on the gel filtration column , it appears that only monomeric PSMα3-FITC molecules interact with the lipoprotein particles ( Figure S4 ) . In conclusion , several serum lipoproteins can associate with PSM . We further investigated whether serum lipoproteins , by binding PSM , are responsible for antagonizing the functions of PSM . Therefore , we depleted normal human serum for lipoproteins by density ultracentrifugation . Compared to untreated serum , lipoprotein-depleted serum lost the ability to inhibit the PSM induced lysis of neutrophils ( Figure 5A ) , indicating that lipoproteins are indeed the serum components that inhibit PSM . Next , we tested the effect of purified HDL and LDL on PSM-function . Normal HDL or LDL protein levels in serum are between 1 and 1 . 3 mg/ml . Purified HDL and LDL at physiologically relevant concentrations , representing serum levels of 1 and 10% , inhibited the PSM-mediated lysis of neutrophils ( Figure 5B ) . In addition , HDL and LDL inhibited the calcium mobilization induced by synthetic PSM ( Figure 5C ) . Addition of recombinant ApoA1 or ApoB1 did not inhibit the PSM induced activation ( Figure S5A ) or lysis ( Figure S5B ) of neutrophils . To determine the serum lipoprotein with the highest neutralizing capacity , we separated human serum by size exclusion chromatography . Subsequently , the separate fractions were tested for inhibition of a high or a low lethal dose of PSMα3 ( Figure 6 ) . At the high lethal dose ( 50 µM ) only the fractions containing HDL inhibited neutrophil lysis , whereas at a lower still lethal dose ( 10 µM ) also LDL/VLDL could inhibit the lysis of neutrophils , indicating that HDL is the most potent inhibitor within human serum . Next , human serum was spiked with PSMα2 , and HDL , LDL , VLDL and lipid-free serum were isolated by density ultracentrifugation and analyzed by HPLC ( Figure 6C ) . Approximately 80% of the spiked PSMα2 was found in the HDL-containing fractions , whereas approximately 15% and 5% of PSM was recovered in LDL- and VLDL-containing fractions , respectively . We did not detect PSM in the lipid-free serum . Finally , we investigated whether PSM could also be recovered in serum constituents when S . aureus was cultured in blood . Therefore , we cultured S . aureus MW2 overnight in freshly drawn lepirudin-anticoagulated blood , isolated the HDL fraction from the clarified blood the following day and analyzed it by HPLC/LC MS . Figure 6D shows the presence of PSMα1 , PSMα2 , PSMα3 , PSMα4 and δ-toxin in the purified HDL fraction , indicating that PSM produced by viable S . aureus could be neutralized by lipoproteins in a blood environment . Thus far , our data strongly suggest that the functional properties of PSM produced by S . aureus in serum as well as in whole blood are inhibited by lipoproteins . These results conflict with literature describing PSM as key virulence determinants for CA-MRSA , with regard to their cell lytic properties [7] . Therefore , we hypothesized that PSM rather act as toxins in an environment devoid of lipoproteins , such as the neutrophil phagosome . To test whether PSM could act intracellular and are produced by S . aureus after uptake by human neutrophils , we generated a construct in which the promoter of the PSMα operon drives GFP expression . This PSMα promoter-GFP construct was transformed into S . aureus MW2 , the bacterium was cultured over time , and the activation of the PSMα promoter was monitored with a fluorescent plate reader . As expected for an agr-controlled expression , the GFP expression was detected during the late logarithmic growth phase . Under these conditions , we did not observe expression of GFP when we introduced the construct in the MW2 agr KO strain ( data not shown ) . Next , we examined GFP expression upon phagocytosis of MW2 bacteria containing the PSMα promoter-GFP construct by neutrophils . Therefore , the bacteria were first cultured to very early logarithmic phase to prevent initial activation of the PSMα promoter . Then , they were opsonized with human serum , presented to neutrophils adherent to a flow cell , and bacterial fluorescence upon phagocytosis by neutrophils was assayed over time with a fluorescent microscope . Fluorescence within neutrophils was observed 45 min to 2 hours after phagocytosis ( Figure 7+Video S1 ) , indicating activation of the PSMα promoter and thus potential production of PSMα inside neutrophils . We did not observe fluorescence for bacteria found outside neutrophils , except when they formed dense micro-colonies ( data not shown ) . These data indicate potential PSMα expression inside neutrophils after phagocytosis of S . aureus .
Serum transports the humoral components of the innate immune system throughout the human body . Apart from the complement system and the coagulation system , the lipid transport system has been found to play a role in innate immunity as well . ApoB1 , the major protein constituent of VLDL and LDL , can sequester AIP and thereby prevent quorum sensing of S . aureus [15] . We show in this study that serum lipoproteins dampen inflammatory over-stimulation and inhibit the cytolytic activities of the agr-controlled PSM . PSM are sequestered by serum lipoprotein particles and thereby lose their functional properties , which may allow their clearance from the system , analogous to LPS sequestration by HDL [20] . This has great implications for our understanding of PSM function during infection . All classes of plasma lipoproteins are normally present in interstitial tissue fluids at approximately 6% of the plasma concentration for VLDL to 20% for HDL [21] . This interstitial lipoprotein concentration even increases in case of inflammation . Although PSM can lyse neutrophils and S . aureus can produce PSM in very large quantities [6] , [7] , the majority of these PSM will most likely rapidly be neutralized by serum lipoproteins , making it unlikely that the high concentrations required to cause neutrophil lysis can be reached . Therefore , we propose that the role of PSM as important secreted extracellular toxins should be reconsidered . We do not rule out a possible role for the membrane-lytic effects of PSM within neutrophils after phagocytosis of S . aureus . Intracellular lysis could account for the fact that experimental mouse studies with PSM knockout S . aureus strains show reduced virulence [4] , [7] . It has recently been described that the concentration of AIP , responsible for gene transcription upon quorum sensing , can reach the critical concentration within cells , allowing the agr system to function within the neutrophil [22] , [23] . In line with these studies , we show that the promoter for the PSMα operon is activated after phagocytosis of S . aureus by neutrophils . This strongly suggests that PSMα cytolytic peptides are produced within the neutrophil phagosome and may have a function there . Since the neutrophil phagosome is devoid of lipoprotein particles , PSM may act intraphagosomally and allow S . aureus to escape from the phagosome and thereby avoiding killing . Similarly , staphylococcal alpha toxin ( Hla ) , has been demonstrated to allow endosomal escape after phagocytosis [24] , [25] . However , further research is needed to test whether PSM have a similar mechanism of action . Next to the neutrophil phagosome , other niches within the human body lacking serum lipoproteins may allow for extracellular PSM functions . PSM are thought to lyse cells by disrupting the cell membrane . S . aureus delta toxin is proposed to form a cation-selective membrane pore with a central hydrophilic channel by the multimerization of 6 monomers and an outer hydrophobic interaction with membrane lipid [18] , [26] . The intrinsic structural properties of the amphipathic alpha helix of PSM likely mediate a similar multimerization in response to a lipid bilayer . This implies that PSM randomly insert into lipid membranes without specific targeting of host cells . In line with this , PSMγ and PSMδ from S . epidermidis have been shown to perforate synthetic POPC/POPG ( ( 1-palmitoyl-2-oleoyl-sn-glycero-3-phosphocholine ) / ( 1-palmitoyl-2-oleoyl-sn-glycero-3-[phospho-rac- ( 1-glycerol ) ] ) lipid vesicles [27] . PSM thus seem to target lipid layers without the presence of membrane proteins , suggesting that there is no species specificity . This is in contrast to other staphylococcal cytolytic toxins , such as PVL , which show strong human specificity [28] . Staphylococcal two-component toxins are proposed to induce receptor-mediated lysis , which explains why they are not inhibited by human serum lipoproteins . PSM seem to have high affinity for lipids . Therefore , it is not surprising that their biological actions are inhibited by the major humoral lipid transportation system , the lipoprotein particles . The two most abundant serum lipoprotein particles are HDL and LDL . Although the normal serum concentration of LDL ( ApoB1 1 . 6 mg/ml ) is higher than the HDL ( ApoA1 1 . 2 mg/ml ) , the combined surface of the HDL particles exceeds 3 times the surface of LDL particles . The surface size difference therefore most likely accounts for the higher inhibitory potential of HDL compared to LDL . As we clearly show that PSM not only interact with HDL , but also with LDL and VLDL , we argue that the Apo proteins within the lipoprotein particles do not play a specific role in binding the PSM . Although we tried , we could indeed not detect a direct interaction between purified APO-proteins and synthetic PSM . In addition , PSM are unable to bind to lipoprotein particles in the presence of a detergent , and PSM can interact with synthetic lipid vesicles [27] . Therefore , we expect that PSM bind to the lipid contents of the lipoprotein particles without the need for a specific interaction with Apo proteins within the complexes . PSM have the capacity of attracting neutrophils in the nanomolar range , exceeding their lytic capacity over a 1000 fold . There is increasing evidence that S . aureus can survive inside neutrophils [29] and that intracellular survival contributes to pathogenesis [30] . Neutrophil attraction to the site of infection may thus possibly be advantageous for the bacterium . On the one hand , it has been demonstrated that neutrophils are necessary to control the infection; to subvert neutrophil attraction through FPR2 S . aureus can also secrete two FPR2 antagonists , FLIPr or FLIPr-like [12] , [13] and prevent excessive PSM-induced neutrophil migration . The higher PSM production of CA-MRSA strains , as compared to HA-MRSA strains , might however tilt the balance in favour of neutrophil migrating towards the infection side . Then , a higher PSM production inside the neutrophils upon S . aureus phagocytosis may rescue CA-MRSA from phagosomal killing , contributing to its enhanced virulence . Serum lipoproteins have a well-known function in dampening immune responses . As for PSM , lipopolysaccharide ( LPS ) and lipoteichoic acid can be inactivated by human serum lipoproteins [31]–[33] . Also comparable to PSM , LPS forms large aggregates in aqueous conditions . Moreover , both PSM and LPS are taken up by HDL as monomeric molecules , resulting in their inactivation . The process of LPS- and PSM-inactivation by HDL displays different kinetics , since it takes hours for HDL to take up LPS [34] , whereas we show uptake of PSM by HDL within seconds . For LPS , the serum components LPS-binding protein ( LBP ) and soluble CD14 ( sCD14 ) catalyze the process of LPS transfer . Both LBP and sCD14 bind to the toxic Lipid A moiety of LPS and facilitate the transfer of monomeric LPS to HDL or to cellular expressed CD14 , resulting in inactivation of LPS and activation of the LPS receptor TLR4 , respectively . Thus far , there is no evidence that a similar serum transfer system exists for the transfer of PSM to lipoproteins or the recognition of PSM by FPR2 . Our data show that isolated HDL , without the addition of other serum components , is sufficient for PSM transfer to HDL . Additionally , we show no enhancement of the FPR2 activating capacity of PSM in the presence of lipoprotein deficient serum , as compared to buffer only . These results strongly suggest that no serum component is needed for the PSM inactivation by lipoproteins or the PSM recognition by FPR2 . High production of PSM by CA-MRSA strains is proposed as the causative factor for the enhanced virulence of CA-MRSA strains , as compared to HA-MRSA strains [6] , [7] . Our current study shows strong interaction and neutralization of PSM by serum lipoproteins , even when PSM are produced by growing S . aureus in whole blood . These results strongly suggest that the contribution of PSM to the enhanced virulence of CA-MRSA strains is not due to PSM acting as extracellular toxins . PSM can only function intracellular or in other lipoprotein-free niches in the body . Low non-toxic concentrations of PSM might attract neutrophils to the site of infection enabling uptake of S . aureus . Once inside the cell , production of PSM might help S . aureus in its escape from the phagosome , aiding in its survival and virulence . The antimicrobial activity of PSM against other bacteria might also create a niche for staphylococcal colonization outside the human body . Future studies are needed to shed more light on the exact functions of PSM and their contributions to CA-MRSA virulence .
Informed written consent was obtained from all donors and was provided in accordance with the Declaration of Helsinki . Approval was obtained from the medical ethics committee of the University Medical Center Utrecht ( Utrecht , The Netherlands ) . PSM peptides were synthesized with the recently published sequences [7] by Genscript at 95% purity . PSMα1 ( MGIIAGIIKVIKSLIEQFTGK ) , PSMα2 ( MGIIAGIIKFIKGLIEKFTGK ) , PSMα3 ( MEFVAKLFKFFKDLLGKFLGNN ) , PSMα4 ( MAIVGTIIKIIKAIIDIFAK ) , PSMβ1 ( MEGLFNAIKDTVTAAINNDGAKLGTSIVSIVENGVGLLGKLFGF ) , PSMβ2 ( MTGLAEAIANTVQAAQQHDSVKLGTSIVDIANGVGLLGKLFGF ) , δ-toxin ( MAQDIISTISDLVKWIIDTVNKFTKK ) were all synthesized with an N-terminal formyl methionine residue . Peptide stocks were prepared at 2 mM dissolved in H2O except PSMα4 , which was dissolved in 50% ( v/v ) MeOH/H20 . Peptide grade TFA , and HPLC grade MeOH were purchased from Biosolve . fMLP , FITC-isomers , and C5a were obtained from Sigma Aldrich . IL-8 was purchased from PeproTech . HDL , LDL , VLDL and ApoB1 were purchased from ( Millipore ) . Apo-A1 was purchased from ( Calbiochem ) . PVL components LukS and LukF were kindly provided by Gerard Lina , Centre National de Référence des Staphylocoques , Lyon , France . In this study the Staphylococcus aureus strains COL , N315 , MRSA252 , Newman , MW2 and USA300 were used . MW2 agr knockout [35] was a kind gift of Alexander Horswill , the University Of Iowa , Iowa , USA . S . aureus strains were cultured overnight in Müller-Hinton broth ( MHB ) with shaking at 37°C . Alternatively , MW2 was grown in MHB until an OD660 of 1 . 0 . Bacteria were washed with PBS and 106 CFU/ml was added to freshly drawn whole human blood anticoagulated with 50 µg/ml lepirudin ( Refludan , Schering ) . Bacteria were cultured overnight in whole blood with shaking at 37°C . Bacterial culture supernatants or whole blood culture plasma were clarified by centrifugation , filtered through 0 . 22 µm pore size filter and stored at −20°C in aliquots until use . Human neutrophils were isolated by means of the Ficoll-Histopaque gradient method . Venous heparinized blood was diluted with an equal volume of PBS , and subsequently layered on a gradient of Ficoll ( Amersham Biosciences ) and Histopaque ( Sigma Aldrich ) . After centrifugation for 20 min at 400 g and 21°C , polymorphonuclear cells ( neutrophils ) were collected from Ficoll and Histopaque interfaces . Cells were washed with cold RPMI-1640 containing 25 mM HEPES , L-glutamine ( Biowhittaker ) , and 0 , 05% human serum albumin ( HSA; Sanquin ) ( RPMI-HSA ) . Erythrocytes were lysed by applying a hypotonic shock to the neutrophil pellet with distilled H2O for 30 sec , followed by 10× concentrated PBS to restore the isotonicity . The cells were washed and resuspended in RPMI-HSA . HL-60 cells stable transfected with the FPR2 ( HL-60/FPR2 ) , were kindly provided by F . Boulay ( Laboratoire Biochimie et Biophysique des Systemes Integres , Grenoble , France ) . Cells were cultured in RPMI-1640 supplemented with 10% fetal bovine serum ( FCS ) , 2 µm , 100 units/ml penicillin , 100 µg/ml streptomycin , and 600 µg/ml G418 . Human pooled serum was obtained from at least 20 healthy donors and stored until use at −70°C . Serum was inactivated by heating at 56°C for 20 min . Isolation of lipid free serum was performed as described [1] . Briefly , EDTA-plasma or clarified lepuridin-plasma was applied on a gradient of potassium bromide and ultracentrifuged at 166 . 000 g for 22 h at 4°C . The lipid free serum fraction was isolated from the gradient with a density heavier than 1 . 25 . The lipid fractions with a density between 1 . 063 and 1 . 210 and between 1 . 019 and 1 . 063 were used as the fractions containing HDL and LDL , respectively . Fractions were dialyzed against PBS , filtered ( 0 . 22 µm ) and stored at 4°C until use . The concentration of HDL and LDL is expressed as the equivalent concentration of cholesterol in micrograms per milliliter . The purified lipoprotein and lipid free serum fractions obtained from the clarified lepirudin plasma were subject to a HPLC analysis . The lipid free serum fraction from the EDTA-plasma was subjected to a hexane extraction to remove the remaining lipid particles from the serum . The serum fraction and hexane 3∶1 ( v/v ) were incubated for 30 min while vigorously shaking . The aqueous partition was dialyzed against PBS . The protein content was measured with a standard BCA kit and concentrated with an Amicon 10 kDa cut-off filter ( Millipore ) to adjust the protein content to the level present in normal human serum . Calcium mobilization with isolated human neutrophils and HL-60/FPR2 cells was performed as previously described [13] . For this purpose , cells ( 5×106 cells/ml ) were loaded with 2 µM Fluo-3-AM for 20 min at room temperature , protected from light with gentle agitation . The cells were washed , resuspended in RPMI-HSA ( without FCS ) to 5×106 cells/ml . Stimuli were prepared by incubating 25 µl of 10 times concentrated agonist with 25 µl 10 times concentrated heat inactivated serum , HDL , LDL or buffer for 30 min at room temperature . Before stimulation , cells were diluted to 1×106 cells/ml in a volume of 200 µl . The basal fluorescence level for Fluo-3 was monitored at 530 nm for 8 sec after which 50 µl of pre-incubated stimulus was added . The sample tube was rapidly placed back to the sample holder and the fluorescence measurement continued up to 52 sec . Cells were gated based on scatter parameters to exclude cell debris and the mean fluorescence value at basal level was subtracted from the value at peak level ( at 30 sec ) . The different fluorescent values were expressed as percentage of the maximal response for each individual stimulus . Alternatively , various concentrations of culture supernatants or synthetic PSM , pre-incubated with 1% or 0 , 1% heat inactivated serum for 30 min , were added to Fluo-3-labeled HL-60/FPR2 cells followed by flow cytometry . For inhibition kinetics , 10 times concentrated PSM were incubated with 0 . 1% serum and at different time-points of incubation samples were added as stimuli for HL-60/FPR2 cells in the flow cytometer . Lysis of human neutrophils by filter-sterilized S . aureus culture supernatants or synthetic PSM was measured as described [36] , [37] . Clarified culture supernatants were pre-incubated with different concentrations of human serum for 10 min at room temperature . Pre-treated supernatants were transferred to a 96-wells ELISA plate ( Nunc ) containing 3×106 neutrophils in a total volume of 100 µl RPMI-HSA and were incubated for 15 min at 37°C . Neutrophil lysis was determined by release of lactate dehydrogenase ( LDH ) using the CytoTox 96 Non-Radioactive Cytotoxicity kit ( Promega ) . Alternatively , synthetic PSM were incubated for 10 min at room temperature with serum , serum gel filtration fractions or lipoproteins and subsequently tested in the neutrophil lysis assay . To study the PSM interaction with serum components , synthetic PSMα3 was labeled with fluorescein isothiocyanate ( FITC ) , by incubating 1 mg/ml PSMα3 with 100 µg/ml FITC in 0 . 1 M sodium carbonate buffer ( pH 9 . 6 ) for 1 hour at 37°C . FITC-labeled PSMα3 ( PSMα3-FITC ) was separated from unbound FITC using a HiTrap desalting column ( Amersham Biosciences ) . To determine the retention volume of PSMα3 or PSMα3-FITC alone , 100 µg/ml was loaded onto a Superdex 200 10/300GL ( GE Healthcare ) equilibrated with PBS . The interaction with serum components was studied by incubating 100 µg/ml PSMα3-FITC with 10% human serum , 2 mg/ml HDL or 1 mg/ml LDL . In some experiments , 500 µl fractions were collected after column passage and fluorescence was quantified with a platereader fluorometer ( Flexstation , Molecular Devices ) . Protein content was measured at OD280 nm and the FITC-extinction was measured at OD492 nm on an AKTA explorer ( Amersham ) . Serum and lipoproteins had minimal auto extinction at OD492 nm allowing measurement of the association of PSMα3-FITC with serum components at OD492 nm . The column was calibrated using the HMW Calibration kit ( GE Healthcare ) containing Thyroglobulin ( 669 kDa ) , Ferritin ( 440 kDa ) , Aldolase ( 158 kDa ) , Conalbumin ( 75 kDa ) and Ovalbumin ( 43 kDa ) . In some experiments , 100% heat inactivated serum was separated by gel filtration and subsequently 0 . 5 ml fractions were collected for further analysis . In other experiments , 100 µg/ml PSMα3-FITC in PBS containing 0 . 1% sodium-desoxycholate ( PBS-DOC ) was applied to the gel filtration column equilibrated with PBS-DOC . PSMα1 and PSMα3 were immobilized on CNBr-Sepharose ( Amersham ) fast flow according to the manufacturer's suggestions . The coupling density was 2 . 5 mg/ml of resin for both PSM . Following quenching of excess reactive groups with 1 M ethanolamine ( pH 8 . 0 ) for 2 h at room temperature , the affinity resins were washed and stored as 50% slurries in PBS . To test for binding to serum proteins , heat inactivated human serum was diluted 1∶20 in PBS and mixed individually with 20 µl of each affinity resin in a total volume of 0 . 5 ml . After a 30 min incubation at room temperature under vigorous agitation , the resins were pelleted by centrifugation at 3500 g for 2 min , and washed five times with 1 ml of PBS or PBS containing 0 . 1% Tween20 ( PBST ) . After the last wash , each resin was resuspended in 20 µl of 2× Laemmli sample buffer , mixed briefly , and heated at 95°C for 5 min . Following sample preparation , the proteins contained in each sample were separated by 12 . 5% SDS-PAGE and visualized by instant blue staining . For protein identification , the bands of interest were excised from the gel and subjected to in-gel proteolysis by trypsin as described by [38] . The resulting tryptic fragments were extracted , separated by capillary liquid chromatography ( LC ) , and characterized by tandem mass spectrometry ( MS/MS ) . Proteins were identified by comparing the observed fragmentation ion patterns against a data base of human proteins using the MASCOT software package . Analytical HPLC was performed using an automatic HPLC system ( Shimadzu ) with an analytical reversed-phase column , an UV detector operating at 214 nm with a flow rate of 0 . 75 mL/min . A Phenomenex Gemini C18 ( 110 Å , 5 µm , 250×4 . 6 mm ) column was used . TFA buffers ( buffer A: H2O∶MeOH , 95∶5 , v∶v; buffer B: MeOH∶H2O , 95∶5 , v∶v , both containing 0 . 1% TFA ) . Elution was effected with either a linear gradient from 100% A to 100% B over 60 min or a linear gradient from 40%A to 100%B over 45 min . The molecular mass in the respective peak was determined using electrospray mass spectrometry ( ESI-MS ) , which was performed on a Thermo Finnigan LCQ DECA XP MAX ion trap mass spectrometer and respective PSM were identified and matched in both retention time and mass synthetic PSM . δ-toxin and PSMα2 were not separated in both systems . The PSMα promoter-GFP construct was made similar to the method described previously [39] . Shortly , 270 bp upstream of the PMSα1 start codon ( excluding the SD sequence ) was amplified by PCR using Phusion polymerase ( Finnzymes ) , using the following primers 5′-AGAATTCGCATGCCTAACGTGTTATTCGTTTTAAACTTAT-3′ ) and 5′-GGATCCTCTAGATTTGCTTATGAGTTAACTTCATTGTA-3′ ( Life Technologies ) and chromosomal DNA from strain Newman as template . Purified PCR products were digested with XbaI and EcoRI ( New England Biolabs ) and ligated in the likewise digested shuttle vector pSK236-GFP-uvr [39] . The ligation mixture was introduced in E . coli Top10F′ using the CaCl2 method [40] . Colonies were checked for GFP expression using an ImageQuant LAS4000 ( GE Healthcare Life Sciences ) , and positive clones were checked by restriction analysis and sequencing of the insert . Correct plasmids were introduced into S . aureus strain RN4220 by electroporation as described by [41] , re-isolated , and introduced by electroporation in S . aureus MW2 and the MW2 Agr knockout . To measure the GFP expression of the S . aureus strains MW2 the MW2 agr KO in culture , the strains containing the PSMα promoter-GFP construct were grown in MHB with 10 µg/ml chloramphenicol . Overnight cultures were diluted 1∶10000 and grown to an OD660 of 0 . 1 . The cultures were transferred to a clear 96 well flat bottom polystyrene tissue culture plates ( Greiner ) using 150 µl culture/well . The plate was grown in a Fluostar Omega plate reader ( BMG labtech ) at 37°C with constant double orbital shaking ( 400 rpm ) in between measurements . Both the absorbance at 660 nm and GFP fluorescence ( excitation 485 nm/emission 520 nm ) were measured every 10 minutes for each well . The signal from 4 identical wells was averaged and corrected for blank wells containing only medium . To detect GFP expression in neutrophils after phagocytosis of S . aureus , an overnight culture of S . aureus MW2 , containing the PSMα promoter-GFP construct , was diluted 10000× in MHB and allowed to grow till OD660 0 . 08 . Then , the bacteria were collected by centrifugation and washed once in PBS . Bacteria were opsonized in 10% human serum in RPMI for 5′ at 37°C , washed once in PBS and resuspended in RPMI/HSA to an OD660 of 0 . 01 , which corresponds to 5×106 bacteria/ml . A three-channel flow cell for inverted microscopes ( 24×50 mm borosilicate cover glass , size 1 , 5 ( VWR International BV , The Netherlands ) mounted at the underside ) with channel dimensions of 1×4×40 mm was assembled and sterilized as described [42] . To promote adherence of neutrophils , the channels were coated with 25% human serum in RPMI by flowing in the serum , stopping the flow , clamping off the channels on both sides , and incubating overnight at 4°C . Before introduction of neutrophils , the channels were flushed with RPMI to remove unbound serum . Freshly isolated neutrophils were diluted to 5×105/ml and 200 µl was injected into each channel of the serum-coated flow cell . Neutrophils were allowed to adhere for 30 minutes at RT , after which the opsonized bacteria were injected in the channel at a ratio of 10 bacteria: 1 neutrophil . The channel was clamped on both sides , the tubes were cut and the flow cell was transferred to the microscope stage . Neutrophils and bacteria were imaged using a Leica TSC SP5 inverted microscope equipped with a HCX PL APO 40×/0 . 85 objective ( Leica Microsystems , The Netherlands ) . The microscope was encased in a dark environment chamber that was stably kept at 37°C . 15 minutes after phagocytosis images were acquired using the camera every 5 minutes in both the bright field and the GFP channel ( I3 filter cube ) for 3 hours to follow GFP production . In post processing the bright field image was not altered and the green channel was adjusted in LAS AF ( Leica ) to contrast +10 , brightness −10 and gamma 1 . 3 to reduce the green background before merging of both channels . | Infections with methicillin-resistant Staphylococcus aureus ( MRSA ) are difficult to treat because of resistance against standard antibiotics . In contrast to the traditional healthcare-associated ( HA- ) MRSA strains , community-associated ( CA- ) MRSA strains cause severe infections in otherwise healthy individuals . CA-MRSA strains display enhanced virulence , spreading more rapidly and causing more severe illness than HA-MRSA strains . Enhanced virulence of CA-MRSA is thought to be associated with the production of several toxins , such as Phenol Soluble Modulins ( PSM ) . PSM have been described to activate , attract and lyse neutrophils . Thus far , previous studies characterizing the functions of PSM were performed in the absence of body fluids . In the current study , we show that human serum strongly inhibits many functions attributed to PSM . We demonstrate that serum lipoprotein particles are responsible for the binding and inhibition of PSM , even when PSM are produced by growing S . aureus in whole blood . Finally , we show production of PSM by S . aureus within neutrophils , suggesting that PSM may play a role intracellularly in a serum-free environment . These findings significantly contribute to our understanding of the function of PSM and strongly suggest that PSM , instead of performing as extracellular toxins , most likely act as intracellular toxins . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"immunity",
"immunology",
"biology",
"microbiology",
"bacterial",
"pathogens"
] | 2012 | Inactivation of Staphylococcal Phenol Soluble Modulins by Serum Lipoprotein Particles |
Recent studies have highlighted the ability of HIV to escape from cytotoxic T lymphocyte ( CTL ) responses that concurrently target multiple viral epitopes . Yet , the viral dynamics involved in such escape are incompletely understood . Previous analyses have made several strong assumptions regarding HIV escape from CTL responses such as independent or non-concurrent escape from individual CTL responses . Using experimental data from evolution of HIV half genomes in four patients we observe concurrent viral escape from multiple CTL responses during early infection ( first 100 days of infection ) , providing confirmation of a recent result found in a study of one HIV-infected patient . We show that current methods of estimating CTL escape rates , based on the assumption of independent escapes , are biased and perform poorly when CTL escape proceeds concurrently at multiple epitopes . We propose a new method for analyzing longitudinal sequence data to estimate the rate of CTL escape across multiple epitopes; this method involves few parameters and performs well in simulation studies . By applying our novel method to experimental data , we find that concurrent multiple escapes occur at rates between 0 . 03 and 0 . 4 day−1 , a relatively broad range that reflects uncertainty due to sparse sampling and wide ranges of parameter values . However , we show that concurrent escape at rates 0 . 1–0 . 2 day−1 across multiple epitopes is consistent with our patient datasets .
During Human Immunodeficiency Virus 1 ( simply HIV hereafter ) infection , cytotoxic T lymphocyte ( CTL ) responses play a significant role in shaping viral dynamics and evolution [1] . CTLs , which are activated CD8+ T cells , identify and target HIV-infected cells by recognizing parts of viral proteins called epitopes that are displayed on the surface of infected cells in combination with MHC-I molecules [2] . However , as has been noted since the 1990s [3 , 4] , the error-prone viral reverse transcriptase generates mutations , resulting in altered proteins , and leading to the loss of recognition of infected cells by CTLs . Such mutations and their rise in frequency , referred to as CTL escape , can occur throughout infection but are especially prominent in the weeks following peak viral load [5–7] . Besides playing an important role in shaping viral dynamics and evolution , CTL escapes are a major problem in the development of an effective vaccine [1 , 8–10] . HIV-specific CTL responses first arise about 3 weeks into infection , several days prior to peak viral load , and initially target roughly 3–5 epitopes [9 , 11] . Typically , at times near peak viral load , T cell responses to 1–2 of the roughly 3–5 epitopes are dominant as measured through ELISpot or peptide-MHC tetramers , with other responses present at low levels [11 , 12] . In the weeks following peak viral load CTL response to other epitopes increase , a phenomenon we refer to as broadening [11–13] . During this time , some CTL responses decline in magnitude , often in association with the rise in frequency of mutations at the targeted epitope , while other CTL responses rise in magnitude [9 , 11 , 12] . Overall , the first two months of HIV infection are marked by several asynchronous but concurrent CTL responses , reflecting a broad CTL response [11–13] . Viral escape from CTL-mediated killing follows a temporal pattern similar to CTL response , although not all responses elicit an escape and escapes do not always occur in the same order as the CTL responses [13] . The first escape mutations rise to significant frequency 1–3 weeks after peak viral load and are usually restricted to the 1–2 epitopes targeted by the dominant CTL responses around the time of peak viral load . Escape mutations at other epitopes reach significant frequencies later , about 4–6 weeks after peak viral load [12–15] . Previous studies have measured the rate of CTL escape as a means of quantifying the strength of CTL response [12 , 15–20] . The escape rate is the difference between the growth rates of the escape variant and wild type populations or , more explicitly , the difference between the slope of the escape variant and wild type ln-VL curves [19 , 21] , where ln-VL denotes the natural logarithm of the viral load attributed to the respective variants . From a population genetics perspective , the escape rate is the selective advantage of the escape variant relative to the wild type [22] . When escape mutations do not reduce replicative fitness , which we define as the growth rate of the virus in the absence of CTL response , the selective advantage of the escape variant is typically attributed to CTL killing , making the escape rate equal to the rate at which an infected cell is killed by CTLs targeting the given epitope [19 , 21 , 23] . When escape mutations do reduce replicative fitness , the CTL kill rate is greater than the escape rate . Thus , the escape rate provides a quantitative estimate of the in vivo strength of CTL response . A common approach for estimating escape rates , introduced in [19 , 21 , 23] and which we refer to as the logistic model , considers escape dynamics at each epitope separately , a simplification that may miss correlations between escapes at separate epitopes and lead to statistically biased estimates . To apply the logistic model , the CTL escape must be sampled at two timepoints , with the escape rate estimate then applying to the interval of time between the first and second sample . However , in many existing datasets , including those we consider below , sampling is temporally sparse with few sample times within the first 2–3 months of infection . For many escapes the first or second sample time provides no information because the escape has not started or is already complete , and for those escapes that are ongoing at both sample times , escape rate estimates are averages over long period of times and may not reflect the dynamics of CTL escape soon after peak viral load . Previous authors have extended the logistic model to cases in which the CTL escape is not sampled at two time points by artificially setting mutant and wild type frequencies to intermediate values at sample times , leading to estimates that may be biased . Methods that estimate escape rates by considering concurrent escape at multiple epitopes have been developed [24–26] , but these methods are computationally complex and depend on highly parameterized models . In this work we consider the CTL response associated with the viral escapes that occur during the first 2–3 months of infection in four previously analyzed patient datasets from the Center for HIV/AIDS Vaccine Immunology ( CHAVI ) cohort: CH40 , CH58 , CH77 , and CH256 [12 , 13] . We analyze the escapes using single genome sequencing and amplification ( SGA ) data that provides linkage information between different epitopes . Importantly , we do not consider viral escapes that are first sampled at times past 2 months post symptoms , corresponding to roughly 3 months post infection . As a result , CTL responses associated with delayed escapes or no escape are not part of our analysis . Pandit and De Boer [27] recently noted concurrent CTL escapes based on haplotype reconstruction using deep sequencing data from one patient . Similarly , we observe concurrent CTL escape across all four patients we consider . Half-genome sequence samples from the four patients reveal common patterns of escape . Multiple variants mutated at different epitopes were sampled concurrently , reflecting concurrent CTL escape through different mutation pathways . Further , samples reveal individual viral variants with mutations at multiple epitopes , reflecting CTL escape through linked epitope mutations . Put together , escape occurred through multiple mutation pathways and linked escape occurred within these different pathways . Using specific examples and simulations , we show that estimates of escape rates based on the logistic model can lead to bias in the presence of multi-epitope escape , with the direction of bias depending on the structure of the escape . Furthermore , we show that application of the logistic model when CTL escape is sampled at a single time point leads to estimates that are severely downward biased ( i . e . , are underestimates ) . Importantly , these biases are unrelated to the presence of replicative fitness costs . Instead , bias associated with the logistic model is due to ignoring concurrent escapes and making strong assumptions regarding mutant frequency when mutants are undetectable . To address these limitations in the logistic model , we introduce a novel method for estimating escape rates that removes bias associated with multi-epitope escape by applying the logistic model to pairs of variants , thereby generalizing the notion of wild type and mutant to the setting of multiple epitopes . This novel method still suffers from bias when CTL escape is captured at a single time point , so we introduce a further extension , involving the introduction of three parameters A , tI , and μ , that allows for unbiased estimation of escape rates given sampling of CTL escape at a single timepoint . Parameter A denotes the number of mutations that occur during the escape , tI is the time at which the first escape mutation occurs , and μ is the per epitope virus mutation rate . For a given patient , different CTL escapes will be parameterized by different values for A , tI and μ that will typically be unknown . The three parameters are the price we pay for requiring only a single sample time during each escape . Through simulation , we show that when A , tI and μ are known , our estimates are not biased by multi-epitope response and single time point sampling . Although we are unable to give a narrow range for the rate at which escapes proceed , our results show that escape rates in the range of 0 . 1–0 . 2 day−1 across multiple epitopes are consistent with our dataset . Such significant escape rates across multiple epitopes would reflect a broad yet still relatively strong CTL killing . If replicative fitness costs associated with escape exist , CTL kill rates are above the range 0 . 1–0 . 2 day−1 , suggesting even stronger CTL killing . Importantly , the wide range of escape rates contained within our lower and upper bound reflects a range of model assumptions consistent with our datasets . More accurate escape rate estimates require either more model assumptions , corresponding to narrower choices for A , tI and μ , or datasets with denser temporal sampling .
Goonetilleke et . al . [12] and Liu et . al . [13] identified putative epitopes targeted by CTL response in 17 patients of the CHAVI cohort including the four patients that we analyze here . Importantly , for every patient in the CHAVI cohort , most putative epitopes with escape mutations in the first several months post infection elicited an experimentally measured epitope-specific CTL response , providing evidence that escape during acute infection is largely driven by CTL mediated selection and does not arise due to neutral evolution [9 , 12] . In each of our four patient datasets , two sample time points fall within the first 2–3 months of infection: we label these sample times t1 and t2 . To analyze early CTL escape , we consider only putative epitopes with mutation in at least one sequence collected at t1 and t2 . Further , to avoid variation unassociated with CTL response , we only consider putative epitopes that meet at least one of two criteria: 1 ) the putative epitope is supported by ELISpot assays in [12 , 13] or 2 ) the putative epitope mutated into multiple haplotypes across the sequences collected at t1 and t2 ( i . e . different sequences collected at t1 and t2 had different mutations on the putative epitope ) and the putative epitope was eventually lost as infection progressed [12 , 28 , 29] . Regarding the second criterion , epitope mutation into multiple haplotypes is a typical pattern of CTL escape , termed epitope shattering in [30] , and the eventual loss of a putative epitope supports positive selection rather than neutral evolution . Following [24 , 25] , we model the pathways of CTL escape in each patient through an escape graph . Vertices of an escape graph represent viral variants that are part of the escape pathway and edges correspond to epitope mutations needed to change one variant into another . Variants are characterized by whether the considered epitopes are mutated or not , and variants are labeled by a string of 0’s and 1’s , e . g . 100000 , with each digit associated with an epitope and a digit of 0 and 1 representing a haplotype with the epitope un-mutated and mutated , respectively . We do not distinguish between different mutations on the same epitope . Fig 1 shows the escape graphs for patients CH40 , CH57 , CH77 , and CH256 , see Methods for details regarding the datasets and construction of escape graphs . The sample frequencies of each variant in the escape graph at t1 and t2 are shown in Fig 1 . Over all patients and all variants sampled , only 3 variants are present at both sample times . Given this pattern , for each patient we distinguish between two groups of variants: initial variants and expansion variants . Initial variants are those variants found at the first sample time while expansion variants are found only at the second sample time . Below , we refer to initial or expansion vertices , rather than variants , when we are discussing the escape graph itself rather than the viral population . The form of the escape graphs reflects the concurrent nature of HIV escape in the four patients and is in-line with the previous results of Pandit and De Boer [27] . For example , as shown in Fig 1D , at t2 the CH256 escape involves four variants that are all children of variant 100000: 110000 , 101000 , 100100 , and 100010 . Escape through these four variants involves mutation at distinct epitopes , reflecting concurrent escape from CTL responses to POL393 , NEF185 , ENV799 , and VIF169 through multiple pathways . For NEF185 , escape occurs concurrently through variants 101000 and 101001 , with variant 101001 linking escape at NEF185 and ENV606 . Similar patterns are seen in the other patients . As shown in Fig 1 , most putative epitopes elicited a response in ELISpot assays , providing strong statistical support for escape driven by CTL mediated selection ( see S1 Text and S1 Table for further statistical details ) . Putative epitopes that did not elicit a response in ELISpot assays may not have been exposed to CTL-mediated selection , but our results are unchanged if those are removed from the escape graphs . At sample times after t2 , a new set of vertices arise from the expansion vertices and the expansion vertices collapse to low frequency or are no longer sampled . For sample times extending up to 6 months , most variants are seen once at intermediate frequencies and then disappear at the next sample time as occurs for the initial vertices in moving from t1 to t2 . This observation has been made previously , although not in the context of linked data [12 , 14 , 20 , 25] . Current methods for estimating escape rates consider escape at each epitope separately by grouping variants into wild type and mutant according to the absence or presence , respectively , of mutation solely at the given epitope [12 , 15 , 19 , 21 , 28] . Letting fWT ( t ) and fMT ( t ) be the frequency of the wild type and mutant groups at time t , the model introduced in [19 , 21] , which we refer to as the logistic model , leads to an escape rate estimate ϵ as the solution of the relation f MT ( t 2 ) f WT ( t 2 ) = f MT ( t 1 ) f WT ( t 1 ) exp [ ϵ ( t 2 - t 1 ) ] , ( 1 ) with sampling occurring at the two timepoints t1 and t2 . Escape rate estimates based on Eq ( 1 ) are problematic for three reasons: the presence of multi-epitope escape biases the estimates in unpredictable ways; when wild type or mutants are at zero frequency or unsampled at t1 and t2 , Eq ( 1 ) cannot be applied; and the presence of mutation biases estimates up . In the presence of multi-epitope escape , variants within and across the wild type and mutant groups can differ at epitopes other than the one used to form the groupings , leading to bias . As a concrete example , consider the hypothetical escape graph shown in panel A of Fig 2 . The graph reflects escape at epitopes a , b , and c , with the associated frequency dynamics shown in panel B . The frequency dynamics were generated assuming equal replicative fitness across variants and constant CTL kill rates of 0 . 4 , 0 . 3 , and 0 . 5 day−1 at epitopes a , b , and c , respectively . Under these assumptions , the escape rate at each epitope should equal the CTL kill rate at the epitope . However , applying Eq ( 1 ) gives escape rate estimates of 0 . 69 , 0 . 25 , and 0 . 62 day−1 for epitopes a , b , and c respectively . The overestimate of the escape rate at epitope a ( 0 . 69 instead of the true 0 . 4 ) is caused by the groupings: the wild type group is composed of variant 000 and the mutant group is composed of variants 100 , 110 , and 101 . The escape of mutant variants 110 and 101 is driven by CTL responses to epitopes a , b and c , but is erroneously attributed solely to CTL response to epitope a , leading to the overestimate . Similarly , the underestimate of the escape rate at epitope b ( 0 . 25 versus the true value of 0 . 3 ) results from a wild type group that includes variant 101 while the mutant group contains variant 110 . The escape rate of mutant variant 110 relative to wild type variant 101 is slowed by response at epitope c and Eq ( 1 ) erroneously attributes the slower escape to weaker response at epitope b , leading to the underestimate . Overall , different escape graph geometries , variant frequencies , and underlying escape rates can lead to an upward or downward bias in the escape rate estimates given by Eq ( 1 ) ( see [31] for similar observations ) . A second problem associated with Eq ( 1 ) arises when either the wild type or mutant groups have sample frequency zero at t1 or t2 . This occurs when no mutations have occurred by time t1 or t2 , when the sweep of the mutants has completed and no wild types exist , or when the modest number of samples possible in SGA fails to capture mutant or wild type variants . In such cases Eq ( 1 ) cannot be applied: a serious limitation for our patient escape graphs because nearly all epitope mutations are first sampled at t2 ( Fig 1 ) . Previous authors have estimated escape rates in such cases by replacing zero frequencies with a non-zero value , usually 1/ ( n + 1 ) where n is the number of sequences sampled [19–21] . Such an approach leads to lower bounds on the escape rates—in other words a downward bias of the estimates—but as we show below , the lower bounds are poor . Finally , Eq ( 1 ) ignores the presence of mutation . When wild type and mutants are at significant frequencies at t1 and t2 , mutation plays a minor role . However , in situations when the mutant population is still small at t1 , mutations significantly increase the number of mutants at t2 relative to t1 and Eq ( 1 ) erroneously attributes these extra mutant variants to a higher escape rate . Table 1 shows the relative error of escape rate estimates using Eq ( 1 ) based on simulations of early acute HIV infection ( see caption of Table 1 and Methods for full details of simulations ) . The simulations include different strengths of CTL response and different escape graph geometries . Column sampled freq gives the relative error of the estimate formed through Eq ( 1 ) with 15 sampled sequences at both t1 and t2 . All three biases discussed above are in effect , but the dominant bias is the small frequency of mutants at t1 , leading to samples in which the mutants are not present and the 1/ ( n + 1 ) substitution is made . The relative errors in this column are all negative , reflecting underestimates of the true escape rates , and extreme: the confidence intervals range between roughly −0 . 50 and −0 . 95 , which corresponds to escape rate estimates that are 50% to 95% of the true escape rate . To make the effect of these underestimates concrete , consider that a 95% underestimate would give an estimate of 0 . 03 day−1 for a true escape rate of 0 . 6 day−1 . Column exact freq gives relative errors for estimates formed through Eq ( 1 ) using the exact wild type and mutant frequencies and restricted to those escape for which mutants and wild types both existed at times t1 and t2 , thereby removing the effect of sampling variance and the 1/ ( n + 1 ) substitution . With the bias associated with the 1/ ( n + 1 ) substitution removed , only the biases associated with multi-epitope escape and mutation still exist , but these lead to significant under or overestimation of the escape rate depending on the specific form of the simulations . Finally , to form the column exact freq/no mutation , we simulated viral dynamics with no mutation between t1 and t2 and then applied Eq ( 1 ) using the exact variant frequencies . With the bias of the 1/ ( n + 1 ) substitution and mutation removed , only the bias associated with multi-epitope escape remains . As can be seen , removing mutation produces lower estimates relative to column exact freq , but the estimates are still biased due to multi-epitope escape , with the direction of bias depending on the geometry of the escape graph . To account for concurrent multi-epitope escape our approach is to associate an escape rate with each edge of the escape graph . Intuitively , each parent-child vertex pair represents a “competition assay” measuring the selective advantage of the child relative to the parent and since there are only two variants considered , the bias associated with multi-epitope escape in Eq ( 1 ) does not occur . Let fP ( t ) and fC ( t ) be the parent and child frequencies at time t for a given parent-child vertex pair . We generalize the logistic model to parent-child vertex pairs through f C ( t 2 ) f P ( t 2 ) = f C ( t 1 ) f P ( t 1 ) exp [ ϵ ( t 2 - t 1 ) ] . ( 2 ) In [19 , 21] , Fernandez et . al . and Asquith et . al . derive Eq ( 1 ) by considering the difference in growth rates between wild types and mutants , see for example Equation ( 1 ) in [19] . Eq ( 2 ) follows from identical arguments , but we consider the parent and child variants instead of wild type and mutant variants . In Eq ( 1 ) , fWT ( t ) + fMT ( t ) = 1 for both t = t1 , t2 , but in Eq ( 2 ) , fP ( t ) + fC ( t ) can sum to any value less than or equal to 1 . Eq ( 2 ) no longer suffers from multi-epitope escape bias , but the zero frequency and mutation biases still exist . Table 2 is analogous to Table 1 , but we estimate escape rates using Eq ( 2 ) . As column exact freq/no mutation in the table shows , estimates are excellent when sampling is exact and no mutation occurs . The errors that do exist in assuming exact frequencies and no mutation arise from numerical error in the simulations and recombination . Zero frequencies are particularly problematic in the context of Eq ( 2 ) because both parent and child variants can be zero at t1 , while in Eq ( 1 ) wild type and mutants cannot both be zero . Substitution of 1/ ( n + 1 ) for both parent and child frequencies in Eq ( 2 ) leads to underestimates that are more extreme than substitution for wild type and mutant frequencies in Eq ( 1 ) , compare the sample freq columns in Tables 1 and 2 . For our patient escape graphs , most vertices are expansion vertices and have zero frequency at t1 . This difficulty exists in our simulations as well , where roughly 90% of all variants have a true frequency of less than 0 . 1 at t1 = 30 days , reflecting the time needed for mutant variants to arise and expand to significant frequencies . As a result , Eq ( 2 ) is not useful for investigation of early escape rates given sampling times that capture CTL escape at a single time point , as is the case in our patient datasets . To estimate escape rates for edges pointing to expansion vertices , we replace Eq ( 2 ) with an estimate that accounts for parent to child mutations and depends on sampled frequencies only at t2: f C ( t 2 ) f P ( t 2 ) = A μ ϵ ( exp [ ϵ ( t 2 - t I ) ] - 1 ) . ( 3 ) In Eq ( 3 ) , A parameterizes the number of parent to child mutations , tI is the time of the first parent to child variant mutation , and μ is the rate at which a parent variant mutates into a child variant . Given data providing the frequencies fC ( t2 ) , fP ( t2 ) and a choice for the parameters A , tI and μ , an estimate ϵ is determined through Eq ( 3 ) . Importantly , Eq ( 3 ) estimates escape rates during the interval [tI , t2] rather than [t1 , t2] . In practice , since Eq ( 3 ) is nonlinear with respect to ϵ , a numerical equation solver must be used to calculate ϵ . To explain Eq ( 3 ) and the parameters A , tI , we consider different models for the ratio fC ( t2 ) /fP ( t2 ) . Let mP ( t ) and mC ( t ) be the parent and child variant population sizes at time t , respectively . A deterministic model of parent-child dynamics assumes growth rates r − ϵ* and r for parent and child variants , respectively , and a parent to child mutation rate μ , leading to the following differential equation system: m˙P= ( r−ϵ* ) mP , m˙C=rmC+μmP . ( 4 ) In Eq ( 4 ) , we let ϵ* be the modeled ( true ) selective advantage of mutants to distinguish it from the escape rate estimate ϵ in Eq ( 3 ) . The selective advantage can arise through CTL killing , replicative fitness costs or other means—the model is unaffected by the underlying cause—and an exact estimate would yield ϵ = ϵ* . Integration of Eq ( 4 ) from time 0 to t2 under the initial condition mC ( 0 ) = 0 and mP ( 0 ) = m0 gives f C ( t 2 ) f P ( t 2 ) ≈ μ ϵ * ( exp [ ϵ * t 2 ] - 1 ) . ( 5 ) Eq ( 3 ) is precisely Eq ( 5 ) when A = 1 and tI = 0 , reflecting the assumptions implicit in Eq ( 5 ) that mutation begins at time 0 and occur deterministically with rate μmP ( t ) . While we have presented Eq ( 4 ) with constant rate r and μ , the derivation of Eq ( 4 ) would still hold if r and/or μ vary . Starting mutation at t = 0 is unrealistic , since parent variants do not even exist at initial infection unless the parent is the founder variant , which is not the case for any of the expansion edges in our patient escape graphs . But we can alter Eq ( 4 ) to account for delayed mutations as done in [32] by specifying a cutoff time tcutoff before which no mutations occur and after which mutations occur at rate μmP ( t ) . Under such a model we would have the relation , f C ( t 2 ) f P ( t 2 ) ≈ μ ϵ * ( exp [ ϵ * ( t - t cutoff ) ] - 1 ) , ( 6 ) which is Eq ( 3 ) with A = 1 and tI = tcutoff . Other models of parent-child variant dynamics and mutation are possible . If we assume that parent variants mutate into child variants at a rate μmP ( t ) once t > tI , but that the actual number of mutations varies around this average due to mutational stochasticity , we arrive at Eq ( 3 ) with A parameterizing an excess or dearth of mutations relative to the average . When A = 1 , the number of parent to child mutations exactly equals the average . When A > 1 and A < 1 , the number of mutations is greater and less , respectively , than what we would expect . Table 3 shows the effect of the parameters A , tI , μ on escape rate estimates . For the simulations presented in Table 1 , we recorded the true A , tI and μ value for each expansion edge . The columns of Table 3 show the relative error of escape rate estimates based on Eq ( 3 ) when we set one of the parameters as specified in the column label while the other two are set to their true values—confidence intervals ( CIs ) have not been included for readability , see S2 Table . For example , in the column A = 0 . 75 , we set A = 0 . 75 while tI and μ were set to their true values . The true value of μ was 10−4 for all parent-child pairs . As the table shows , changing any one of the parameters within a reasonable range can lead to estimates that are upper or lower bounds , reflected by a positive or negative relative error , respectively . We do not know the ‘true’ A or tI for a parent-child pair within a given patient and μ is only roughly known from existing mutation studies [33] . Further , within a given patient different parent-child pairs likely have different ‘true’ A and tI values . Previous approaches for estimating escape rates across multiple epitopes using the standard model and/or birth death processes can be viewed in the context of Eq ( 3 ) , with the model selected providing an implicit choice for the A , tI and μ for each parent-child vertex pair . To apply Eq ( 3 ) to our patient escape graphs , we consider a plausible range for each parameter and develop lower and upper bounds for the escape rates based on these ranges . All escape rates within these lower and upper bounds are consistent with our patient datasets . Conversely , greater certainty in the escape rate estimates requires either more assumptions , i . e . some method of specifying a narrower range for A , tI and μ , or denser temporal sampling that would capture escape variants at two timepoints , thereby eliminating expansion variants and allowing us to apply Eq ( 2 ) . In our simulations , A ∈ [1 , 10] and tI ∈ [16 , 34] for roughly 75% of expansion variants . Since peak viral load generally occurs 21 days post infection [9] , tI falls between 5 days prior and two weeks after peak viral load . As an additional verification of the lower bound on tI , the CTL response likely arises roughly a week prior to peak viral load [14] , making the presence of an expansion variant that is two or more epitope mutations removed from the wild type unlikely at five days prior to peak viral load . Finally , we take μ ∈ [3 × 10−5 , 3 × 10−4] . Given a per nucleotide , per reverse transcription mutation rate of 3 × 10−5 [33] , the range reflects epitope mutations formed by 1 to 10 different single nucleotide mutations . Applying Eq ( 3 ) with A = 10 , tI five days prior to peak viral load and μ = 3 × 10−4 and with A = 1 , tI two weeks post peak viral load and μ = 3 × 10−5 , respectively , gives us lower and upper bounds for the escape rate . We also consider intermediate values for our parameters determined by the medians of A and tI: A = 1 . 25 and tI four days post peak viral load . We choose μ = 10−4 as an intermediate value of μ . Table 4 shows the relative error of these lower and upper bounds applied to the simulations described above , as well as the relative error given the intermediate values of the parameters . Notice that the relative error is negative and positive , respectively , for the lower bound and upper bounds , reflecting under and over estimation of the true escape rate , respectively . The intermediate bounds are roughly accurate , although they tend to slightly overestimate the escape rate . While the parameter ranges we have chosen are relatively broad , the possibility exists that we have missed the true parameter range of HIV infection . Our method does not eliminate parameter dependence , but by choosing a broad range of parameter values , we are more likely to have captured true dynamics than if we simply specified particular values for A , tI and μ . Table 5 provides escape rate estimates for CTL escapes seen in the four patients given the frequencies sampled at t1 and t2 ( Fig 1 ) . To apply Eq ( 3 ) , we only calculate escape rates for edges corresponding to a mutation at a single putative epitope and pointing to expansion vertices , but this restriction still includes most putative epitopes in the escape graphs . We calculate escape rates for 15 putative epitopes of which 14 elicited a response from patient T cells in ELISpot assays or are known epitopes or motifs for the patient HLA type [12 , 13] , supporting the presence of CTL-mediated selection . The sole exception , ENV830 in CH58 , was lost by day 85 post symptoms , mutated through four different haplotypes during days t1 , t2 and eight different haplotypes as infection progressed , dynamics suggestive of epitope shattering [30] . Columns lower bound , intermediate and upper bound correspond to the choices for A , tI and μ described above , with the lower and upper bound representing a range in which we believe the escape rate lies , and the intermediate estimate describing a middle ground in which the parameters are not chosen to be extreme . The column single gives the escape rate estimate produced through Eq ( 1 ) , which considers escape at each epitope separately . The column previous gives previous escape rate estimates from [13 , 20] based on sample times up to 6 months and later , significantly beyond the 2–3 month range we consider . For CH58 and CH77 , the lower bounds demonstrate that escape can proceed concurrently , with rates exceeding 0 . 05 day−1 at multiple epitopes . The lower bounds are less informative for CH40 and CH256 . For both these patients , a single epitope has a lower bound escape rate exceeding 0 . 05 day−1 , but the other bounds are lower , roughly 0 . 03 day−1 . Across all patients , the upper bound escape rates are significant , allowing for the possibility of fast escape at multiple epitopes . However , with the exception of patient CH77 , the upper bounds do not exceed 0 . 4 day−1 , roughly the upper range seen in previous studies that considered escape separately at each epitope [12 , 20 , 28] , suggesting a general limitation in the escape rate that applies across multiple epitopes . The large difference between our lower and upper bounds reflects a large range of dynamics consistent with our patient datasets . Therefore , current acute HIV infection datasets based on SGA sampling and sparse temporal sampling cannot accurately estimate escape rates without significant modeling assumptions . Our intermediate estimates provide one case of such modeling assumptions . The estimates , formed by typical values for A , μ and tI seen in simulation , demonstrate that significant escape rates of roughly 0 . 1–0 . 2 day−1 across multiple epitopes are consistent with our data and reflects reasonable assumptions . The single escape rate estimates tend to fall near our lower bound estimates , possibly reflecting a downward bias of Eq ( 1 ) , as seen in our numerical studies ( see Table 1 ) . In some cases , the single escape rates are negative and so almost certainly reflect such downward bias . Previous estimates made in [13 , 20] also tend to fall near our lower bounds and in some cases far below the lower bounds . For example , escape ENV605 in CH77 was predicted to occur at a rate 0 . 01 day−1 and our novel estimates suggest at least an escape rate of 0 . 22 day−1 , a 22-fold increase . Since single estimates of escape rates are based on samples over a longer time period , their downward bias relative to our estimates suggests that escape rates may indeed slow down as time progresses as has been suggested previously [18 , 20] . Overall , our analysis strongly suggests that escape in acute HIV infection occurs concurrently from multiple CTL responses and proceeds at significant rates . However , whether escape rates reach the ranges of 0 . 1–0 . 2 day−1 suggested by our intermediate values cannot be determined without more assumptions ( leading to narrower rate for parameters A , tI , and μ ) or denser temporal sampling .
We have presented a novel method for estimating the rate of concurrent escape of HIV from multiple CTL responses that can be applied to our dataset and potentially other datasets . The method is based on an escape graph representing the mutation pathways through which HIV evades CTL response . Our method extends the logistic model of [19 , 21] to an escape graph by considering pairs of viral variants corresponding to parent-child vertices on the graph . Through stochastic simulations we have shown that the logistic method can be biased in complex ways by concurrent multi-epitope escape and severely downward biased when escape is captured at a single time point . Our results suggest that CTL escape can occur concurrently at multiple epitopes , with escape rates ranging between 0 . 03 and 0 . 4 day−1 across multiple epitopes . The upper bound of 0 . 4 day−1 is in-line with upper bounds for CTL escapes found using the logistic model [12 , 14 , 15] , suggesting a general limit for the rate of escape whether escape is proceeding concurrently at multiple epitopes or not . Our lower bounds are less informative but are greater than estimates of the escape rates based on methods that do not account for concurrent multi-epitope escapes . If replicative fitness costs are associated with epitope mutations , then CTL kill rates are higher than the escape rate estimates . But this effect would apply to previous estimates as well , so the comparison between our escape rate ranges and previous ranges still has validity . In a similar vein , we have grouped mutations at each epitope , meaning that our escape rate estimates are averages over the different mutation variants at a given epitope . As a result , our escape rate estimates will underestimate the escape rate of some mutation variants and overestimate others . Pandit and De Boer [27] recently demonstrated the presence of concurrent CTL escape in a one patient dataset . Our results provide further confirmation of concurrent CTL escape through four patient datasets . However , the nature of CTL escape may differ between patients and differing conclusions may reflect the use of different datasets . Results by Kessinger et . al . [25] and da Silva [34] suggest non-concurrent escape , meaning that escape occurs one epitope at a time . Kessinger et . al . [25] present simulations showing that escape is largely non-concurrent , but that ( see their Fig 2 ) concurrent escape can occur , although with one variant at high frequency . Combining our results and those of Pandit and De Boer [27] with the analysis of Kessinger et . al . [25] suggests escape dynamics in which concurrent escape occurs and is resolved in favor of a given escape variant , which then serves as the basis of another concurrent escape that is resolved in favor of another escape variant , and so on . Further work will be required to verify this conjecture . Da Silva calculated an effective population size ( Ne ) for HIV of 102–103 during early infection based on a census population on the order of 107 . In simulations of CTL escape based on a Wright-Fisher model with Ne of 102–103 , da Silva observed that escapes do not happen concurrently . Since our simulations assume census population sizes similar to da Silva’s ( i . e . 107 ) , combining our results with those of da Silva raises the possibility that a Wright-Fisher model does not accurately reflect HIV escape dynamics , as has been suggested by previous authors [35 , 36] ( see also [37] for a non-HIV perspective ) . Concurrent viral escape variants may affect each other indirectly through competition for target cells . Previous authors have considered such interactions and the potential role of clonal interference on viral evolution , see the reviews [38 , 39] for further details . However , the role of target cell limitation in early HIV infection is still unclear [40–42] . Importantly , the specific dynamics of interaction between CTL escapes affect our escape rate estimates only through A , tI and μ . Our approach and results come with the statistical caveat that a biased escape graph will produce biased estimates . For example , the escape graph will be biased when many low frequency variants are present . To see why , consider an extreme example of CTL escape at 100 epitopes and imagine that 100 variants exist in the viral population , each variant mutated at a single epitope and each variant at frequency 1% . If we form an escape graph based on sampling 10 sequences , the most likely outcome is 10 different variants , each with sample frequency 10% but with true frequency 1% . Dataset CH77 has a sampling pattern consistent with such biasing . In general , formation of the escape graph is statistically complex and other forms of bias may exist , although such biases did not arise in our simulations . Exploration of this issue requires further work . Much of our approach follows from the sparse sampling of current datasets . While the rise of deep sequencing datasets addresses the shallowness of sampling , understanding the complexity of early escape requires linkage information and would benefit from more sampling time points . Our methods and results highlight the importance of better sampled datasets for understanding early HIV dynamics and evolution .
The left escape graph of Fig 3 shows escape at three epitopes . Multiple paths can be taken to reach variant 111 and recombination may play a role . Our current method does not account for recombination , but in many cases we can nevertheless obtain escape rate estimates . Our goal in estimating escape rates is to quantify the CTL killing rates associated with CTL response at different epitopes . When multiple paths exist between two vertices , multiple edges must correspond to escape at the same epitope: here edges B and E correspond to escape in epitope 2 , while edges C and D correspond to escape in epitope 3 . In such cases , we specify a subgraph that includes a single escape rate for each epitope and we refer to such subgraphs as escape trees since each vertex possesses a single parent . The right graph of Fig 3 shows such an escape tree that is a subgraph of the escape graph on the left . The methods for estimating escape rates discussed above apply to escape trees , but issues of recombination associated with multiple paths are eliminated . Ideally , we would estimates escape rates at all edges and this would provide information about CTL response in the context of different haplotypes . For example , in Fig 3 , edges B and E correspond to escape at the same epitope , but the escape involves different haplotypes . Escape rates associated with the two edges need not be equal and estimating both might provide valuable information addressing the additivity of CTL killing across multiple epitopes . However , analyzing escape graphs with multiple paths would require more parameters to model recombination , an extension we have not explored . In our datasets , the escape graphs of patients CH40 and CH77 require no pruning to reduce them to escape trees . The escape graph of patient CH58 requires the removal of two vertices , but the associated variants are at very low frequency at t1 , 0 . 02 and 0 . 04 respectively , and are not sampled at t2 . In CH256 we remove three vertices to produce an escape tree , with all three corresponding variants unsampled at t1 and sampled at 0 . 04 frequency at t2 , see S1–S4 Figs for details . To perform simulations of viral escape , we use the model of Batorsky et . al . [32] , developed for modeling chronic HIV infection , as a basis for a model of acute infection . The Batorsky et . al . model tracks individual viral genomes over time . Viral genomes , which correspond to provirus in infected cells , produce offspring , mutate and recombine . Selection is also modeled based on genome haplotype at L loci . As in the Batorsky et . al . model , we model selection at L loci , representing the CTL epitopes , and we account for the effects of mutation and recombination . In addition , we allow the number of viral genomes to vary , to account for viral expansion and contraction before and after peak viral load . Our model also includes time-varying selection as a model of CTL response . For simplicity , we assume all viral variants have equal replicative fitness . For each simulation we track the CTL kill rate at a given epitope , providing us with the true escape rate ( i . e . ϵ* in the notation of Eq ( 4 ) ) . Given a simulation , we then simulate sampling 15 sequences at t1 = 30 and t2 = 60 , form an escape graph based on the simulated sequences , and then estimate the escape rate , ϵ , based on the inference method we are considering and the escape graph . Combining the true and estimated escape rates provides us with the relative error shown in the Tables 1–4 . We produced 1 , 000 simulated escape graphs for each combination of strong and weak CTL response models and full and linear escape graphs ( see caption of Table 1 for definitions ) using a tau-leaping approach [43] . Table 6 gives the simulation parameters and their values . See S5 and S6 Figs for an example of a single simulation and the corresponding escape graph generated . To model CTL response at L = 6 epitopes—roughly matching the number seen in our patient datasets—we use a CTL dynamics model introduced by De Boer et . al . in [44 , 45] . Letting ki ( t ) be the CTL kill rate at time t of a variant possessing epitope i , we set ki ( t ) = 0 for t ≤ Ton , i , where Ton , i represents a time at which CTL response to epitope i initiates . For t > Ton , i , ki ( t ) varies according to k ˙ i = f ( N i ) α k i ( 1 - k i k max , i ) - ( 1 - f ( N i ) ) β k i , ( 7 ) with ki ( Ton , i ) = 0 . 01 giving the initial condition of the response . ( See Table 6 for meaning of parameters . ) De Boer et . al . modeled different types of CTLs and tracked the CTL population size ( see equations 4–7 in [45] ) , but here we group all CTLs and assume that the CTL population varies proportionally with the CTL kill rate , allowing us to track ki ( t ) . In our simulations , the CTL response to epitope 1 occurs early and is relatively strong compared to responses to the other epitopes , which expand later . Strong and weak CTL simulations differ based on the value of kmax , i . Rates of expansion and contraction for the immune response ( α and β ) are taken from parameters for the CD8 T cell response to a virus in mice [44 , 45] . These values appear to be much higher than rates of T cell expansion and contraction in humans following vaccination [46] . In our simulations , higher values of these parameters allow for rapid change in CTL killing efficacy . The impact of the slower dynamics on the kinetics of escape will be investigated elsewhere . At time 0 , the population starts with 1 variant possessing all 6 epitopes . The population size N—not to be confused with effective population size—is expanded from 1 to 107 within the first 21 days , following estimates of the total number of infected cells in the body , and then collapses to 104 . 5 over the following two weeks and subsequently holds steady [47] . The reduction from 107 to 104 . 5 variants is larger than estimated in [47] , but matches better the magnitude of viral load decline seen during acute infection [1] . Variants other than 000000 arise through mutation and recombination occurring at rates μ and ρ . All mutations occur independently at each epitope and to each variant according to a Poisson process run at rate μ = 10−4 . Recombination occurs at rate ( ρ/5 ) NfA fB with ρ = 1 . 4 × 10−5 and where fA , fB are the frequencies of the recombining variants A and B . When recombination does occur , we assume 5 breakpoints uniformly selected over a 10000 nucleotide genome , matching the number of breakpoints measured through in-vitro studies [48–50] . Overall this gives a per nucleotide per day recombination rate of ρ , matching the rate measured in [51] . We assume all epitopes are 1000 base pairs apart . In the simulations , when a recombination between variants A and B occurs , we increased the resulting recombined variant population by 1 and decrease either the population of A or B variants by 1 , where we choose to decrease either A or B with probability 1/2 . For full details regarding patient datasets CH40 , CH58 , CH77 , and CH256 see [12 , 13] and references therein . Briefly the patients were identified during acute infection: viral load data suggests CH58 and CH256 were identified several days prior to peak viral load while CH40 and CH77 were identified several days after peak viral load . Blood samples were taken at various time points after the onset of symptoms and at each time point SGA was used to sequence the 5’ and 3’ ends of viral DNA , meaning that our data covers the whole genome but does not include linkage information between the 5’ end ( GAG , POL ) and 3’ end . For patients CH40 and CH58 , we apply our methods to the first two time points sampled after the onset of symptoms . Escape at CH77 was very fast and broad , encompassing 9 loci by the first timepoint sampled , day 14 post symptoms . To study this early escape , we assume that 10 days prior to symptoms , a time likely about 5 days prior to peak viral load , sampling would have been homogeneous for the founder variant . We then use 10 days prior to symptoms as our first timepoint and day 14 past symptoms as our second timepoint . For CH256 , the first timepoint sampled was homogeneous for the founder variant and the second timepoint possessed variation at a single loci , so we use the second and third timepoints as our two sample times . To choose lower bounds for tI , we estimate the patients’ peak viral load times and picked a time 5 days earlier . S3 Table summarizes the lower bound values for tI , t1 , and t2 that we use for each patient dataset along with the number of samples available for the 5’ and 3’ end at each time point . For CH77 all of the putative epitopes are on the 3’ end , so construction of the escape graph follows directly from the data . For the other three patients we construct a full escape graph by attaching 5’ edges onto the 3’ escape graphs . CH58 and CH256 each has only a single putative epitope on the 5’ end , meaning we form the full escape graph by adding a single edge to the 3’ escape graph . CH40 has putative epitopes evenly split between the 5’ and 3’ ends . Since we lack linkage information , the full escape graph represents a guess on our part . But we attach 5’ escapes to parts of the 3’ escape graphs near the root vertex , if the 5’ escapes are actually attached further away from the root our estimated escape rates would be higher . S1–S4 Figs show the 5’ and 3’ escape graphs , as well as the full escape graph we constructed . | Since the early 1990s , cytotoxic T lymphocytes ( CTLs ) have been known to play an important role in HIV infection with CTLs targeting HIV epitopes and , in turn , HIV escapes arising through mutations in the targeted epitopes . Over the past decade , studies have shown that CTL responses concurrently target multiple HIV epitopes , yet the effect of concurrent responses on HIV dynamics and evolution is not well understood . Through an analysis of patient datasets and a novel statistical method , we show that during early HIV infection concurrent CTL responses drive concurrent HIV escapes at multiple epitopes with significant pressure , suggesting a complex picture in which HIV simultaneously explores multiple mutational pathways to escape from broad and potent CTL response . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [] | 2015 | Broad CTL Response in Early HIV Infection Drives Multiple Concurrent CTL Escapes |
Congenital or neonatal cardiomyopathies are commonly associated with a poor prognosis and have multiple etiologies . In two siblings , a male and female , we identified an undescribed type of lethal congenital restrictive cardiomyopathy affecting the right ventricle . We hypothesized a novel autosomal recessive condition . To identify the cause , we performed genetic , in vitro and in vivo studies . Genome-wide SNP typing and parametric linkage analysis was done in a recessive model to identify candidate regions . Exome sequencing analysis was done in unaffected and affected siblings . In the linkage regions , we selected candidate genes that harbor two rare variants with predicted functional effects in the patients and for which the unaffected sibling is either heterozygous or homozygous reference . We identified two compound heterozygous variants in KIF20A; a maternal missense variant ( c . 544C>T: p . R182W ) and a paternal frameshift mutation ( c . 1905delT: p . S635Tfs*15 ) . Functional studies confirmed that the R182W mutation creates an ATPase defective form of KIF20A which is not able to support efficient transport of Aurora B as part of the chromosomal passenger complex . Due to this , Aurora B remains trapped on chromatin in dividing cells and fails to translocate to the spindle midzone during cytokinesis . Translational blocking of KIF20A in a zebrafish model resulted in a cardiomyopathy phenotype . We identified a novel autosomal recessive congenital restrictive cardiomyopathy , caused by a near complete loss-of-function of KIF20A . This finding further illustrates the relationship of cytokinesis and congenital cardiomyopathy .
We present a small Caucasian family with three children ( S1 Fig ) . The parents are not consanguineous . Two of the children , one male ( II-2 ) and one female ( II-3 ) , were diagnosed in late fetal life with a congenital heart defect categorized as restrictive cardiomyopathy of the right ventricle ( RV ) . In the male index patient , the diagnosis of a small RV with severe pulmonary stenosis was made at the postmenstrual age ( PMA ) of 35 weeks . Due to secondary hydrops foetalis , with chylothorax and ascites , labour was induced at 35 weeks and 2 days . At birth , weight was 2400g ( 25th-50th centile ) , length 48cm ( 75th centile ) and head circumference 31 , 8cm ( 25th-50th centile ) . Postnatal echocardiography ( Fig 1 ) confirmed the diagnosis of a bipartite RV with agenesis of the apex , a functional pulmonary stenosis , moderate pulmonary insufficiency ( grade 2/4 ) and severe tricuspid insufficiency ( grade 3/4 ) . The peak instantaneous gradient ( PIG ) of 45mmHg on day 0 measured over the severe tricuspid insufficiency was lower than expected and this was thought to be due to the bipartite right ventricle with dysfunction and thus inability of the ventricle to generate sufficient pressure for anterograde flow . The pulmonary valve leaflets appeared thicker on echocardiography , the infundibulum was normal . Due to a pulmonary circulation dependent on a patent ductus arteriosus , IV prostaglandin was started . On day 1 percutaneous dilatation of the pulmonary valve was performed and the ductus arteriosus was stented . The gradient over the pulmonary valve was not measured in the cathlab as it was extremely difficult to have a stable position over the pulmonary valve , but no waist was seen with a 6mm balloon . On day 5 a Rashkind balloon septostomy of the intra-atrial septum was performed . Due to persistent ascites , pronounced hepatomegaly and increased transaminases , an MRI of the liver and liver biopsy was performed at 40 days postnatal age . This showed billirubinostasis , hypoplasia of the portal veins and associated hyperplasia of the portal arteries . A preliminary diagnosis of a ductal plate abnormality was made . During the subsequent weeks , the heart function of both ventricles progressively decreased . At the age of 3 months the decision was made to start palliative care and the patient demised at the age of 93 days . Autopsy confirmed the cardiac diagnosis . The right ventricle was hypoplastic with the cardiac apex formed solely by the left ventricle . The right atrium and tricuspid annulus showed important dilation; the tricuspid valves were slightly thicker and curled towards the atrium; consistent with severe tricuspid insufficiency . The leaflets of the pulmonary valve were confirmed as being tricuspid and mildly dysplastic . Microscopic examination showed pronounced subendocardial to transmural ischemic fibrosis of the myocardium ( S2A Fig ) . The myocardial tissue was hypertrophic with hydropic swelling and myocytolysis ( S2C Fig ) . The endocardium showed fibrous thickening and there were prominent intramyocardial sinusoids ( S2B Fig ) . The myocardium of the left ventricle ( LV ) was grossly normal and not hypertrophic , except for endocardial fibrosis which was clearly less pronounced compared to the RV . The coronary arteries were normal . Pronounced chronic venous congestion of the liver was noted with cardiac fibrosis . The venous centrolobular walls were severely thickened with formation of centro-central fibrous septae . Ductal proliferation was present , but ductal plate malformation could not be confirmed given the normal central bilious ducts in the larger portal fields . In a following pregnancy , at the PMA of 32 weeks , the diagnosis of restrictive right ventricular cardiomyopathy with RV dysfunction was made in the female fetus . She was born at the PMA of 37 weeks . Her weight , length and head circumference at birth were within normal range; 3 , 450 kg ( 25th-50th centile ) , 49 , 5cm ( 10th-25th centile ) and 33 , 8 cm ( 3rd-10th centile ) respectively . She was admitted in NICU due to cyanosis and cardiac decompensation with pronounced ascites . Postnatal echocardiography confirmed the diagnosis of a restrictive cardiomyopathy . The pulmonary valve was morphologically normal , but decreased anterograde flow as well as moderate tricuspid insufficiency was present ( grade 2/4 ) secondary to RV dysfunction . No hepatic abnormalities were present . In the following weeks the heart function progressively decreased , at the age of 71 days ( 2 months ) the patient demised . An autopsy was not performed . Array-CGH is both patients and the parents were normal .
Linkage analysis was performed on the entire family , and maximal LOD-score ( MLS ) of 0 , 727 was obtained in 27 regions ( S3 Fig ) . These regions contained a total of 1273 genes , obtained from Ensembl ( http://www . ensembl . org/ ) . Whole exome sequencing was performed on both affected siblings and the unaffected sibling . After filtering the variants in the genes in the linkage regions , under a hypothesis of autosomal recessive inheritance , we identified 1 gene with a homozygous variant ( PCDHA9 ) and 2 genes ( ZNF587 and KIF20A ) with compound heterozygous variants ( S1 Table ) . The PCDHA9 gene contained a nonsynonymous variant ( c . 1006C>G: p . L336V ) which was absent in the 1000 genomes , but with an allele frequency of 51% in local exomes and 61 . 8% in the ExAC database . In ZNF587 two missense variants were detected , c . 956C>G ( p . T319S ) and c . 1676G>A ( p . R559Q ) with an allele frequency of respectively 1% and 6% in local exomes . In KIF20A we identified a missense variant ( c . 544C>T: p . R182W ) , changing an arginine to a tryptophan , and a frameshift mutation , creating a premature stop codon ( c . 1905delT: p . S635Tfs*15 ) . The c . 544C>T substitution in exon 6 results in a single amino acid substitution ( p . R182W ) within the motor domain of the protein . Arginine and tryptophan are members of different chemical amino acid groups , and the R182 amino acid is highly conserved in 98 out of 100 vertebrates . The variant c . 544C>T: p . R182W was predicted to be damaging by in silico tools SIFT , Polyphen and MutationTaster . The c . 1905delT in exon 15 results in a frameshift that introduces a premature stop codon 15 amino acids downstream . These observations suggest that both variants are likely to affect protein function . These variants were absent in the population control exomes . In the ExAC Browser database , containing genetic data of 60 706 humans of various ethnicities , the missense variant was found in 2 individuals , respectively of South Asian and European origin . The frameshift variant was present in 32 individuals of African descent . [12] . Sanger sequencing validated the presence of both variants in the affected siblings and confirmed a heterozygous carrier status in both parents ( maternal c . 544C>T and paternal c . 1905delT ) . Both variants were absent in the unaffected sibling . Variants in other known cardiomyopathy genes according to our local cardiomyopathy panel were absent in the two affected siblings . Quantitative real-time PCR ( qPCR ) was used to investigate the effects of the KIF20A variants on its expression by comparing KIF20A cDNA-levels amplified from mRNA isolated from patient and control fibroblasts . Student’s T-test was used to test significance in expression level . Both patients had a significantly reduced expression level to 40–60% of control levels ( Fig 2A ) . To investigate the effect of the variants on KIF20A protein levels , immunoblotting was performed using unrelated controls and patient fibroblasts . Both affected individuals had a reduced amount of endogenous KIF20A protein compared to controls ( Fig 2B ) . Antibodies for the N-terminal and C-terminal part of the protein gave identical results , indicating that the frameshift mutation leads to elimination of the transcript by nonsense-mediated mRNA decay . The localization of the remaining KIF20A in dividing patient fibroblasts ( c . 544C>T: p . R182W ) was then examined . These cells have approximately half the levels of KIF20A when compared to control fibroblasts , but retain normal levels of other cell division proteins ( Fig 2C ) . In control cells , KIF20A localizes to the spindle midzone in anaphase and telophase of dividing cells where it promotes recruitment of the Aurora B kinase ( Fig 3A ) . In both patients KIF20A was aberrantly targeted to chromatin and failed to support translocation of Aurora B to the spindle midzone ( Fig 3A ) . As a consequence of the inability to relocate Aurora B to the spindle midzone , Aurora B phosphorylation of a key anaphase central spindle protein KIF23 was reduced ( Fig 3B , arrows ) . This failure to move from chromatin to the anaphase spindle microtubules suggested that the missense mutation ( c . 544C>T: p . R182W ) perturbed the kinesin motor activity . This possibility was therefore tested using microtubule-stimulated ATPase assays . Purified wild type or R182W mutant KIF20A proteins were tested over a range of concentrations in microtubule-stimulated ATPase assays . Plots of the initial rate of ATPase hydrolysis as a function of the concentration of motor domain show that the KIF20A R182W missense mutation has greatly reduced microtubule activated motor activity ( Fig 4A ) . This reduction in ATPase activity is typical of kinesin “rigor” mutants , which have a point mutation in the ATP binding site . As a result , the rigor motor can bind to microtubules but cannot hydrolyse ATP; this ATP-bound form of the motor is locked on the microtubule and does not support microtubule motility [13 , 14] . To further pursue this idea , a wild type KIF20A E245A “rigor” mutant and the missense mutation present in the patient cells R182W were transfected into HeLa cells where the endogenous copy of KIF20A was removed by siRNA . In the absence of any KIF20A , Aurora B is trapped on chromatin and is not present on the central spindle Fig 4B . Expression of wild type KIF20A rescues the transport of Aurora B to the central spindle . However , neither the patient R182W mutation nor the rigor E245A supported efficient Aurora B transport and this remains trapped on chromatin in dividing cells . Together these results indicate that the missense variant ( c . 544C>T: p . R182W ) is a near-complete loss-of-function mutation creating an ATPase defective form of KIF20A . The whole heart of kif20a morphants was smaller and significant pericardial edema was evident in all transverse sections of the morphants when compared with controls ( Fig 6A ) . The atrioventricular ( AV ) valve in morphants appeared morphologically normal , the bulbus arteriosus ( BA ) was smaller and the atrial and ventricular walls were thicker compared to controls . Pooling of blood was present anterior to the atrium of the morphants , suggesting a decreased function . The ventricular wall thickness was significantly increased compared to controls ( Fig 6A ) . This correlates with the pathological hypertrophy seen in the human patients . To better characterize the heart phenotype in kif20a morphants , we evaluated several cardiac parameters including heart rate and fractional shortening ( FS ) . At 3dpf the heart rates were similar in the controls and kif20a morphants . At 4dpf the morphants showed a significant increase in heart rate ( p = 0 . 009 , Fig 6B ) , most likely as a response to progressive heart failure and decreased stroke volume , since cardiac output equals heart rate x stroke volume . Although an increased ventricular fractional shortening was seen in the kif20a morphants compared to the controls ( 40 . 6% ± 10 . 38 vs . 30 . 49% ± 4 . 91 , p = 1 , 121 E-03 ) ; more outliers were observed in the kif20a morphants ( Table 1 and Fig 6B ) . The end-systolic diameter ( ESD ) of the atrium in kif20a morphants was significantly smaller ( p = 0 . 001 ) ; this is most likely due to increased force necessary to empty the atrium into a more rigid ventricle . These findings suggest that kif20a is required for normal heart function in zebrafish embryos .
We report a family with two siblings presenting with a novel lethal congenital heart disease . It was characterized by fetal-onset restrictive cardiomyopathy predominantly affecting the right ventricle and leading to irreversible heart failure and early death . Given the occurrence of the same distinct phenotype in siblings of both sexes with unaffected parents , autosomal recessive inheritance was likely . This phenotype is unique and to our knowledge has not been reported in literature previously . After exclusion of mutations in known CM genes , linkage analysis and exome sequencing was performed to identify the genetic basis . We were able to identify functional variants in the KIF20A gene as the most likely cause . Two compound heterozygous variants were found; one variant was a missense mutation ( c . 544C>T: p . R182W ) , the other a frameshift mutation , creating a premature stop codon ( c . 1905delT: p . S635Tfs*15 ) . There is no known phenotype of constitutional KIF20A mutations in humans . In mice , homozygosity is lethal in all pups at an early age of 3–4 weeks , but no phenotypic details have been reported [15] . In a zebrafish model we showed that translational blocking of the zebrafish kif20a gene resulted in a cardiomyopathy phenotype and that kif20a is required for proper function , suggesting KIF20A has an evolutionary conserved function in heart development . Future studies using more specific genetic knockout models could provide additional valuable information about its function . Kinesin family member 20A ( KIF20A ) , a . k . a . Mitotic kinesin-like protein 2 ( MKLP2 ) or Rab6-interacting protein ( RAB6KIFL ) is one of the kinesin-like proteins . These proteins are microtubule-associated motors that play important roles in intracellular transport and cell division [16] . It is required for chromosomal passenger complex-mediated cytokinesis and translocation of the chromosomal passenger complex ( CPC ) from the chromatin to the central spindle in metaphase , anaphase and telophase [17] . Functional studies in patient fibroblasts revealed reduced protein levels associated with deficient transport of the Aurora B and the CPC which remains trapped on chromatin in dividing cells . This was due to the missense variant causing a near complete loss-of-function of the ATPase function of KIF20A . It is not excluded that a complete loss-of-function is embryonically lethal , and that the minimal residual function of one allele in this family allowed survival beyond fetal life . KIF20A is highly expressed in human testis and thymus , it is moderately expressed in human cardiac myocytes ( www . genecards . org ) . There are no data on expression in zebrafish or cardiac expression in mice [18] . The crucial role of PLK1 in cardiomyocyte proliferation has been shown in zebrafish . This cardiomyopathy phenotype predominantly affected the right ventricle . Currently , it is not known why this occurs . It might be related to a different origin of the right ventricle which is formed by the second heart field , compared to the left ventricle which originates from the primary heart field . However , this might also be secondary to distinct differences in function of the fetal left and right heart . Unlike the adult circulation , in the fetus , the stroke volume of the fetal LV is not equal to the stroke volume of the RV as a result of intracardiac and extracardiac shunting . The RV receives around 65% of the venous return and the LV about 35% [19] . A cardiomyopathy affecting predominantly the RV could thus lead to significant morbidity and possible mortality during the fetal or early neonatal period . This could lead to early unexplained mortality and underreporting of this specific phenotype . Additional phenotypic features becoming apparent at a later age would also be difficult to detect . The specific link with the cell cycle and this cardiopathy in humans is still unclear . Previously we and others reported mutations in the ALMS1 gene as a cause of mitogenic cardiomyopathy . This links cardiomyopathy to ciliopathy and the cell cycle [20 , 21] . These reports open a new mechanism for future research in cardiomyopathies and cytokinesis .
Informed oral consent was given by the family for further genetic studies . This study was approved by the Ethics Committee of the University Hospitals Leuven , KU Leuven . The zebrafish experiments were carried out in accordance with the Guide of Care and Use of Experimental Animals of the Ethical Committee of KU Leuven . The Ethical Committee of KU Leuven approved all animal experiments . Genotyping was done on DNA extracted from peripheral white blood cells , obtained from the parents and both the unaffected and affected siblings . A dense SNP marker set derived from the 250k Affymetrics SNP typing platform was used in a recessive model . Genome wide parametric linkage analysis with Merlin software was performed ( http://www . sph . umich . edu/csg/abecasis/Merlin/tour/parametric . html ) . Whole exome sequencing was done on both patients and the unaffected sibling . Genomic DNA was sheared by sonication , platform-specific adaptors were ligated , and the resulting fragments were size selected . The library was captured using the SeqCap EZ Human Exome Library v2 . 0 ( Roche NimbleGen ) , and 2 x 76 bp paired-end reads were generated on a HiSeq2000 ( Illumina ) . Reads that did not pass Illumina’s standard filters were removed prior to alignment . Remaining reads were aligned to the reference human genome ( hg19 ) , using the Genome Analysis ToolKit ( GATK ) pipeline . After duplicate removal , local realignment and base quality score recalibration , the data was used for variant calling with GATK Unified Genotyper ( 2 . 4–9 ) . Annovar was used for functional annotation of detected variants . Quality filtering was applied by excluding variants found in less than 5 reads and variants detected in less than 15% variant reads . From the variant files , we only retained variants in genes from the linkage regions . Exonic variants and only intronic variants located less than 6 bp from the intron-exon boundary were included . Synonymous variants were excluded . Variants occurring with a frequency of <1% in the 1000 genomes project or with an unknown frequency were included . Variant filtering was done under the hypothesis of autosomal recessive inheritance , thus retaining only homozygous or compound heterozygous variants in both affected siblings , but not in the unaffected sibling . All remaining calls were checked for correct calling using Integrative Genomics Viewer ( IGV , Broad Institute , Cambridge , MA , USA ) . Primary fibroblasts from patients and unrelated controls were grown from skin biopsy and cultured in Dulbecco’s modified Eagle medium DMEM/F12 ( Life Technologies ) supplemented with 10% fetal bovine serum ( Clone III , HyClones ) , 1% streptomycine and 0 , 02% Fungizone at 37°C under 5% CO2 . The PCR was performed for KIF20A ( GenBank NM_005733 ) and the house-keeping gene GAPDH ( Glyceraldehyde 3-phosphate dehydrogenase , GenBank NM_002046 ) , which was used as an endogenous control for normalization . qPCR primers were designed using Genscript software ( https://www . genscript . com/ssl-bin/app/primer ) . All primers were synthetized by Integrated DNA Technologies . Student’s T-test was used to test significance in expression level . General laboratory chemicals and reagents were obtained from Sigma-Aldrich and Thermo Fisher Scientific . Sheep antibodies were raised to the KIF20A motor domain ( N ) or neck plus C-terminus ( C ) domains . Rabbit antibodies to KIF4A , KIF23 , PRC1 and KIF23 pS911 peptide were described previously [22–25] . Specific antibodies were purified using the antigens conjugated to Affi-Gel 15 , eluted with 0 . 2 M glycine-HCl , pH 2 . 8 , and then dialyzed against PBS before storage at −80°C . Commercially available antibodies were used to AIM1 ( mouse 611083; BD ) . Affinity-purified primary and secondary antibodies were used at a final concentration of 1 μg/ml . Secondary antibodies conjugated to Horseradish Peroxidase ( HRP ) were obtained from Jackson ImmunoResearch Laboratories , Inc . Secondary antibodies for microscopy conjugated to Alexa Fluor 488 , 555 , and 647 were obtained from Invitrogen . DNA was stained with DAPI ( Sigma-Aldrich ) . Human KIF20A was amplified directly from human testis cDNA . The KIF20A R182W mutant was created using QuikChange mutagenesis according to the instructions from Agilent Technologies . Mammalian expression constructs for N-terminally GFP-tagged KIF20A was made using pcDNA5/FRT/TO vector ( Invitrogen ) . Hexahistidine-tagged bacterial expression constructs for the motor domain ( 1–507 ) of wild type KIF20A or R187W KIF20A were made in pQE32 ( QIAGEN ) . HeLa cells were cultured in DMEM containing 10% ( vol/vol ) bovine calf serum ( Invitrogen ) at 37°C and 5% CO2 . For synchronization , cells were treated for 18 hours with 2mM thymidine , washed three times in PBS , and twice with growth medium . For plasmid transfection and siRNA transfection Mirus LT1 ( Mirus Bio LLC ) and Oligofectamine ( Invitrogen ) were respectively used . The siRNA duplexes used targeted the following sequences: control 5′-CGTACGCGGAATACTTCGA-3′ , KIF20A 3’-UTR 5′-CCACCTATGTAATCTCATG-3′ . Microscopy was performed as described previously [25] . The motor domains of KIF20A wild type and R187W were expressed in Escherichia coli strain JM109 and purified after induction for 3 hours with 0 . 5mM IPTG . Cell pellets were washed once in ice-cold PBS , and then lysed in 20 ml of IMAC20 ( 20mM Tris-HCl , pH 8 . 0 , 300mM NaCl , 20mM imidazole ) and protease inhibitor cocktail ( Sigma-Aldrich ) for 20 minutes on ice . Cell lysis was performed using an Emulsifex C5 cell breaker system ( Avestin Europe GmbH ) . Cell lysate was clarified by centrifugation and loaded onto a 1-ml HisTrap FF column ( GE Healthcare ) at 0 . 5 ml/min . The column was then washed with 30 ml of IMAC20 , and eluted with a 20-ml linear gradient from 20 to 200mM imidazole in IMAC20 collecting 1 ml fractions . Peak fractions were buffer exchanged using 5 ml Zeba Desalt Spin columns ( Perbio ) into TND ( 20mM Tris-HCl , pH 8 , 300mM NaCl , and 1mM DTT ) . Protein samples were snap-frozen in 15-μl aliquots and stored at −80°C for further use . A commercial enzyme–linked inorganic phosphate assay was used to measure kinesin ATPase activity ( Cytoskeleton , Inc . ) as described previously [25] . In brief , a microtubule premix was created at room temperature by mixing 1ml of reaction buffer ( 15mM PIPES-KOH pH 7 and 5mM MgCl2 ) , 10μl of 2mM paclitaxel , 80μl of preassembled microtubules ( 1mg/ml tubulin , 15mM PIPES-KOH pH 7 , 5mM MgCl2 , 1mM GTP , and 20μM paclitaxel ) , 240μl of 1mM 2-amino-6-mercapto-7-methylpurine riboside , and 12μl of 0 . 1U/μl purine nucleoside phosphorylase . Reactions were set up in 96-well plates by mixing the protein of interest in a total volume of 7 . 5μl TND with 147 . 5μl of the microtubule premix at room temperature . To start the assay , 10μl of 10mM ATP was added to each well . Final assay volume was 165μl of 12mM PIPES-KOH pH 7 , 4mM MgCl2 , 0 . 61mM ATP , and 14 . 5mM NaCl . This was then rapidly transferred to a 37°C plate reader ( Tristar LB 941; Berthold Technologies ) set to read absorbance at 360nm . Readings were acquired every 30 seconds during 1 hour . An inorganic phosphate standard curve was created in the same assay buffer and used to convert absorbance to nmol hydrolysed ATP . | Inborn heart defects can be divided into structural heart defects and diseases affecting the heart muscle , called cardiomyopathies . Congenital or neonatal cardiomyopathies are commonly associated with a poor prognosis and have multiple etiologies . In two siblings , a male and female , we identified an undescribed type of lethal congenital restrictive cardiomyopathy affecting the right ventricle . We hypothesized that this was most likely due to a novel autosomal recessive condition . To identify the cause , we used powerful genetic tools; linkage analysis combined to Whole Exome Sequencing ( WES ) , which analyses the protein coding parts of the human genome . A compound heterozygous mutation was found in KIF20A , a gene which has never been associated with human pathology previously . Further functional studies confirmed that the found variants resulted in a loss-of-function of KIF20A . Studies in zebrafish as an animal model were consistent with a role of KIF20A in cardiac development and function . These findings provide a functional link between cytokinesis and cardiomyopathy , opening a new mechanism for future research in genes involved in cell division . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] | [
"cardiomyopathies",
"fish",
"medicine",
"and",
"health",
"sciences",
"cardiovascular",
"anatomy",
"cardiac",
"ventricles",
"vertebrates",
"animals",
"fibroblasts",
"animal",
"models",
"osteichthyes",
"organisms",
"developmental",
"biology",
"mutation",
"model",
"organisms",... | 2018 | Compound heterozygous loss-of-function mutations in KIF20A are associated with a novel lethal congenital cardiomyopathy in two siblings |
Innate recognition of invading intracellular pathogens is essential for regulating robust and rapid CD4+ T cell effector function , which is critical for host-mediated immunity . The intracellular apicomplexan parasite , Toxoplasma gondii , is capable of infecting almost any nucleated cell of warm-blooded animals , including humans , and establishing tissue cysts that persist throughout the lifetime of the host . Recognition of T . gondii by TLRs is essential for robust IL-12 and IFN-γ production , two major cytokines involved in host resistance to the parasite . In the murine model of infection , robust IL-12 and IFN-γ production have been largely attributed to T . gondii profilin recognition by the TLR11 and TLR12 heterodimer complex , resulting in Myd88-dependent IL-12 production . However , TLR11 or TLR12 deficiency failed to recapitulate the acute susceptibility to T . gondii infection seen in Myd88-/- mice . T . gondii triggers inflammasome activation in a caspase-1-dependent manner resulting in cytokine release; however , it remains undetermined if parasite-mediated inflammasome activation impacts IFN-γ production and host resistance to the parasite . Using mice which lack different inflammasome components , we observed that the inflammasome played a limited role in host resistance when TLR11 remained functional . Strikingly , in the absence of TLR11 , caspase-1 and -11 played a significant role for robust CD4+ TH1-derived IFN-γ responses and host survival . Moreover , we demonstrated that in the absence of TLR11 , production of the caspase-1-dependent cytokine IL-18 was sufficient and necessary for CD4+ T cell-derived IFN-γ responses . Mechanistically , we established that T . gondii-mediated activation of the inflammasome and IL-18 were critical to maintain robust CD4+ TH1 IFN-γ responses during parasite infection in the absence of TLR11 .
Toxoplasma gondii is an obligate intracellular apicomplexan parasite capable of infecting most nucleated cells of warm-blooded animals , including humans , resulting in persistent cysts residing within the skeletal muscle , cardiac tissue , and brain [1–3] . Notably , in the United States toxoplasmosis is a leading cause of foodborne-related deaths [4 , 5] . A detailed understanding of the cellular and molecular mechanisms responsible for host resistance against T . gondii has been well-established in murine infection models [6] . Parasite recognition by TLRs is critical for IL-12 production by dendritic cells ( DCs ) , which is critical for a rapid and robust CD4+ T helper 1 ( TH1 ) response leading to production of IFN-γ [7–10] . During T . gondii infection both IL-12 and IFN-γ cytokines are essential for host resistance to T . gondii . The cytokine IFN-γ is critical for macrophage activation and induction of IFN-γ-inducible genes including p47 immunity related GTPases ( IRGs ) and p65 guanylate-binding proteins ( GBPs ) , which are required for intracellular parasite clearance [11–15] . Robust IL-12 production during mouse infection is largely attributed to direct recognition of T . gondii profilin by TLR11 which forms a heterodimer complex with TLR12 and leads to the activation of Myd88-dependent signaling pathways [16–21] . Unlike in mice , humans lack both these innate parasite ligand receptors; however , in most cases , humans that become infected with T . gondii are relatively resistant to this pathogen , unless patients become immunocompromised in CD4+ T cell effector functions [22 , 23] . At present , TLR11-independent mechanisms of T . gondii recognition required for establishing T cell immunity are largely unknown; however , several recent studies including our own have suggested that CCL2-mediated recruitment of monocytes and the inflammasome-dependent release of mature IL-1β and IL-18 represent the initial steps in driving a TLR11-independent protective immunity to the parasite [24–28] . Activation of inflammasome sensors results in recruitment of the adaptor molecule apoptosis-associated speck-like protein containing a CARD ( ASC ) that is essential for activating caspase-1 and for processing IL-1β and IL-18 into biologically active forms [29–31] . Data from several groups have identified that both the NLRP1 and NLRP3 inflammasomes play a role in caspase-1-dependent release of IL-1β in response to T . gondii invasion in vitro [26 , 28 , 32] . Additional studies have shown that NLRP3 , ASC , and IL-18 play a role in host resistance against T . gondii in vivo [32]; however , the precise T . gondii-derived stimuli that can initiate inflammasome activation remain undetermined . In this report , we set out to determine the mechanisms of TLR11-dependent and independent regulation of CD4+ TH1 response during T . gondii infection . Our data revealed that mice with individual deficiencies in TLR11 , NLRP3 , ASC , or caspase-1 and -11 ( Casp1/11 ) did not increase host mortality and had a minimal impact on CD4+ T cell-derived IFN-γ production . Strikingly , combined deficiency in TLR11 and Casp1/11 resulted in rapid susceptibility to parasitic infection caused by impaired T cell-derived IFN-γ responses during infection . Mechanistically , we revealed that that in the absence of TLR11 , inflammasome activation and IL-18-mediate CD4+ TH1-derived IFN-γ responses to T . gondii . Correspondingly , in the absence of Casp1/11 , TLR11-dependent IL-12 production is sufficient to generate robust TH1 responses in vivo . These results provided an explanation for the high susceptibility of Myd88-deficient mice , which is not observed in T . gondii infected TLR11- or Casp1/11-deficient mice . TLR11-dependent activation of Myd88 was sufficient for establishing TH1 immunity to the parasite due to large amounts of IL-12 necessary for TH1 polarization without inflammasome-dependent release of IL-18 . Similarly , IL-18 and possibly other Casp1/11-dependent mediators were capable of driving TH1 immunity and provided a partial protection against the parasite .
Inflammasome activation plays a major role in host defense against intracellular pathogens , including T . gondii , in part via inflammasome-dependent release of IL-1β and IL-18 [26 , 32 , 33] . Despite recent reports establishing that parasite infected cells can lead to the rapid activation of both the NLRP1 and NLRP3 inflammasomes [26 , 28 , 32 , 34 , 35] , the precise molecular and cellular events responsible for this activation in vivo remains undefined . To examine the role of the inflammasome in controlling host resistance to T . gondii in vivo , we assessed the susceptibility of mice lacking NLRP3 ( Nlrp3-/- ) , ASC ( Asc-/- ) , or Casp1/11-/- . We observed that upon intraperitoneal ( i . p . ) infection with cysts of the ME49 strain of T . gondii , mice deficient in the examined inflammasome components were relatively resistant to the acute stage of the infection ( Fig 1A ) . Moreover , analysis of parasite burden on day 30 post-infection revealed that the examined inflammasome activation was not essential for the control of T . gondii during the persistent chronic stage of infection , as measured by cyst burden in the brains of infected mice ( S1A Fig ) . Inflammasomes play an important role during the initial encounter between host cells and invading pathogens; therefore , we examined if the inflammasome was required for local parasite restriction at the site of infection . Our results revealed that while NLRP3-deficiency had no effects on the parasite burden , mice lacking either ASC or Casp1/11 exhibited elevated local parasite loads compared to WT and Nlrp3-/- mice ( Fig 1B ) . As a consequence , the absence of ASC or Casp1/11 , but not NLRP3 , also resulted in elevated T . gondii dissemination during the acute stage of infection , as indicated by an increased pathogen burden in the liver and spleen of the infected mice ( Fig 1B ) . These data revealed that , while the examined inflammasome components played a limited role in host survival during the acute stage of infection , both ASC and Casp1/11 participated in restriction of T . gondii tachyzoites during the acute stage of the infection . Nevertheless , the control of the parasite burden during the chronic stage of the infection in the brain did not require NLRP3 , ASC , or Casp1/11 . The cytokine IFN-γ is essential for parasite clearance and host resistance [8] , yet whether the inflammasome contributes to an IFN-γ response during parasite infection remains unclear . We first examined if IL-12 , a key innate cytokine that drives IFN-γ production in NK , ILC1 , and TH1 cells , was augmented by the inflammasome components NLRP3 , ASC , or Casp1/11 in T . gondii-infected mice . We observed that deficiency in NLRP3 , ASC , or Casp1/11 resulted in partially reduced circulating levels of IL-12 in the infected mice ( S1B Fig ) . Similarly , T . gondii infected Nlrp3-/- , Asc-/- , and Casp1/11-/- mice demonstrated modest reductions of IFN-γ within the sera on day 8 post-infection ( S1C Fig ) . In agreement with the ELISA data , IFN-γ transcripts in the spleen were reduced in the absence of ASC and Casp1/11 , further confirming a role for these inflammasome components in the regulation of immunity to T . gondii ( S1D Fig ) . Next , we examined if local IFN-γ induction was also compromised . Unexpectedly , IFN-γ transcript levels locally at the site of infection in the peritoneum were not reduced in any of the infected inflammasome-deficient mice , and the elevated IFN-γ transcript levels in Asc-/- mice was most likely caused by the increased pathogen burdens observed in those animals ( S1D Fig and Fig 1B ) . The cytokine IFN-γ plays a critical role in host defense by initiating a series of genes essential for parasite elimination and cell recruitment . Therefore , we examined if the absence of the inflammasome disrupted IFN-γ-mediated gene expression during T . gondii infection . In agreement with the largely unimpaired IFN-γ production at the site of infection , we observed robust induction of IFN-γ-inducible genes in both WT and inflammasome-deficient mice ( S1E Fig ) . Similarly , lower levels of splenic IFN-γ transcripts in the absence of the inflammasome resulted in a reduction of splenic Irgm3 and Cxcl10 transcripts ( S1F Fig ) . Thus , our data revealed a partial and limited role for ASC and Casp1/11 in the regulation of systemic IFN-γ responses during T . gondii infection . Unexpectedly , the examined inflammasome components were dispensable for the regulation of IFN-γ production and the expression of IFN-γ-induced genes at the site of infection . A potent TH1 response is a hallmark of T . gondii infection . It has been previously established that TLR-dependent production of IL-12 plays a major role in TH1 polarization , whereas IL-18 and IL-1β are able to enhance the robust IFN-γ production by CD4+ T cells [36–38] . However , it remains undetermined how the inflammasome impacts parasite-mediated TH1 effector function . Therefore , we assessed the contribution of the inflammasome in mediating the CD4+ TH1 response during T . gondii infection . We observed no reduction in the frequency or absolute cell numbers of peritoneal CD4+IFN-γ+ T cells in the absence of NLRP3 , ASC , or Casp1/11 by day 8 post-infection ( Fig 2A–2C ) . However , Asc and Casp1/11 deficiency resulted in a reduction in the amount of IFN-γ produced by peritoneal CD4+ TH1 cells ( Fig 2D ) . Similarly , we established that the examined inflammasome components NLRP3 , ASC , and Casp1/11 played no discernible role in the induction of the systemic TH1 response analyzed in spleens of the infected mice ( Fig 2E–2H ) ; however , deficiency in NLRP3 resulted in exacerbated frequency and total cell numbers of splenic CD4+IFN-γ+ T cells , and the amounts of IFN-γ produced by splenic CD4+ Th1s as measured by intracellular staining for this cytokine compared to Asc-/- and Casp1/11-/- mice ( Fig 2F–2H ) . In addition to CD4+ T cells , several other cell types contribute to IFN-γ production in response to T . gondii infection , including CD8+ T cells , NK cells , and neutrophils . Our data reveals that the examined inflammasome components played no significant role in the regulation of IFN-γ responses by CD8+ T cells , NK1 . 1+ NK cells , or Ly6G+ neutrophils locally or in the spleen of the infected mice compared to WT controls ( S2 Fig ) . Direct innate recognition of T . gondii profilin by the TLR11/12 heterodimer complex , regulates a major Myd88-dependent mechanism of IL-12 mediated immunity against T . gondii [6] . Therefore , we hypothesized that an intact TLR11 signaling pathway masks a role for inflammasomes in regulating TH1 immunity towards T . gondii and host resistance to the parasite due to its dominant effect on the activation of innate immune cells [39] . To test this hypothesis , we examined a role for Casp1/11 in the absence of TLR11 . Our results revealed that while neither TLR11 nor Casp1/11 alone were absolutely essential for host survival during acute toxoplasmosis , a combined deficiency in TLR11 and Casp1/11 ( TLR11xCasp1/11-/- ) resulted in rapid acute mortality , comparable to that seen in Myd88-/- mice ( Fig 3A ) . Analysis of the parasite burden revealed that while TLR11 was the primary mediator of parasite control at the site of infection , Casp1/11 plays an important role in cooperation with TLR11 in providing systemic Myd88-dependent immunity towards this intracellular pathogen ( Fig 3B ) . The rapid susceptibility of TLR11xCasp1/11-/- mice suggests that Casp1/11 is required for strong induction of IFN-γ in the absence of TLR11 during parasite infection . This was evident from the analysis of IFN-γ expression in peritoneal cavity and spleen of the infected TLR11xCasp1/11-/- mice ( Fig 3C ) . While both TLR11 and Casp1/11 cooperate in host resistance to T . gondii , these signaling pathways play distinct roles in programming host defense . We observed that TLR11 , but not Casp1/11 played a role in the regulation of systemic IL-12 responses . This was evident from the analysis of IL-12/23p40 detected in the sera of infected TLR11-/- and TLR11xCasp1/11-/- mice ( Fig 3D ) . Consistent with a role for Casp1/11 in the regulation of the circulating IFN-γ , we also observed a significant reduction in serum IFN-γ levels detected in TLR11xCasp1/11-/- mice compared to TLR11-deficienct animals ( Fig 3D ) . These data strongly suggest that in the absence of TLR11 , the Casp1/11-dependent inflammasome pathway significantly contributes to controlling T . gondii and systemic IFN-γ production during infection . We then examined if the Casp1/11-dependent cytokines , IL-18 and IL-1β , were playing a compensatory role in the absence of TLR11-mediated IL-12 during T . gondii infection . We observed on day 5 of parasite infection circulating IL-18 was significantly elevated in TLR11-/- mice compared to both WT and TLR11xCasp1/11-/- mice ( S3A Fig ) . Moreover , TLR11-/- mice sustained these elevated levels of IL-18 compared to TLR11xCasp1/11-/- mice by day 8 post-infection ( S3A Fig ) . Contrariwise , we were unable to detect any circulating IL-1β on days 5 and 8 of T . gondii infection in WT , TLR11-/- , TLR11xCasp1/11-/- mice ( S3B Fig ) . Overall , our results revealed a pivotal role for Casp1/11 in controlling T . gondii and systemic IFN-γ production in the absence of TLR11 . Among the several classes of innate immune receptors , TLRs are known to play a central role in the regulation of TH1 effector choices . At the same time , both IL-1 and IL-18 are known to be required for the sustained IFN-γ production by CD4+ T cells and the combined deficiency in TLR and IL-1R signaling contributes to a compromised TH1 immunity seen in Myd88-deficient mice infected with the intracellular pathogens [10] . We next investigated the relative contribution of TLR11 and Casp1/11 in the regulation of TH1 immunity to T . gondii . Infection of TLR11-/- mice unexpectedly revealed TLR11-deficiency played no obvious role in the initiation of a CD4+ TH1 response towards T . gondii . This was in sharp contrast to TLR11xCasp1/11-deficient mice that demonstrated a profound reduction in both frequency and absolute numbers of CD4+IFN-γ+ T cells at the site of infection ( Fig 4A–4C ) . Furthermore , Casp1/11-deficiency in the absence of TLR11 resulted in a significant reduction of IFN-γ production among peritoneal CD4+ T cells as measured by intracellular staining for this cytokine ( Fig 4D ) . Both the frequencies of CD4+IFN-γ+ and the amounts of IFN-γ produced by CD4+ T cells were similar in TLR11xCasp1/11-/- and Myd88-/- mice ( Fig 4B and 4D ) , strongly suggesting that both of these signaling pathways cooperate in a Myd88-dependent manner that is essential for TH1 effector function during T . gondii infection . Analogous to our results from the peritoneum , our data showed both frequency and total cell number of splenic CD4+IFN-γ+ T cells , and the amounts of IFN-γ produced by CD4+ T cells were not compromised by a single TLR11- or Casp1/11-deficiency ( Figs 2E–2H and 4E–4H ) . Instead , deficiency in both TLR11 and Casp1/11 resulted in the reduced frequencies and total TH1 cells , and the amounts of IFN-γ produced by CD4+ T cells ( Fig 4E–4H , S4 Fig ) . Additionally , in TLR11xCasp1/11-/- mice we observed a reduction in the absolute numbers of IFN-γ producing peritoneal CD8+ T cells and NK cells , similar to Myd88-/- animals ( S5A Fig ) . However , a reduction in the absolute numbers of IFN-γ producing splenic CD8+ T cells and NK cells was only observed in Myd88-/- mice ( S5B Fig ) . Overall , the analysis of TLR11- and Casp1/11-deficient mice allowed us to reveal that inflammasome activation plays a major role in the regulation of TH1 immunity when TLR11-dependent recognition is compromised . Our results also demonstrated that T . gondii activation by the inflammasome Casp1/11-dependent pathway is sufficient for Myd88-dependent activation of TH1 immunity . This provided an explanation for largely unimpaired activation of CD4+ T cells observed in TLR11-deficient mice infected with the parasite . Inflammasome recognition of intracellular pathogens is essential for the release of mature forms of IL-1β and IL-18 . The cytokine IL-18 is critical for the production and secretion of IFN-γ by immune cells [36 , 38 , 40] . Therefore , we hypothesized in the absence of TLR11 , IL-18 is required for robust CD4+ TH1-derived IFN-γ responses during parasite infection . To determine if the compromised CD4+ T cell-derived IFN-γ response of TLR11xCasp1/11-/- mice is a result of IL-18 deficiency , we neutralized either IL-1β or IL-18 in TLR11-/- mice during T . gondii infection . Our results revealed IL-1β does not significantly contribute to the amounts of IFN-γ produced by TH1s in TLR11-deficient mice ( S6 Fig ) . Simultaneously , antibody-mediated blocking of IL-18 in TLR11-/- mice resulted in a dramatic reduction in the amounts of IFN-γ produced by both peritoneal and splenic CD4+ T cells , similar to TLR11xCasp1/11-/- mice ( S6 Fig ) . To confirm IL-18 is critical to augment CD4+ T cell-derived IFN-γ production in the absence of TLR11 , we administered IL-18 to TLR11xCasp1/11-/- mice during T . gondii infection and examined their CD4+ TH1 responses . Our results revealed administering IL-18 to TLR11xCasp1/11-/- mice during parasite infection dramatically augmented the frequency and absolute numbers of CD4+IFN-γ+ T cells both locally in the peritoneum and in the spleen , compared to non-treated controls ( Fig 5 , S7 Fig ) . Our data also demonstrate that TLR11xCasp1/11-/- mice given IL-18 during infection augments the absolute numbers of peritoneal and splenic CD8+IFN-γ+ T cells , IFN-γ+ NK cells , and IFN-γ+ neutrophils ( S8A and S8B Fig ) . Finally , our results demonstrate that IL-18 does contribute in reducing parasite burden both at the site of infection and in peripheral tissue ( S8C Fig ) , but failed to rescue the survival of TLR11xCasp1/11-/- mice infected with the parasite ( S8D Fig ) . Our data revealed that in the absence of TLR11 , inflammasome activation and IL-18 is sufficient and necessary in augmenting peritoneal and splenic IFN-γ production by CD4+ and CD8+ T cells along with NK cells during T . gondii infection and limiting pathogen burden . Nevertheless , other Casp1/11-dependent cytokines are required for host survival during T . gondii infection .
Activation of inflammasomes by pathogen stimuli recognition results in the oligomerization and recruitment of the adaptor molecule ASC , caspase-1 activation , release of the active forms of IL-1β and IL-18 , and induction of the cell-mediated death pathway , pyroptosis . Accordingly , pathogen ligand recognition by inflammasomes is considered an essential host innate immune pathway to identify invading microbes . Recent reports have demonstrated that T . gondii is recognized by the NLRP1 and NLRP3 inflammasomes resulting in IL-1β and IL-18 production both in vitro and in vivo [26 , 32] . It has also been established that TLR recognition of the parasite by DCs is required for IL-12 production [16 , 18 , 20] . The T . gondii-driven IL-12 production is essential for generating a robust CD4+ TH1-derived IFN-γ response , resulting in the induction of IFN-γ-mediated genes required for parasite clearance [11 , 12 , 41–43] . Additionally , previous reports have shown that the inflammasome-dependent cytokine IL-18 can work in synergy with IL-12 during T . gondii infection , augmenting IFN-γ responses and contributing to parasite restriction [32 , 37 , 38] . Hence , we hypothesized that a deficiency in NLRP3 , ASC , or Casp1/11 would significantly abrogate CD4+ T cell-derived IFN-γ responses , resulting in rapid host mortality . Unexpectedly , our results did not support the original hypothesis , instead revealing that inflammasome-deficient mice do not have increased mortality rates during parasite infection and only have a modest reduction of IL-12 and IFN-γ . Additionally , we observed no change in the frequency or absolute cell numbers of peritoneal or splenic CD4+ TH1 cells . Thus , our data demonstrates that the inflammasome plays a limited role in parasite restriction and is dispensable for murine host resistance against T . gondii when TLR11-mediated immunity remains intact . Two previous studies have reported contradicting results as to the role of the inflammasome in host immunity towards T . gondii . The data reported by Hitziger and colleagues are consistent with our own results , indicating inflammasome deficiency alone does not play a major role in parasite restriction [44] . However , Gorfu and colleagues’ data indicate that the absence of either the NLRP1 or NLRP3 inflammasome results in increased parasite burden and host mortality [32] . While additional work is required to resolve these inconsistencies , a largely dispensable role for caspase-1 and -11 in the survival of T . gondii infected mice strongly suggest a limited role for inflammasomes alone in host resistance to the parasite when TLR11-sensing remains intact . Innate recognition of T . gondii by TLRs is critical for robust IL-12 production and host defense . Consequently , it has been shown that TLR11 and TLR12 , the only known innate receptors to uniquely recognize T . gondii profilin and who significantly contribute to IL-12 production , are not absolutely required for host resistance or for TH1 effector function [16 , 18 , 20] . Nevertheless , it is important to note that we observed a higher susceptibility of TLR11-/- mice than previously reported [16] , as on average half of the infected mice succumb to T . gondii during an acute stage of the infection . Considering that the same ME49 parasites were used in both studies , the difference in susceptibility to T . gondii may in part be caused by distinct microbiota that regulate immunity to the parasite and other pathogens [45 , 46] . Analysis of UNC93B1-deficient mice , which lack all endosomal TLR signaling , including TLR11 and TLR12 , resulted in diminished IL-12 responses and a delayed T . gondii-mediated IFN-γ response . Yet , this delayed TLR-independent parasite-driven IFN-γ response is insufficient for host protection [18 , 20] . Host immunity in these TLR or UNC93B1 knockouts can be rescued by administering exogenous IL-12 [18 , 20] . Paradoxically , Myd88-deficient mice , are not only highly susceptible to parasite infection , but cannot be rescued with exogenous administration of IL-12 , strongly suggesting that signaling from both IL-12 and the inflammasome-mediated cytokines IL-1β and IL-18 are required for host resistance against T . gondii [10] . Furthermore , it has been reported that both IL-1β and IL-18 play a role in augmenting IFN-γ production during T . gondii infection [37 , 47] . Hence , we hypothesized that in the absence of TLR recognition , inflammasome activation is required for host resistance against T . gondii . The results from TLR11xCasp1/11-/- mice demonstrated that in the absence of both robust IL-12 and IL-18 production , CD4+ T cell-derived IFN-γ responses were compromised and insufficient for host protection resulting in uncontrolled parasite replication . Therefore , since TLR11xCasp1/11-/- mice have intact IL-18R signaling , we examined if exogenous IL-18 administered to TLR11xCasp1/11-/- mice during parasite infection would be sufficient to restore CD4+ TH1-derived IFN-γ responses . Our results demonstrated that in the absence of TLR11 , IL-18 is sufficient and necessary for robust CD4+ T cell-derived IFN-γ responses . Hence , our data establishes two significant attributes of inflammasome activation during parasite infection: 1 ) in the absence of TLR11 recognition , the inflammasome and IL-18 is critical for robust TH1 effector function; and 2 ) they provide pivotal evidence as to why IL-12 alone is insufficient to rescue Myd88-/- mice during T . gondii infection . Our findings illustrate the mechanism by which triggering inflammasome activation in the absence of TLR11 enables host resistance against T . gondii infection . Dissimilar to murine innate recognition of T . gondii , humans lack a functional TLR11 , and TLR12 is absent from the human genome [48]; however , parasite infection in immunocompetent individuals is generally asymptomatic , indicating TLR11- and TLR12-independent innate recognition of T . gondii is sufficient for human immunity against the parasite . Both CCL2-dependent recruitment of human monocytes and NLRP1 , NLRP3 , ASC , and caspase-1-dependent release of IL-1β and IL-18 by these cells suggest that T . gondii-triggered activation of the inflammasome is required for resistance against the parasite when TLR11-dependent immunity is absent [24 , 28 , 49] . Unlike for TLR11 and TLR12 , the precise T . gondii ligand that is recognized by inflammasomes remains unknown . However , reports have demonstrated that heat killed T . gondii or mycalolide B-treated parasites are unable to trigger IL-1β release , indicating that active parasite invasion is required for inflammasome-mediated cytokine release [24 , 32] . Intriguingly , while it has been demonstrated that T . gondii induces both pyroptosis and IL-1β secretion in rat macrophages , mouse macrophages do not undergo pyroptosis during parasite invasion in vitro [27 , 50] . Additionally , reports have indicated that predominantly type II strains of the parasite , such as ME49 , which was used in this current study , promote substantial cytokine release compared to the majority of other identified T . gondii strains [24 , 32] . It has also been shown that the dense granule protein , GRA15 , from type II strains of the parasite plays an important role for IL-1β release in both human monocytes and mouse bone marrow-derived macrophages [24 , 32] . It has yet to be determined if GRA15 is directly recognized by inflammasome sensors resulting in caspase-1 activation and IL-1β release; or if GRA15 , a known inducer of NF-κB activation , leads to the induction of immature IL-1β and a second T . gondii-dependent stimulus triggers caspase-1-dependent cytokine release [51] . Furthermore , recent reports show that T . gondii-mediated potassium efflux can trigger the rapid release of IL-1β in infected monocytes [28] . While the precise T . gondii stimuli that triggers inflammasome activation remains unknown , our results established that inflammasome activation and IL-18 are required for host protection and robust CD4+ T cell-derived IFN-γ responses in the absence of TLR11 .
C57BL/6 , Nlrp3-/- , and Casp1/11-/- mice were obtained from Jackson Laboratory ( Bar Harbor , ME ) . Asc-/- , TLR11-/- and Myd88-/- mice have been previously described [16 , 52 , 53] . TLR11-/- mice were crossed with Casp1/11-/- mice to generate TLR11xCasp1/11-/- . All control and experimental mice were age- and sex-matched within all individual experiments . This study included both male and female mice , and the data derived from male and female mice identified no sex-specific differences in the performed experiments . All mice were maintained at in the pathogen-free American Association of Laboratory Animal Care-accredited animal facility at the University of Rochester Medical Center , Rochester , NY . All animal experimentation was conducted in accordance with the guidelines of the University of Rochester’s University Committee on Animal Resources ( UCAR ) , the Institutional Animal Care and Use Committee ( IACUC ) . All mice were i . p . infected with an average of 20 T . gondii cysts of the ME49 strain . At day 8 post-infection , the animals were necropsied . In some experiments , mice were injected i . p . with 200 ng of IL-18 ( R&D ) on days 0 , 1 , 2 , and 3 , or mice were injected with 200 μgs of anti-IL-1β or anti-IL-18 ( BioXCell ) on days 0 , 2 , 4 , and 6 . Total RNA was isolated from the peritoneal exudate cells and the spleens of naïve or T . gondii infected mice using Trizol . cDNA was prepared using iScript cDNA Synthesis Kit ( Bio-Ras , Hercules , CA ) . Optimized primers targeting each gene were designed using Primer3 Software [54] . These primers included the following: IFN-γ ( 5’-ACTGGCAAAAGGATGGTGAC-3’ , 3’-TGAGCTCATTGAATGCTTGG-5’ ) , Irgm3 ( 5’-CTGGAGGCAGCTGTCAGCTCCGAG-3’ , 3’-GTCCTTTAGAGCTTTCCTCAGGGAGGTCTTG-5’ ) , Cxcl10 ( 5’-GACGGTCCGCTGCAACTG-3’ , 3’-GCTTCCCTATGGCCCTCATT-5’ ) , and HPRT ( 5’-gcccttgactataatgagtacttcagg-3’ , 5’-ttcaacttgcgctcatcttagg-3’ ) [55] . cDNA was amplified using SSOFast Eva Green Supermix ( BioRad ) . A MyiQ Real-Time PCR Detection System ( BioRad ) was used to obtain Ct values . The relative expression of samples was determined after normalization to HPRT using ddCt method [56] . To determine T . gondii pathogen loads , total genomic DNA from animal tissue was isolated by using the DNeasy Blood and Tissue Kit ( Qiagen ) according to manufacturer’s instructions . PCR were performed by using SSOFast Eva Green Supermix ( BioRad ) . Samples were measured by qPCR using a MyiQ Real-Time PCR Detection System ( BioRad ) , and data from genomic DNA was compared with a defined copy number standard of the T . gondii gene B1 . The IFN-γ ( ThermoFisher ) , IL-12/23p40 ( ThermoFisher ) , IL-1β ( ThermoFisher ) , and IL-18 ( R&D ) concentration in the sera or cell supernatant was analyzed by standard sandwich ELISA kit according to manufacturer’s instructions . To assay the responses of mice infected with T . gondii the PECs and spleens were harvested from C57BL/6 , Nlrp3-/- , Asc-/- , Casp1/11-/- , TLR11-/- , TLR11xCasp1/11-/- , and Myd88-/- mice and on day 8 post-infection . Single-cell suspension of PECs and spleens were restimulated with PMA ( 20ng/mL ) and ionomycin ( 1μg/mL ) ( Sigma-Aldrich ) for 4 hours in the presence of GolgiPlug ( Brefeldin A , BD Biosciences ) . Alternatively , bone marrow-derived DCs ( BMDCs ) were generated in the presence of GM-CSF ( Invitrogen ) and plated at 2 . 5x106 cells/well in a 96-well plate and pulsed with frozen ME49 tachyzoite antigen for 18 hours and then 2 . 5x106 splenic T cells were added for 6 hours in the presence of GolgiPlug . After isolation or in vitro restimulation , the cells were washed once in phosphate-buffered saline + 1% fetal bovine serum and stained with fluorochrome-conjugated antibodies . Cell fluorescence was measured using an LSRII flow cytometer , and data were analyzed using FlowJo Software ( Tree Star ) . All data were analyzed with Prism ( Version 7; GraphPad ) These data were considered statistically significant when P-values were <0 . 05 . | It is currently estimated that one third of the world’s population is seropositive for the parasite Toxoplasma gondii and this parasite can lead to serious illness and death in immunocompromised patients , and is one of the leading causes of foodborne-related deaths in the United States . Host immunity against the parasite has largely been attributed to recognition of the parasite-derived protein , profilin , by the innate Toll-like receptors ( TLRs ) , TLR11 and TLR12 . T . gondii also triggers inflammasome activation in a caspase-1-dependent manner resulting in cytokine release . However , how these innate recognition systems regulate TH1 immunity and host resistance remains largely unknown . Therefore , using genetically modified mice , we investigated TLR11-dependent and -independent host immunity against the parasite . Our research establishes that in the absence of TLR11 , inflammasome activation and subsequent production of the inflammasome-dependent molecule , IL-18 are critical for host immunity to the parasite . These data provide novel mechanistic insight into how TLR and inflammasomes cooperate in regulation of TH1 immunity and host protection . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] | [
"blood",
"cells",
"innate",
"immune",
"system",
"medicine",
"and",
"health",
"sciences",
"immune",
"cells",
"immune",
"physiology",
"cytokines",
"toxoplasma",
"gondii",
"spleen",
"immunology",
"cell-mediated",
"immunity",
"parasitic",
"diseases",
"parasitic",
"protozoan... | 2019 | TLR11-independent inflammasome activation is critical for CD4+ T cell-derived IFN-γ production and host resistance to Toxoplasma gondii |
Mosquito-borne alphaviruses such as chikungunya virus and Ross River virus ( RRV ) are emerging pathogens capable of causing large-scale epidemics of virus-induced arthritis and myositis . The pathology of RRV-induced disease in both humans and mice is associated with induction of the host inflammatory response within the muscle and joints , and prior studies have demonstrated that the host complement system contributes to development of disease . In this study , we have used a mouse model of RRV-induced disease to identify and characterize which complement activation pathways mediate disease progression after infection , and we have identified the mannose binding lectin ( MBL ) pathway , but not the classical or alternative complement activation pathways , as essential for development of RRV-induced disease . MBL deposition was enhanced in RRV infected muscle tissue from wild type mice and RRV infected MBL deficient mice exhibited reduced disease , tissue damage , and complement deposition compared to wild-type mice . In contrast , mice deficient for key components of the classical or alternative complement activation pathways still developed severe RRV-induced disease . Further characterization of MBL deficient mice demonstrated that similar to C3−/− mice , viral replication and inflammatory cell recruitment were equivalent to wild type animals , suggesting that RRV-mediated induction of complement dependent immune pathology is largely MBL dependent . Consistent with these findings , human patients diagnosed with RRV disease had elevated serum MBL levels compared to healthy controls , and MBL levels in the serum and synovial fluid correlated with severity of disease . These findings demonstrate a role for MBL in promoting RRV-induced disease in both mice and humans and suggest that the MBL pathway of complement activation may be an effective target for therapeutic intervention for humans suffering from RRV-induced arthritis and myositis .
Arthritogenic alphaviruses , such as Ross River virus ( RRV ) and chikungunya virus ( CHIKV ) , are mosquito-borne viruses that cause severe polyarthritis and myositis in humans . RRV causes annual disease outbreaks in Australia and has caused sporadic epidemics of debilitating polyarthritis , including one outbreak involving over 60 , 000 people in Oceania [1] . RRV is transmitted to humans primarily by the Aedes and Culex species of mosquitoes that generally populate marsh areas , and CHIKV transmission has been traditionally mediated by the urban Aedes aegypti , though the virus has recently adapted for efficient transmission by the widely distributed Aedes albopictus species [2] , leading to an increased risk for CHIKV spread into new areas , as illustrated by recent outbreaks in Italy and southern France [3] . The expansion of CHIKV into an additional mosquito vector and the subsequent epidemic has highlighted the ability of the arthritic alphaviruses to move into new geographic areas and cause large-scale outbreaks of acute and persistent arthralgia and myalgia in humans . RRV-induced arthritic disease presents predominantly as painful stiffness , inflammation , and swelling in peripheral joints that can last months after initial infection and the host inflammatory response is thought to play a major role in disease pathogenesis . Inflammatory monocytes constitute the bulk of leukocytes isolated in synovial aspirates from RRV-infected patients [4] , [5] , and macrophage-cytotoxic drugs have been shown to drastically reduce disease progression and severity in mice [6] , [7] . In addition , mice lacking C3 , the central complement factor that is essential for complement activation , exhibit reduced RRV-induced disease and tissue destruction [8] , implicating a role for complement in development of the disease . Consistent with studies in mice , synovial aspirates from patients with RRV-induced arthritis have been shown to contain increased levels of the C3 cleavage product C3a [8] . Although macrophage recruitment to infected tissues is markedly increased after RRV infection , the role played by complement is independent of inflammatory cell recruitment . Rather , tissue destruction and disease progression requires complement receptor 3 ( CR3 ) , suggesting that complement interactions with CR3 on inflammatory cells promote tissue destruction in RRV-infected tissues [9] . However , it is currently unclear how the complement system is activated following RRV infection . There are three main activation pathways of the complement cascade; the classical , alternative , and lectin dependent pathways , that all converge on factor C3 and lead to activation of complement effector functions ( reviewed in [10] ) . The classical pathway is initiated by C1q interactions with antigen-bound complexes of IgG and IgM , and the proteases C1r and C1s cleave C4 and C2 to generate the C3 convertase C4b2b . Binding of factor B ( fB ) and spontaneously hydrolyzed C3 initiates the alternative pathway , and fB binding to C3b leads to formation of the alternative C3 convertase C3bBb that can amplify complement activation . In the lectin pathway , mannose binding lectin ( MBL ) or the ficolins bind to carbohydrate moieties on foreign bodies , such as viruses , or to host cells and apoptotic cells , and the MBL-associated serine proteases ( MASPs ) cleave C4 and C2 to form the C3 convertase C4b2b . Cleavage and processing of C3 by the C3 convertases produce several C3-derived components that are potent activators of the immune system . One such component is iC3b , which is a ligand for several complement receptors , such as CR3 . Binding of iC3b to CR3 on cells such as monocytes/macrophages , neutrophils , and NK cells , results in activation of these cells , leading to enhanced phagocytosis and cytotoxic activity against iC3b-opsonized cells [10] . MBL is a soluble C-type lectin that can initiate the complement cascade through binding of the carbohydrate recognition domains ( CRD ) to cell-surface sugars expressed on bacteria and viruses and some endogenous host ligands ( reviewed in [11] ) . The complement system is notorious for having both a protective and pathologic role and frequently leads to additional tissue injury and damage once activated . Similarly , MBL appears to be able to have the ability to protect as well as harm the host cells . In the context of sterile inflammatory diseases such as myocardial and gastrointestinal ischemic reperfusion injury , MBL and the lectin pathway mediate development of disease through complement-mediated regulation of pro-inflammatory cytokines and inflammation , leading to exacerbated pathology and tissue injury [12]–[15] . In contrast to the pathologic role of MBL in sterile inflammatory diseases , MBL is thought to primarily play a protective role in response to infectious pathogens . MBL has been shown to be essential for host protection from many different viral and bacterial infections either through direct binding to pathogens or by limiting spread through complement effector functions . MBL has been shown to bind directly to many different viruses , including human immunodeficiency virus ( HIV ) , Ebola virus , and arboviruses such as dengue virus and West Nile virus ( WNV ) , and MBL can either directly neutralize these viruses through activation of complement or interfere with their binding to host cells ( reviewed in [16] ) . Furthermore , studies using mice deficient in both MBL genes ( MBL-A−/− and MBL-C−/−; MBL-DKO ) have revealed that MBL can have a protective role during WNV and herpes simplex virus infections [17]–[19] . Though MBL neutralizes flaviviruses , such as WNV and dengue virus [18] , and plays a protective role during WNV infection [19] , these same studies found no detectable interactions between MBL and alphaviruses , suggesting that MBL does not play a protective role during alphavirus infection . However , the role of MBL role in the pathogenesis of alphavirus-induced arthritis/myositis has not been evaluated . The goal of this study was to further assess the role of the host complement system in the pathogenesis of alphavirus-induced inflammatory disease and to determine which complement activation pathways are required for virus-induced disease . In a mouse model of RRV-induced arthritis and myositis , mice deficient in either the classical or alternative pathways developed severe disease , while mice deficient in both genes of MBL ( MBL-DKO ) were resistant to disease , suggesting that MBL plays a major role in RRV-induced disease . Similar to previous findings with C3−/− mice [8] , RRV-infected MBL-DKO mice had similar levels of viral burden and inflammation compared to wild-type ( WT ) but exhibited significantly less complement deposition , tissue damage , and disease . Further analysis found that MBL levels are enhanced in RRV infected tissues and that MBL binds to RRV infected cells , suggesting that RRV infection leads to MBL deposition and subsequent complement activation . Importantly , studies in human patients suffering from RRV-induced disease found that levels of MBL were elevated in the serum of RRV-infected patients compared to healthy controls . In addition , serum and synovial fluid MBL levels correlated with the severity of RRV disease , while no differences were observed in classical or alternative pathway activation , suggesting that MBL contributes to RRV-induced disease in human populations .
Complement activation products are elevated in the synovial fluid of persons suffering from RRV-induced arthritis and complement activation is required for virus-induced arthritis/myositis in a mouse model [8] , [9] , [20] . Although other alphaviruses , such as the neurovirulent Sindbis virus , have been shown to activate complement via both the classical and alternative pathways [21] , the pathway ( s ) leading to complement activation by arthritic alphaviruses is currently unknown . Therefore , mice deficient in key components of the classical ( C1q−/− ) , alternative ( factor B , fB−/− ) , or lectin ( MBL-A/C−/− , MBL-DKO ) pathways were assessed for their susceptibility to RRV-induced disease . C1q−/− , fB−/− , MBL-DKO , or WT C57BL/6 mice were inoculated with RRV in the footpad and assessed for weight loss and scored for hind limb function as previously described [20] . RRV-infected WT mice showed signs of hind-limb weakness by 5 days post infection ( dpi ) and developed severe hind-limb weakness by 7 dpi through 10 dpi ( Figure 1A ) . RRV causes disease independently of B cells and antibody [20] , suggesting that the classical pathway does not contribute to disease during RRV infection . Consistent with this , C1q−/− mice exhibited severe disease signs and hind-limb weakness similar to WT animals ( Figure 1A ) . Likewise , fB−/− mice developed severe RRV-induced disease ( Figure 1A ) , demonstrating that the alternative pathway of complement activation is not required for RRV-induced disease , though it is important to note that RRV-infected fB−/− mice tended to develop more severe disease compared to WT mice , suggesting that the alternative pathway is activated and may actually play a protective role during RRV infection . In contrast to C1q−/− and fB−/− mice , MBL-DKO mice infected with RRV developed mild hind-limb weakness and exhibited reduced weight loss compared to WT mice ( Figure 1 , A and B ) . Mock-infected WT and MBL-DKO mice did not differ in weight gain ( Figure S1 ) and showed no signs of disease throughout the course of infection ( data not shown ) . While we cannot rule out a minor contribution of the classical and alternative activation pathways or activation of the lectin pathway through ficolins in development of RRV disease , this data demonstrates that the lectin pathway initiated by MBL plays an essential role in driving RRV-induced disease . Following infection of WT C57BL/6 mice , RRV replicates to high levels within both the joints and skeletal muscle and elicits an inflammatory infiltrate into these tissues [20] . Following the onset of inflammatory cell infiltration , wild type mice develop severe destructive myositis , which is a major aspect of virus-induced disease in this mouse model , and we have previously shown that muscle cell killing and disease is dependent upon both C3 activation and CR3 [8] , [9] . Therefore , to confirm the role of MBL in driving RRV-induced disease , inflammatory pathology was assessed within the quadriceps muscles of RRV-infected fB−/− , C1q−/− , MBL-DKO , or WT mice by H&E staining of paraffin embedded sections . At 10 dpi , a time point of peak RRV disease , we observed similar inflammation and tissue pathology in fB−/− , C1q−/− , and WT mice ( Figure 1C ) . Inflammatory cells are present in the quadriceps muscles of infected mice from all strains , as indicated by the solid arrowheads . We observed tissue damage in the quadriceps muscles in fB−/− , C1q−/− , and WT mice as evidenced by the degeneration of the fibrous architecture of the skeletal muscle ( marked by open arrowheads ) . In contrast , RRV infected MBL-DKO mice maintained the architecture of the skeletal muscle with very little tissue damage , despite the presence of inflammatory cells ( Figure 1C ) , which was strikingly similar to previous results demonstrating an essential role for C3 in RRV-induced disease [8] . To confirm that MBL-DKO mice have decreased tissue damage following RRV infection compared to WT mice , we used Evans Blue dye ( EBD ) uptake to detect areas of damage . Consistent with the clinical scores and histological analyses at 10 dpi , RRV-infected WT mice had abundant EBD positive muscle fibers within the quadriceps muscle whereas EBD positive cells were rare in RRV-infected MBL-DKO mice ( Figure 1D ) , further demonstrating that MBL is required for the induction of tissue damage during RRV-induced disease . To determine if MBL is deposited onto tissues following RRV infection , we evaluated tissue homogenates of quadriceps muscle from RRV-infected WT mice at 7 dpi for levels of MBL by immunoblot analysis . We observed an increase in the amount of MBL in the quadriceps muscle of infected mice compared to that of mock-infected mice ( Figure 2A ) , indicating that RRV infection results in elevated amounts of MBL within target tissues . Given the enhanced MBL within RRV infected muscle tissue , we next evaluated whether MBL would directly bind to RRV or RRV infected cells . Studies with two other alphaviruses , CHIKV and Sindbis virus , found no evidence for interactions with MBL , while mosquito derived West Nile virus was efficiently bound and neutralized by MBL [18] . Consistent with these findings , we were unable to detect direct binding between MBL and either mammalian cell or mosquito cell derived RRV virions by ELISA ( data not shown ) and we also found no evidence for MBL-mediated neutralization of RRV ( Figure S2 ) . Given the lack of detectable interactions between MBL and the RRV virion , we assessed whether MBL bound to RRV-infected cells . Differentiated C2C12 murine skeletal muscle cells were infected with RRV for 24 hours and then incubated with medium containing either C57BL/6 wild-type serum or MBL-DKO serum for 30 minutes . Cell lysates were harvested and analyzed for presence of MBL-C by immunoblot analysis . As shown in Figure 2B , the amount of MBL-C deposition was enhanced in cells infected with RRV compared to mock infected cells , indicating that RRV infection results in increased MBL binding to cells . No deposition of MBL was detected onto cells incubated with MBL-DKO serum , indicating the specificity of detection . Therefore , though MBL does not appear to interact with the RRV virion , RRV infection does lead to MBL deposition on infected cells . MBL , but not the alternatively or classical complement activation pathways , was required for RRV-induced disease and tissue pathology ( Figure 1 ) , and the phenotype in MBL-DKO mice is strikingly similar to C3−/− mice , suggesting that MBL plays a major role in driving complement activation during RRV infection . Therefore , we directly assessed whether MBL was required for RRV-dependent complement deposition . Western blot analysis of skeletal muscle from wild type or MBL-DKO mice indicated that C3 levels , including the α and β chains of C3 were present at reduced levels in the skeletal muscle of RRV infected MBL-DKO mice compared to wild type mice ( Figure S3A–B ) . However , since inflammatory macrophages produce C3 [22] , western blot analysis was not able to clearly differentiate between complement deposition within the tissue and de novo production of complement by the infiltrating inflammatory cells in both wild type and MBL-DKO animals . Therefore , we directly assessed the impact of MBL deficiency on complement deposition within the RRV infected muscle by performing immunohistochemistry on quadriceps muscle from WT and MBL-DKO mice using an anti-mouse C3 antibody . Abundant C3 staining localized to damaged skeletal muscle at 7 dpi in WT mice while C3 staining was substantially reduced in muscle from RRV-infected MBL-DKO mice ( Figure 2C ) . Importantly , we observed comparable C3 staining between RRV-infected C1q−/− , fB−/− , and WT mice ( Figure S3C ) , suggesting that neither C1q nor fB are required for C3 deposition on muscle tissue following RRV infection . Therefore , these results suggest that MBL is the major mediator of complement activation and deposition within RRV infected muscle tissue . Prior studies with C3−/− and CR3−/− mice demonstrated that complement activation and CR3-dependent signaling is essential for RRV-induced disease and tissue destruction , but complement deficiency had no effect on viral burden or tropism . To determine whether this was also the case in MBL-DKO mice , we evaluated WT and MBL-DKO mice for viral load within the quadriceps muscles , ankle joints , and serum . As shown in Figure 3A , MBL-DKO mice exhibited no significant difference in the amount of infectious virus in the quadriceps muscle through days 7 and 10 dpi , which represent the times when RRV-induced muscle destruction peaks . Furthermore , analysis of the viral distribution within wild type and MBL-DKO animals by in situ hybridization found no differences in the localization of RRV specific signal between the two mouse strains . ( Figure 3D ) . Therefore , the differences in RRV-induced tissue destruction ( Figure 1C–D ) or C3 deposition ( Figure 2C ) within the RRV infected muscle of MBL-DKO mice cannot be explained by differences in viral replication . In addition to evaluating viral titers within the skeletal muscle , we also assessed viral loads within the serum and ankle joints . Viral titers within the ankle joints were similar between MBL-DKO and WT mice through 7 dpi ( Figure 3B ) , though we did observe a small , but statistically significant decrease in viral titer within the ankle joints of MBL-DKO mice compared to wild type mice at day 10 post infection . MBL-DKO mice had higher amounts of virus in the serum at 1 dpi compared to WT mice ( Figure 3C ) , indicating that MBL may play some role in initial control of viremia . However , virus was cleared from the serum of infected animals at similar rates in both WT and MBL-DKO mice ( Figure 3C ) , suggesting that MBL does not play a major role in serum clearance of RRV or direct neutralization of virus in the serum . The impact of this initial increase in serum viremia in MBL-DKO mice on downstream disease through antibody production is unclear , although it is important to note that both RAG-1−/− and μMT mice develop disease similar to WT mice [8] , indicating that the antibody response is not required for development of disease . Prior studies demonstrated that complement activation drives inflammatory tissue destruction , but does not regulate inflammatory cell recruitment during RRV infection [8] . However , as MBL may regulate the host inflammatory response independently of its effects on complement activation , we quantified and analyzed the inflammatory cell populations within the muscle of WT and MBL-DKO animals at 7 and 10 dpi , which are the times of peak inflammation in WT mice [20] . Consistent with prior findings in C3−/− mice , RRV infected MBL-DKO mice exhibited no statistically significant differences in either total number of leukocytes ( Figure 4A and Figure S4A ) or the composition of the inflammatory infiltrates at either 7 or 10 dpi ( Figure 4B and Figure S4B ) . Representative flow cytometry plots of the various cell types are shown in Figure S5 . The total numbers of CD4+ T cells , CD8+ T cells , and NK cells at both 7 and 10 dpi were not significantly different between RRV-infected WT and MBL-DKO mice ( Figure 4B and Figure S4B ) . Given the role of inflammatory macrophage in development of RRV-induced disease [6] , we compared total numbers of cells with staining characteristics of inflammatory macrophages ( F4/80+ CD11b+ Gr-1lo B220− ) at both 7 and 10 dpi , and observed no difference at 7 dpi , and interestingly , a significant increase in numbers of these cells in RRV-infected MBL-DKO mice at 10 dpi . While we cannot rule out the possibility that minor populations of inflammatory cells are differentially regulated by MBL , these data suggest that MBL does not affect the major populations of inflammatory infiltrates recruited to the skeletal muscle following RRV infection . Although inflammatory cell recruitment was largely unaffected by either MBL or C3 deficiency [8] we have previously shown CR3 is also required for RRV-induced disease and that a subset of inflammatory genes expressed in the inflamed muscle of RRV-infected mice are dependent upon both C3 and CR3 , including the calgranulins S100A8 and S100A9 , IL-6 and the enzyme arginase I ( ArgI ) [9] . Therefore , to determine if MBL affected expression of these genes in the same manner as C3 and CR3 , expression levels were assessed in the quadriceps muscle from both WT and MBL-DKO mice at 7 dpi . As shown in Figure 4C , RRV-infected WT mice exhibited significantly higher expression of S100A8 and S100A9 compared to RRV-infected MBL-DKO mice , indicating that expression of these genes is also regulated by MBL during RRV infection . Interestingly , the S100A8/S100A9 complex has been associated with inflammatory arthritis ( reviewed in [23] ) however , the role of these proteins in RRV disease requires further investigation . Expression of TNFα and IL-1β , which were shown to be C3-independent [9] , were also unaffected in MBL-DKO mice ( Figure 4C ) . Expression of IL-6 and Arg I , which we have previously shown to be C3 and CR3-dependent , were unaffected in MBL-DKO mice ( IL-6 ) or slightly reduced ( Arg I ) ( Figure 4C ) . Expression of these genes may reflect residual complement activation in the absence of MBL ( Figure S3 ) , though this requires further study . Interestingly , expression of IL-10 , which is largely C3-independent following RRV infection [9] , was dependent on MBL and suggests that MBL may be interacting with pathways other than the complement system to mediate IL-10 expression ( Figure 4C ) . Prior studies have shown that the C3 cleavage product C3a is elevated in synovial fluid of RRV polyarthritis patients [8] . To determine if circulating levels of MBL are elevated in RRV-infected patients , we compared serum MBL levels from patients during convalescence to serum MBL levels in RRV-seronegative controls . As shown in Figure 5A , RRV patients had significantly higher levels of circulating MBL compared to healthy controls . Since MBL levels are highly variable in human populations , we also assessed serum and synovial fluid MBL levels in a small cohort of patients clinically characterized as having severe or mild RRV-induced polyarthritis . MBL levels correlated with severity of RRV disease ( Figure 5B ) , with higher levels of MBL observed in patients classified as having severe disease . Severity of disease also correlated with increased levels of C4a in the synovial fluid ( Figure 5C ) , which could result from complement activation through either the lectin or classical pathway . However , analysis of the level of C1q-C4 complexes formed within the synovial fluid , an activation marker of the classical pathway [24] , showed no difference between patients with severe or mild disease ( Figure 5C ) , suggesting that the higher C4a levels in severe RRV disease was primarily due to activation through the lectin pathway . In addition , we did not observe a difference in levels of Bb , an activation marker of the alternative pathway ( Figure 5C ) , further supporting the hypothesis that the MBL pathway primarily mediates complement activation following RRV infection . Importantly , when we assessed MBL levels in a cohort of patients suffering from non-inflammatory osteoarthritis , we found no evidence for elevated MBL levels ( mean MBL levels of 57 . 8±24 . 8 ng/ml [n = 5 patients with severe osteoarthritis] ) compared to MBL levels of 485±163 . 7 ng/ml in patients with severe RRV induced disease and 218 . 5±82 . 4 ng/ml within the synovial fluid of patients with mild RRV-induced disease , suggesting that elevated levels of MBL are not simply the result of arthritis symptoms within the joints . Although additional studies with a larger cohort of patients are required to determine whether MBL levels associate with the severity of RRV-induced arthritis , and whether this effect reflects a causal role for MBL in human disease , these results , combined with the knockout mouse studies strongly suggest that the MBL pathway of complement activation plays a major role in the pathogenesis of RRV-induced inflammatory disease .
Alphaviruses such as CHIKV and RRV represent significant emerging disease threats that cause large-scale outbreaks of severe chronic and persistent arthralgia/myalgia in human populations . Though alphavirus-induced arthralgia and myalgia is often debilitating , the mechanisms by which these viruses cause arthritis/arthralgia are not fully understood . Previous studies have shown that inflammatory macrophages play a major role in the pathogenesis of both RRV and CHIKV [6] , [25] , that the host complement cascade is essential for the induction of muscle destruction by these inflammatory cells during RRV infection , and that this process is dependent on complement receptor 3 ( CR3 ) [8] , [9] . The data presented here demonstrate that RRV infection results in the deposition of MBL within RRV infected tissues , and that MBL , but not the alternative or classical complement activation pathways , was essential for RRV induced complement activation and subsequent inflammatory tissue destruction and disease . Consistent with this , humans suffering from severe RRV disease exhibited increased levels of MBL , but not markers of the classical or alternative complement activation pathways in their synovial fluid . Therefore , these studies demonstrate that MBL plays a key role in promoting the pathogenesis of alphavirus-induced inflammatory disease , and suggest that MBL may represent a target for therapeutic intervention in the treatment of alphavirus-induced arthritis/myositis . MBL has generally been associated with a protective role during viral infection , either through its ability to neutralize viruses directly or via the downstream activation of complement . In the context of arbovirus infection , MBL contributes to direct neutralization of mosquito-derived West Nile virus and both mammalian and mosquito-derived dengue virus through interactions between MBL and viral N-linked glycans [18] , [26] , and MBL contributes to protection from West Nile virus-induced disease in vivo [19] . Though the host complement cascade has been shown to play a protective role during neurotropic alphavirus infection [21] , [27] , [28] these processes are dependent upon either the classical or alternative complement activation pathways [21] . Furthermore , though Fuchs , et al . , found that MBL bound and neutralized WNV virions , they found no evidence for MBL interactions with two alphaviruses , CHIKV and Sindbis virus [18] , which is supported by our inability to demonstrate direct binding or neutralization of RRV virions by MBL . Therefore , our findings demonstrate a novel role for MBL in the pathogenesis of alphavirus-induced arthritis/myositis and indicate that this pathway , which plays a protective role against many viral infections , is actually a major driver of RRV-induced tissue pathology and disease . In addition to its prominent role in the pathogenesis of RRV-induced disease , the complement cascade is linked to a number of host autoimmune inflammatory disorders , including rheumatoid arthritis ( reviewed in [29] ) . However , the role of MBL in these processes is less clear . Though MBL has been shown to contribute to ischemic injury in mouse models of cardiac or intestinal reperfusion injury [12]–[15] , MBL has not been directly linked to inflammatory arthritis . There are conflicting reports associating MBL polymorphisms with rheumatoid arthritis in humans [30]–[34] , however in mouse models of collagen-induced arthritis , which serves as a model of RA , MBL is dispensable for complement activation and arthritis induction [35] , [36] . Therefore , MBL appears to be playing a unique role in the pathogenesis of RRV-induced arthritis/myositis that is not shared with other arthritic syndromes , though further comparisons between these different disease states are needed to clarify this issue . The studies presented here demonstrate that MBL-dependent complement activation promotes RRV-induced disease and raises several questions relating to the mechanism of RRV activation of complement , the role of MBL polymorphisms in determining disease severity , and the therapeutic potential of MBL inhibition to treat RRV-infected patients . The CRD of MBL recognizes terminal carbohydrates , such as mannose and glucose , which can be found on glycosylated proteins in bacteria and viruses . The RRV glycoproteins contain three N-linked glycosylation sites that are glycosylated with a combination of high mannose and complex glycans [37] and may serve as ligands for MBL , leading to complement activation . While we did not observe direct binding of MBL to virions , we did observe an increase in the amount of MBL deposited onto infected tissues and on virally infected cells ( Figure 2 ) , indicating that some aspect of RRV infection induces MBL deposition and complement activation . Alphaviruses bud from the plasma membrane of infected cells and the viral glycoproteins are prominently exposed on the surface of the cell . Therefore , it is possible that MBL is recognizing and binding to viral glycoproteins on infected tissues during viral egress , resulting in complement activation directly onto the tissue rather than binding to free virus . Alternatively , viral infection may lead to the modification of host cell N-linked glycans or other cellular components , thereby promoting MBL deposition and complement activation; however , both of these possibilities require further investigation . Given the central role of MBL in development of severe RRV-induced disease , specific inhibition of the MBL activation pathway of the complement system in RRV-patients may be a strategy to alleviate disease . Current therapy involves administration of non-steroidal anti-inflammatory drugs , and given the role that complement plays in mediating severe RRV-induced disease , treatment with complement inhibitors may provide an attractive alternative to nonspecific anti-inflammatory drugs . However , prolonged inhibition of the complement system can leave patients susceptible to other infectious diseases , especially as treatment of disease symptoms may require several months for some individuals [38] . Our results suggest that a more focused approach targeting MBL may prove effective in limiting RRV-induced arthralgia/myalgia , while limiting the general immune suppression associated with complement inhibition . Inhibitors targeting the MBL pathway of complement through inhibition of MASP-2 are in development [39] , and may be useful in treatment of RRV-induced disease in humans . In addition to raising the possibility of targeting MBL in the treatment of RRV or other alphavirus-induced arthraligias/myalgias , these studies raise the issue of whether polymorphisms in MBL affect susceptibility to RRV-induced disease . Common genetic polymorphisms within the promoter region and exon 1 of the human Mbl2 gene lead to variations in serum MBL levels or functional deficiency of MBL ( reviewed in [40] ) . Human patients with severe RRV disease have higher levels of MBL within the synovial fluid and serum; however , it is unclear if levels of MBL are elevated in response to severe RRV infection or if naturally higher levels of MBL contribute to the development of severe disease . Preliminary analysis of a small cohort of RRV patients does not associate Mbl2 polymorphisms with severity of RRV disease ( S . Mahalingam , B . Piraino , B . Cameron , L . Herrero , and A . Lloyd , unpublished data ) , however a larger cohort of RRV-infected individuals must be analyzed before we can conclude whether MBL polymorphisms associate with RRV-induced disease severity or if up-regulation of MBL in response to viral infection contributes to disease pathogenesis in humans . In summary , the data presented in this study demonstrate the role for MBL in promoting severe disease following RRV infection through complement activation and subsequent destruction of RRV infected tissue . Numerous studies have shown a protective role for MBL and the complement system in response to a diverse set of viruses . Our results demonstrate a novel role of MBL following viral infection in which MBL contributes to development of severe disease , and these findings suggest that MBL may be a therapeutic target for treatment in humans suffering from RRV-induced polyarthritis or other alphavirus-induced arthritides .
Some of the studies described in this manuscript did involve human samples . For human serum samples , all serum samples had been submitted for diagnostic testing with written and oral informed patient consent at CIDMLS , Westmead Hospital and The Royal Melbourne Hospital ( Melbourne , Australia ) . Samples were de-identified by the testing laboratory before being used in the research project . Synovial samples were collected from adult patients ( age range , 30–45 years ) residing in the Murray-Goulburn Valley ( Victoria , Australia ) who had acute cases of RRV-induced polyarthritis in accordance with human subjects protocols approved by the Royal Melbourne Hospital Human Ethics Committee . All individuals received and completed written informed consent forms prior to collection of materials . Mouse studies were performed in strict accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health . All mouse studies were performed at the University of North Carolina ( Animal Welfare Assurance # A3410-01 ) using protocols approved by the UNC Institutional Animal Care and Use Committee ( IACUC ) . All studies were performed in a manner designed to minimize pain and suffering in infected animals , and any animals that exhibited severe disease signs was euthanized immediately in accordance with IACUC approved endpoints . The viral stocks used in this study were generated from the infectious clone of the T48 strain of RRV ( pRR64 ) , kindly provided by Richard Kuhn ( Purdue University ) as described in [20] . Briefly , viral RNA was generated through in vitro transcription of SacI-linearized pRR64 using the mMessage mMachine SP6 kit ( Ambion ) and electroportated into BHK-21 cells ( ATCC ) . Viral titer was determined by plaque assay on BHK-21 cells . BHK-21 cells were grown in α-MEM ( Gibco ) supplemented with 10% donor calf serum ( DCS ) , 10% tryptose phosphate , L-glutamine , penicillin , and streptomycin . C2C12 cells were grown in DMEM ( Gibco ) supplemented with 20% fetal bovine serum ( FBS ) , L-glutamine , penicillin , and streptomycin prior to differentiating . To differentiate cells into myotubes , confluent C2C12 cells were maintained in DMEM supplemented with 2% horse serum , L-glutamine , penicillin , and streptomycin . All mice used in this study were maintained and bred in house at the University of North Carolina ( UNC ) in accordance with UNC Institutional Animal Care and Use Committee guidelines . C57BL/6 and MBL-DKO mice were purchased from The Jackson Laboratories ( Bar Harbor , ME ) ; C1q−/− mice were a generous gift from Dr . Marina Botto ( Imperial College London , UK ) ; fB−/− mice were generously provided by Dr . Charles Jennette ( UNC ) . While RRV is classified as a biosafety level-2 agent , due to the exotic nature of the virus , all animal studies were performed in a biosafety level-3 facility . Twenty-four day old mice were inoculated with 103 PFU of RRV in diluent ( phosphate buffered saline supplemented with 1% DCS ) into the left rear footpad . Mice were weighed daily and assigned a clinical score based on hind limb weakness and altered gait on the following scale: 0 = no disease; 1 = mild loss of hind limb grip; 2 = moderate loss of hind limb grip; 3 = severe loss of hind limb grip; 4 = no hind limb grip and mild inability to right; 5 = no hind limb grip and complete inability to right; 6 = moribund . Mice were infected with RRV as described above , and at indicated times post infection mice were sacrificed , perfused with 1× PBS , and indicated tissues were dissected out , weighed , and homogenized with glass beads in diluent . Viral titer within infected tissues were determined by plaque assay on BHK-21 cells from tissue homogenates . Mice were infected with RRV as described above , and at indicated times post infection mice were sacrificed , perfused with 4% paraformaldehyde ( PFA ) , pH 7 . 3 . Tissues were paraffin embedded and 5 µm sections were generated and in situ hybridization was performed as previously described [20] using an RRV-specific or EBER-specific 35S-labeled RNA probe . At desired times post infection , mice were sacrificed and perfused with 4% paraformaldehyde ( PFA ) , pH 7 . 3 . Tissues were paraffin embedded and 5 µm sections were generated and stained with hematoxylin and eosin ( H&E ) to examine tissue pathology and inflammation . Sections were visualized by bright field light microscopy ( Olympus BX61 ) . At 10 days post infection mice were injected with 1% Evans blue dye in PBS into the peritoneal cavity ( 50 µl/10 g mouse weight ) . At 6 hours post injection , mice were sacrificed and perfused with 4% PFA . Quadriceps muscle tissues were embedded in optimal cutting temperature compound ( OCT ) and frozen in an isopentane histobath , 5 µm sections were generated , mounted with ProLong Gold with DAPI ( Invitrogen ) and sections were analyzed by fluorescence microscopy ( Olympus BX61 ) . At 7 dpi , mice were sacrificed and perfused with 4% PFA . Quadriceps muscles were removed , paraffin embedded , and 5 µm sections were generated . Sections were deparaffinized in xylene , rehydrated through an ethanol gradient , and probed with a goat anti-mouse C3 polyclonal antibody ( 1∶500 Cappel ) using the Vectastain ABC-AP kit ( Vector Labs , CA ) and Vector Blue Alkaline phosphatase substrate kit ( Vector Labs , CA ) according to the manufacturers' instructions . Sections were counterstained with Gill's hematoxylin . At indicated times post infection , mock-infected and RRV-infected mice were sacrificed , and perfused with 1× PBS . Quadriceps muscles were removed and homogenized in radioimmunoprecipitation lysis buffer ( RIPA; 50 µM Tris pH 8 . 0 , 150 mM NaCl; 1% NP-40 , 0 . 5% deoxycholate , 0 . 1% SDS and 1× complete protease inhibitor cocktail ( Roche ) ) by glass beads . Protein concentration was determined by Bradford protein assay and 25–30 µg of protein was run onto a 10% SDS-PAGE gel . Protein was transferred onto a PVDF membrane , and membranes were blocked in 5% milk , 0 . 1% Tween-20 in PBS . Membranes were probed with goat anti-mouse MBL-A ( 1∶1000 R&D Systems ) , goat anti-mouse MBL-C ( 1∶1000 R&D Systems ) , goat anti-mouse C3 polyclonal antibody ( 1∶1000 Cappel ) , mouse anti-RRV ( 1∶1000 ATCC ) , or goat anti-mouse actin polyclonal antibody ( 1∶500 SCBT ) , washed with PBS containing 0 . 1% Tween-20 and incubated with rabbit anti-goat antibody or sheep anti-mouse antibody conjugated to horseradish peroxidase ( 1∶ 10 , 000 Sigma ) . Membranes were washed again and protein visualized by ECL ( Amersham ) according to manufacturer's instructions . Densitometry was performed using ImageJ software ( NIH ) . Differentiated C2C12 cells were either mock-infected or infected at an approximate MOI of 20 with RRV . At 24 hpi , culture medium was removed and cells were incubated in differentiation medium containing either 10% serum from WT or MBL-DKO mice for an additional 30 minutes . Cells were washed and harvested in RIPA lysis buffer , and cell lysates were analyzed by immunoblot analysis . To determine the composition of the inflammatory cell infiltrates within the quadriceps muscle , at indicated times post infection mice were sacrificed and perfused with 1× PBS . Both quadriceps muscles were removed , minced , and digested with RPMI containing 10% fetal bovine serum ( FBS ) , 15 mM HEPES , 2 . 5 mg/ml collagenase A ( Roche ) , 1 . 7 mg DNase I ( Roche ) for 2 hours at 37°C with shaking . Cells were strained through a 40 µm strainer and washed twice with wash buffer ( HBSS containing 1% sodium azide and 1% FBS ) and total viable cells were determined by trypan blue exclusion . To stain cells for flow cytometry , cells were incubated with anti mouse FcgRII/III ( 2 . 4G2; BD Pharmingen ) and stained with combinations of the following antibodies: fluorescein isothiocyanate ( FITC ) -conjugated anti-mouse CD3 , phycoerythrin ( PE ) -conjugated anti-NK1 . 1 , PE-Cy5 anti-CD45 ( leukocyte common antigen ) , PE-Cy7 anti-F4/80 , Allophycocyanin ( APC ) -conjugated anti-CD49b , eF450-conjugated anti-CD11b , APC anti-major histocompatibility complex class II antigens ( MHC II ) , and eF780-conjugated anti-CD45 ( B220 ) ( eBiosciences , San Diego , CA ) , FITC anti-Ly-6G , and PE anti-SigLecF ( BD-Pharmingen , San Diego , CA ) , and PE-Texas Red-conjugated anti-CD45 ( B220 ) , and PE-Texas Red anti-CD11c ( Molecular Bioprobes , Invitrogen ) . Cells were fixed with 2% PFA ( pH 7 . 3 ) and analyzed on a CyAn flow cytometer ( Becton Dickinson ) , and data was analyzed using Summit software . At 7 dpi following RRV infection , mice were sacrificed and perfused with 1× PBS . Quadriceps muscles were removed and homogenized in Trizol ( Invitrogen ) using glass beads . RNA was extracted using Invitrogen PureLink RNA purification kit , and mRNA expression of indicated genes was measured by quantitative real-time PCR . Raw data values were normalized to 18S rRNA levels . Convalescent serum samples from five patients presenting with acute , serologically confirmed ( seroconversion by neutralization , IgM and IgG ) RRV-infection and thirteen samples from healthy individuals were provided by CIDMLS , Westmead Hospital ( Sydney , Australia ) . All serum samples had been submitted for diagnostic testing with informed patient consent at CIDMLS , Westmead Hospital and The Royal Melbourne Hospital ( Melbourne , Australia ) . Samples were de-identified by the testing laboratory before being used in the research project . Needle biopsy was performed to collect synovial fluid samples from adult patients ( age range , 30–45 years ) residing in the Murray-Goulburn Valley ( Victoria , Australia ) who had acute cases of RRV-induced polyarthritis . Samples were collected and prepared aseptically in the laboratories of Echuca Hospital ( Murray-Goulburn Valley; Victoria , Australia ) and The Royal Melbourne Hospital and was performed in accordance with The Royal Melbourne Hospital Human Ethics Committee . Severe RRV-induced disease was defined as a patient presenting with intense swelling , severe joint pain and myalgia affecting both the knee joints and joints of the fingers . Mild RRV-induced disease was defined as a patient presenting with minor swelling , localized in the knees , and no additional symptoms . For osteoarthritis samples , synovial fluid aspirates were obtained from 5 patients with osteoarthritis from the John James Hospital ( Canberra , Australia ) . Sample collection was performed in accordance with the AustralianCapital Territory Health Community Care Human Research Ethics committee . Samples were obtained at the time that knee joint arthroplasty was performed , and joints were aspirated before arthrotomy . The diagnosis given to patients was primary osteoarthritis with no evidence of an inflammatory arthropathy . These samples were de-identified prior to analysis . Levels of MBL in serum and synovial fluid were determined using a commercially available ELISA kit according to the manufacturer's instructions ( R&D Systems ) . Levels of C4a in the synovial fluid was determined using BD OptEIA ( BD ) . Bb levels were determined using Microvue Bb Plus ( Quidel ) . The levels of the C1q-C4 complex were determined as described in [24] . Clinical scores and percent of starting weight at 10 dpi between C1q−/− , fB−/− , MBL-DKO , and wild-type mice were analyzed for statistically significant differences by Mann-Whitney analysis with multiple comparisons corrections ( clinical scores; p<0 . 01 is considered significant ) , and by one-way ANOVA with Bonferroni's correction ( percent of starting weight; p<0 . 05 is considered significant ) . Viral burden , total number of infiltrating cells , and gene expression data at each time point between wild-type and MBL-DKO mice was analyzed for statistically significant differences by Mann-Whitney analysis or t-test ( p<0 . 05 is considered significant ) . Levels of MBL , C4a , C1q-C4 complexes , and Bb in serum and synovial fluid from human patients were analyzed by Mann-Whitney analysis for statistical significance ( p<0 . 05 is considered significant ) . Statistical analyses were performed using GraphPad Prism 5 . | Arthritogenic alphaviruses such as Ross River virus ( RRV ) and chikungunya virus are transmitted to humans by mosquitoes and cause epidemics of debilitating infectious arthritis and myositis in various areas around the world . Studies in humans and mice indicate that the host inflammatory response is critical for development of RRV-induced arthritis and myositis , and the host complement system , a component of the host inflammatory response , plays an essential role in the development of RRV-induced disease through activation of complement receptor 3 ( CR3 ) -bearing inflammatory cells . Of the three main complement activation pathways , only the lectin pathway activated by mannose binding lectin ( MBL ) was essential for RRV-induced complement activation , tissue destruction , and disease . Furthermore , we found that levels of MBL were elevated in human patients suffering from RRV-induced polyarthritis and MBL levels correlated with disease severity . Taken together , our data implicates a role for MBL in mediating RRV-induced disease in both humans and mice , and suggests that therapeutic targeting of MBL may be an effective strategy for disease treatment in humans . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"virology",
"biology",
"microbiology"
] | 2012 | Mannose Binding Lectin Is Required for Alphavirus-Induced Arthritis/Myositis |
We use the qualitative insight of a planar neuronal phase portrait to detect an excitability switch in arbitrary conductance-based models from a simple mathematical condition . The condition expresses a balance between ion channels that provide a negative feedback at resting potential ( restorative channels ) and those that provide a positive feedback at resting potential ( regenerative channels ) . Geometrically , the condition imposes a transcritical bifurcation that rules the switch of excitability through the variation of a single physiological parameter . Our analysis of six different published conductance based models always finds the transcritical bifurcation and the associated switch in excitability , which suggests that the mathematical predictions have a physiological relevance and that a same regulatory mechanism is potentially involved in the excitability and signaling of many neurons .
Detailed computational conductance-based models have long demonstrated their ability to faithfully reproduce the variety of electrophysiological signatures that can be recorded from a single neuron in varying physiological or pharmacological conditions . But the predictive value of a computational model is limited unless its analysis sheds light on the core mechanisms at play behind a computer simulation . Because conductance-based models are nonlinear dynamical models , their analysis often requires a drastic reduction of dimension . The reduced model is amenable to the geometric methods of dynamical systems theory , but the mathematical insight is often gained at the expense of physiological interpretability; hence the need for methodological tools that can relate mathematical predictions of low-dimensional models to physiological predictions in detailed conductance based models . In recent work [1] , we used phase plane analysis and dynamical bifurcation theory to characterize in reduced-order neurodynamics models a switch of excitability that is consistent with many physiological observations . More precisely , a transcritical bifurcation governed by a single parameter was shown to organize a switch from restorative excitability , extensively studied in most models inspired from the Hodgkin-Huxley model , to regenerative excitability whose distinct electrophysiological signature include spike latency , plateau oscillations , and afterdepolarizeation potentials . The main contribution of the present paper is to show that this transcritical bifurcation , and the associated excitability switch , exist in a number of high-dimensional conductance-based models and that the resulting mathematical predictions have physiological relevance . Although purely mathematical in nature , the detection of the transcritical bifurcation relies on an ansatz that leads to a simple physiological interpretation: the switch of excitability is determined by a balance between restorative ( those providing a negative feedback ) and regenerative ( those providing a positive feedback ) ion channels at the resting potential . Because this simple balance equation can take many different physiological forms , it is potentially shared by very different neurons . We use the balance equation to provide an algorithm to trace the transcritical bifurcation in arbitrary conductance-based models . We apply the algorithm to detailed conductance-based models of six neurons known to exhibit drastic changes in their electrophysiological signatures depending on environmental conditions: the squid giant axon [2] , the dopaminergic neuron [3] , the thalamic relay neuron [4] , the thalamic reticular neuron [5] , the aplysia R15 model [6] , and the cerebellar granule cell [7] . In each case , the algorithm identifies a transcritical bifurcation that occurs close to the nominal model parameters and its predictions are consistent with experimental observations . After defining a novel classification of ion channels based on their restorative or regenerative nature , we briefly review the planar model presented in [1] and how its transcritical bifurcation qualitatively captures the switch between restorative and regenerative excitability . As a generalization of this low-dimensional case , we mathematically construct the same bifurcation in generic conductance based models and derive the balance condition determining the regenerative or restorative nature of the model . This construction and its electrophysiological predictions are firstly illustrated on the squid giant axon . An algorithm for generic conductance-based models is subsequently derived and different models analysed .
Conductance-based models of neurons describe the dynamic interaction between the membrane potential and - possibly many - gating variables that control the ionic flow through the membrane . The gating of ion channels occurs on many different timescales . However , gating timescales can be grouped in three families , according to their influence on neuronal excitablity [8]: In view of their importance for neuronal excitability , we focus only on slow gating variables to classify ion channels: when the slow channel provides a negative feedback on membrane potential variations , we term the associated channel a slow restorative ion channel . When the slow variable instead enhances a potential variation by positive feedback , the associated ion channel is termed slow regenerative ( a characterization in terms of partial derivatives is postponed to the next sections ) . Ion channels that do not possess a slow gating variable are neither restorative nor regenerative and are called neutral . Neutral ion channels solely regulate the “quantity” of excitability without affecting its “quality” . Table 1 shows a classification of many known ion channels according to this criterion . Not surprisingly , potassium channels are the main representatives of slow restorative ion channels . By increasing the total outward current , their activation induces a negative feedback on membrane potentials variations that is responsible for neuron repolarization . On the other hand , physiologically described calcium channels are all slow regenerative . Their activation induces an increase of the total post-spike inward current , in contrast to potassium channels . This is the source , for instance , of afterdepolarization potentials ( ADP ) . Interestingly , sodium channels can be either restorative , regenerative , or neutral according to their fast transient , resurgent , or persistent behavior , respectively . It is important to observe that the restorative ( resp . regenerative ) nature of channels is not solely linked to the outward ( resp . inward ) nature of the current . For instance , transient sodium channels ( although responsible for the regenerative spike upstroke ) are slow restorative , because their slow variable inactivates an inward current , inducing a negative-feedback on membrane potential variations . Similarly , potassium channels can be slow regenerative when their slow inactivation massively decreases the outward current , like in the case of A-type potassium channels . Although elementary , the classification above seems novel . It is motivated by the central message of this paper , that the balance between regenerative and restorative ion channels in slow timescale determines its neuronal excitability type . In the remainder of the paper , we simply write restorative ( resp . regenerative ) for slow restorative ( resp . slow regenerative ) channels . Planar models - that only consist of two state variables - have been instrumental to study excitability since the early days of neurodynamics [9] , [10] . Empirical planar reductions of conductance based models only retain the ( fast ) voltage variable and one slow gating variable . Fast gating variables are set to steady-state ( i . e . their fast variation is approximated as instantaneous ) , adaptation variables are treated as slowly varying parameters , and the sole gating variable aggregates all slow variables , expressing each of them as a ( curve-fitted ) static function of . Motivated by the phase portrait of such an empirical reduction of the Hodgkin-Huxley model augmented with a calcium channel [11] , our recent study [1] explores the neuronal excitability of the planar model ( 1a ) ( 1b ) whose phase portraits are reproduced in Fig . 1 for two distinct values of the parameter ( an indirect representation of the calcium conductance in the high-dimensional model ) . The parameter characterizes the time-scale separation between and . The function has the standard sigmoid shape of conductance-based models and is the half-activation potential . The typical step responses of ( 1 ) are also reproduced in Fig . 1 . The spike generation mechanism in the phase portrait of Fig . 1 left is reminiscent of the of FitzHugh-Nagumo model and of the physiologically grounded planar reduction of Hodgkin-Huxley model by Rinzel [10] . It is associated to a reversible and sudden switch from resting to spiking and has been studied extensively , with finer distinctions depending on the mathematical nature of the underlying bifurcation [12] . For further reading , see [13] , [14 , Section 7 . 1 . 3] , and references therein . The phase portrait in Fig . 1 right is in sharp contrast in that the electrophysiological response to a current input exhibits spike latency , plateau oscillations , and after depolarization potential ( ADP ) . This specific signature , experimentally observed in many families of neurons , is fundamentally associated to the bistability illustrated in the phase portrait: namely , the robust coexistence of two stable attractors ( a hyperpolarized resting potential and a limit cycle of periodic action potentials ) and a saddle-separatrix that sharply separates their basins of attraction . The time evolution shown in the top figure is a consequence of this phase portrait and cannot be observed in FitzHugh-Nagumo like phase portraits . The distinction between the two phase portraits , the associated excitability types , and their relation with Hodgkin's excitability classification [15] are further discussed in [1] and later in the paper . A simple mathematical distinction between the two phase portraits shown in Fig . 1 is drawn from the Jacobian linearization of the model at the stable resting point :The product of the partial derivatives is negative on the left phase portrait ( ) , capturing the restorative nature of the gating variable , whereas it is positive on the right phase portrait ( ) , capturing the regenerative nature of the gating variable . This difference is schematized in the block diagrams of Fig . 2 . Accordingly , excitability in planar models is called restorative ( resp . regenerative ) when the gating variable provides negative ( resp . positive ) feedback close to the resting point: We start by grouping gating variables of a given conductance-based model according to their time scales . The family collects fast gating variables . The gating variable denotes an arbitrary member of this family . Similarly , the family collects slow gating variables , whereas collects adaptation variables . For a given ion channel type , the standard notation ( resp . ) is adopted for the activation ( resp . inactivation ) gating variable of the associated ionic current . With these notations , a general neuron conductance-based model reads ( 5a ) ( 5b ) ( 5c ) ( 5d ) where the sum in ( 5a ) is over all ion channels in the model , and ( 5b ) , ( 5c ) , ( 5d ) hold for all the associated fast , slow , and adaptation variables , respectively . The activation ( resp . inactivation ) functions are strictly monotone increasing ( resp . decreasing ) sigmoids . In the forthcoming analysis , all adaptation variables are treated as constant parameters , that is , their slow evolution is neglected . We will detect a switch from restorative to regenerative excitability by mimicking the two-dimensional algorithm of the previous section . We first impose the bifurcation condition , where denotes the Jacobian of the subsystem ( 5a ) , ( 5b ) , ( 5c ) . The algebraic condition writes ( 6 ) where the sums are over all fast and slow variables , respectively . The particular form of equation ( 6 ) is a direct consequence of the specific structure of conductance-based models , that is , parallel interconnection of two-dimensional feedback loops involving the voltage dynamics ( 6a ) and one of the gating variable dynamics ( 6b ) , ( 6c ) . As for the planar model ( 1 ) , we track the switch between restorative and regenerative excitability by imposing the high-dimensional equivalent of the balance condition ( 2 ) . We therefore look for solutions of ( 6 ) satisfying the ansatz ( 7 ) The two conditions ( 6 ) and ( 7 ) now imply ( 8 ) where denotes the Jacobian of the fast subsystem ( 5a ) , ( 5b ) . We show in Supplementary Material S1 that the corresponding bifurcation is necessarily transcritical [17 , Section 3 . 2] . The singularity condition ( 8 ) is the high-dimensional counterpart of the -nullcline self-intersection in the planar model . It reflects the geometric nature of the transcritical bifurcation , that is , a robust geometrical object that exists independently of the timescale separation and persists in the singular limit of an infinite timescale separation , regardless of the system dimension . Our ansatz makes the proposed analysis robust against the model time constants . The time constants are only used to classify the gating variables in the three physiological groups . We split in two subfamilies: , which contains regenerative slow gating variables , and , which contains restorative slow gating variables . The balance condition ( 7 ) is then rewritten as ( 9 ) to express a balance between restorative and regenerative ion channels . It is the high-dimensional counterpart of ( 2 ) and it provides a rigorous high-dimensional generalization of restorative and regenerative excitability: The Hodgkin-Huxley ( HH ) model [2] provides a non-physiological , but historical and experimentally verified tutorial for tracking a switch of excitability in conductance based models . The model reads ( 10a ) ( 10b ) ( 10c ) ( 10d ) where is the fast sodium channel activation while the sodium channel inactivation and the potassium channel activation are the slow gating variables . We set all time constants to one , because this simplification has no effects on the algebraic conditions ( 7 ) and ( 8 ) . The Jacobian of ( 10 ) reads ( 11 ) The upper-left block is the Jacobian of the fast subsystem . Imposing the singularity condition ( 8 ) yields ( 12 ) while the balance equation ( 9 ) reads ( 13 ) Note that ( 12 ) and ( 13 ) imply the bifurcation condition in ( 11 ) . At first sight , the balance condition ( 13 ) cannot be satisfied because both sodium and potassium channels are restorative channels according to their corresponding kinetics in the model , and in agreement with our proposed classification . This is consistent with the fact that the excitability of the HH model is always restorative in physiological conditions . However , it was long recognized [18] that potassium channels can generate an inward current at steady-state if the extracellular concentration is sufficiently large . Indeed , any change in extracellular potassium concentration induces a change in the potassium reversal potential , as expressed by the Nernst equation . This suggests to use the potassium reversal potential as a bifurcation parameter in HH model in order to satisfy the balance equation ( 14 ) where potassium now acts as a regenerative gating variable provided that . Physiologically , condition ( 14 ) imposes that the potassium Nernst potential is large enough for the regenerativity of the potassium activation to balance the restorative effects of the sodium current inactivation . The two conditions ( 12 ) and ( 14 ) can numerically be solved to determine the critical values and . The value of the applied current at the transcritical bifurcation is then determined from ( 10a ) , which givesThe numerical bifurcation diagram in Fig . 6A confirms the transcritical bifurcation at . That bifurcation diagram is drawn by varying together with applied current , following the affine reparametrization described in Supplementary Material S1 . More precisely , Mathematically , this reparametrization imposes one of the defining conditions of the transcritical bifurcation . Physiologically , its effect is to keep the net current constant at steady-state ( i . e . ) : as is varied , the observed switch in the excitability type does not rely on changes in the net current across the membrane , but solely on changes in its dynamical properties . The bifurcation diagram in Fig . 6A provides informations on the model excitability also far from the transcritical values . For highly hyperpolarized , the model is purely restorative and exhibits the typical excitable behavior of the original Hodgkin-Huxley model [1 , Figure 8] . As is increased , a stable regenerative steady-state is born in a saddle-node bifurcation . At this transition , the system switches to a mixed excitability type . Short current pulses let the system switch between the depolarized restorative stable steady state and the hyperpolarized regenerative stable steady state ( Fig . 6B , middle ) . The associated bifurcation diagram and phase portrait are reproduced in Fig . 6C , D , middle . Finally , a further increase of lets the restorative steady state exchange its stability with a ( regenerative ) saddle at the transcritical bifurcation ( and , soon after , lose stability in a Hopf bifurcation ) and the system switches to regenerative excitability . The regenerative steady state coexists in this case with the spiking limit cycle attractor . Current pulses switch the system asymptotic convergence between the two attractors ( Fig . 6B , right ) . The associated bifurcation diagram and phase portrait are reproduced in Fig . 6C , D , right . The same qualitative excitability switch was described by Rinzel in [10] , who linked the appearance of a bistable behavior to the inward nature of potassium current at steady-state for sufficiently depolarized . In vitro recordings of the squid giant axon with isotonic extracellular concentration show the same transition [18] . The mathematical analysis of the previous section follows an algorithm that allows to detect a transcritical bifurcation in generic conductance based models of arbitrary dimension and to track associated excitability switches . The steps of the algorithm are summarized in Table 2 . For simplicity and conciseness , we restrict our attention to the modulation of only one regenerative ionic current . However , a similar algorithm can be written for an arbitrary modulation of ionic currents ( by variation of maximal conductance ( s ) , adaptation variable ( s ) , or reverse potential ( s ) ) that brings the model to the balance expressed in ( 9 ) . For the sake of illustration , we apply this algorithm to a number of published conductance-based models and show that all these models can switch between restorative and regenerative excitability through a transcritical bifurcation , as sketched in Fig . 7 . Figure 7 indicates two qualitatively distinct paths from restorative to regenerative excitability: one path traversing the mixed excitability region just described with Hodgkin-Huxley model ( Fig . 6A ) and one path switching directly from restorative to regenerative excitability through the TC bifurcation that will be illustrated on the dopaminergic neuron model . As illustrated in Figure 4 , the significance of the transcritical bifurcation is that it delineates in the parameter space the boundary of a specific type of excitability and that this boundary is determined by a simple physiological balance ( Eq . 9 ) between restorative and regenerative channels . Specific to regenerative excitability is the bistable phase portrait of Fig . 1 , right . For the six analyzed conductance-based models , our bifurcation analysis of the full model confirms the existence of a bistable range beyond the transcritical bifurcation , where a regenerative resting state and a spiking limit cycle coexist . In each case , the bistability range is obtained for the nominal time scales of the published model and is robust to a variation of time scales . In each case , the bistabilty range is also neuromodulated , that is , determined by conductance parameters that are known to vary in slower time scales and/or across neurons of a same type . It is important to distinguish this robust and physiologically regulated bistability from other types of bistability that can be encountered in conductance-based models . Figure 11 qualitatively illustrates three typical bistable phase portraits associated to the planar model ( 1 ) that exhibit the coexistence of a stable resting state and of a spiking limit cycle . The first two are associated to restorative excitability and are extensively studied in the literature . See , e . g . [13] , [14] , and references therein . Only the third one is associated to regenerative excitability . The three bistable phase portraits share the common feature of “hard excitation”: as the amplitude of a step input depolarizing current is increased , the response of the neuron abruptly switches from no oscillation to high frequency spiking . Following the historical classification of Hodgkin [15] , the three situations correspond to Class II neurons , as opposed to Class I neurons for which the spiking frequency gradually increases with the depolarizing current amplitude . Hard excitation can be a manifestation either of a switch-like monostable bifurcation diagram or of a hysteretic bistable bifurcation diagram . By definition , the three bistable phase portraits in Figure 11 give rise to hysteretic bifurcation diagrams . But for the two bistable phase portraits associated to restorative excitability , the hysteresis is highly dependent on the time scale separation , i . e . , the ratio between the fast and slow time constants . In the case of the first phase portrait ( subcritical Hopf bifurcation ) , asymptotic analysis shows that the hysteresis vanishes as . In the case of second phase portrait ( saddle-homoclinic bifurcation ) , the situation is even worse because for small the system necessarily undergoes a monostable saddle-node on invariant circle bifurcation . In fact , the second phase portrait is not physiological for neuron conductance based models . For instance , in the hypothetical conductance-based model considered in [14 , Fig . 6 . 44] , the time constant of the potassium activation must be set below to create a saddle-homoclinic bifurcation , which is roughly 40 times smaller than its physiological value and even smaller than the fast time constant . A geometric proof of the generality of this fact is provided in [1] . The conclusion is that hysteresis associated to restorative excitability is at best very small ( if any ) in physiologically plausible conductance based models , which makes their electrophysiological signatures similar to those associated to a switch-like monostable bifurcation diagram . In sharp contrast , the hysteresis associated to regenerative excitability is barely affected by the time-scale separation . Instead it is regulated by conductance parameters whose modulation is physiological ( for instance , a regenerative ion channel density ) . The extended hysteresis is what determines the specific electrophysiological signature of regenerative excitability: a pronounced spike latency , a possible plateau oscillation , and an after depolarization potential . As a consequence , those features cannot be robustly reproduced in physiologically plausible conductance based models of restorative excitability . Because those features are important markers of modern electrophysiology [22] , [23] , the distinction between restorative and regenerative excitability seems physiologically relevant , beyond the possible shared feature of hard excitation . In conclusion , the bistability associated to regenerative excitability is specific in that it produces a robust electrophysiological signature in physiologically plausible parameter ranges and consistent with many experimental observations . It is in that sense that the balance equation delineates a switch of excitability of physiological relevance . Early in the history of neurodynamics [15] , Hodgkin proposed a classification of excitability in three different classes: Class I: The spiking frequency vs . input current amplitude ( f/I ) curve is continuous , i . e . , the spiking frequency continuously increases from zero to high-frequency firing as the input current amplitude rises . Class I excitability is also referred to as “soft” excitation . Class II: The f/I is discontinuous , i . e . , the spiking frequency abruptly switches from zero to high-frequency firing as the amplitude of the applied current is raised above a certain threshold . Class II excitability is also referred to as “hard” excitation . Class III: The spiking frequency is zero for all amplitudes of the applied current . Transient action potentials can be generated in response to high-frequency stimuli . Because regenerative excitability exhibits hard excitation , it is a physiologically distinct subtype of Class II excitability . Bifurcation theory helps relating this physiological classification to mathematical signatures of the associated neuron models . Distinct bifurcations delineate the different excitability classes as well as the different excitability mechanisms within a given class . They are summarized in Fig . 12 . Ermentrout [12] showed that Class I excitability arises from a saddle-node on invariant circle bifurcation , whereas Class II excitability arises from a Hopf bifurcation . Both bifurcations correspond to examples of restorative excitability in the terminology of the present paper and the transition between Class I and II is governed by a Bogdanov-Takens bifurcation . Our recent paper [1] further expands this classification to account for regenerative excitability . Regenerative excitability ( called Type IV in [1] ) arises from a ( singularly perturbed ) saddle-homoclinic bifurcation and the transition from restorative to regenerative excitability always involves a transcritical bifurcation .
Motivated by a geometric analysis of a qualitative phase portrait , we have proposed an algorithm that easily detects a transcritical bifurcation in arbitrary conductance based models . Owing to the special structure of such models , the algorithm leads to solving an algebraic equation of remarkable simplicity and physiological relevance: a balance between slow restorative and slow regenerative ion channels . The condition is also robust because the balance is independent of the detailed kinetics , even though it critically relies on a classification of variables in three well separated time-scales , in good agreement with what is known on ion channels kinetics [8] . The detection of the transcritical bifurcation relies on the sole existence of a physiological balance between restorative and regenerative ion channels . Given that all neuronal models possess restorative sodium and potassium channels , this implies that a transcritical bifurcation exists in every conductance-based model that possesses at least one regenerative ion channel . Moreover , the channel balance , and therefore the TC bifurcation , are readily detectable in a model of arbitrary dimension ( both in the state and parameters ) : the balance ( 9 ) simply defines a hypersurface in the parameter space that can algebraically be tracked under arbitrary parameter variations . An illustration was provided on the GC model above . In spite of its ubiquity and of its physiological significance , we are not aware of an earlier reference to a transcritical bifurcation in conductance based models . A reason for this omission might be accidental: there are no regenerative channels in the seminal model of Hodgkin and Huxley ( unless one modifies the potassium resting potential ) and this model has been the inspiration of most mathematical analyses of conductance-based models . For the same reason , it seems physiologically relevant to distinguish between restorative and regenerative excitability beyond Hodgkin's classification of Class I ( “soft” ) and Class II ( “hard” ) excitability . Regenerative ( and restorative ) excitability faithfully capture the presence ( or the absence ) of specific electrophysiological signatures of modern electrophysiology such as spike latency , afterdepolarization potentials , or robust coexistence of resting state and repetitive spikes . Although purely mathematical in nature , the transcritical bifurcation has a remarkable predictive value in several published conductance based models . In each of the six analysed models , the proposed algorithm identifies a physiological parameter that acts as a tuner of neuronal excitability in a physiologically plausible range and in full agreement with existing experimental data . At the same time , the distinct nature of the regulating parameter , which can be either the maximal conductance or the inactivation gating variable of a regenerative ion channel depending on the neuron model , is associated to distinctly different regulation mechanisms . The classification of gating variables in three distinct time scales is an essential modeling step both for the proposed algorithm and for the reduction of full conductance-based models to low-dimensional models that can be used in population studies [24] . Despite the inherent robustness of time-scale separation analysis , this classification is a limitation of the proposed approach if the model contains ion channels with poorly known kinetics . When all slow ion channels are properly identified , they can be aggregated in a single slow variable to lead to a second order model of the type ( 1 ) , where the single parameter captures the restorative or regenerative nature of the aggregated slow variable . Further reduction to a one-dimensional hybrid model with reset is possible thanks to the time-scale separation between the voltage and the slow variable . This reduction is illustrated in [11] on the thalamic TC neuron and the reduced model remarkably retains the switch of excitability of its high-dimensional counterpart . In contrast , a reduced model will lose the switch of excitability of the full conductance-based model when a regenerative ion channel is treated as a fast gating variable . It is for instance common in model reduction to set the activation of a calcium channel to steady state . This amounts to treat the calcium activation as a fast variable , which makes the channel either “slow restorative” or “neutral” in the terminology of this paper . If the calcium channel is the only source of regenerative excitability , then the reduced model will not retain features of regenerative excitability . In each of the analysed conductance-based models , the balance equation responsible for the switch of excitability is satisfied for a set of parameters that is close to the published parameter values . This observation supports the hypothesis that neuronal excitability is tightly regulated by molecular mechanisms and that the influence of the channel balance condition on neuronal excitability might play a role in neuronal signaling .
Numerical temporal traces of the different neurons ( Figs . 1 , 6 , 8 , 9 ) were reproduced by implementing in MATLAB ( available at http://www . mathworks . com ) the original models as described in the associated papers . The phase portraits in Figures 1 and 3 were hand-drawn using the Open Source vector graphics editor Inkscape ( http://inkscape . org ) . The phase portraits in Figure 6 were numerically drawn with MATLAB by implementing the planar model in [10] and subsequently modified with Inkscape . The bifurcation diagrams in Figures 6 , 7 , and 8 were drawn by implementing the algorithm of Table 2 in MATLAB . No figure or part of figure was reproduced from other published works . | Understanding the changing electrophysiological signatures of neurons in different physiological and pharmacological conditions is a central focus of experimental electrophysiology because a key component of cell signaling in the nervous system . Computational modeling may assist experimentalists in this quest by identifying core mechanisms and suggesting pharmacological targets from a mathematical analysis of the model . But a successful interplay between experiments and mathematical predictions requires new analysis tools adapted to the complexity of high-dimensional computational models nowadays available . We use bifurcation theory to propose a mathematical condition that can detect an important switch of neuronal excitability in arbitrary conductance-based neuronal models and we illustrate its physiological relevance in six published state-of-the art models of different neurons . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"physics",
"systems",
"biology",
"mathematics",
"computational",
"neuroscience",
"biology",
"computational",
"biology",
"nonlinear",
"dynamics",
"biophysics",
"neuroscience"
] | 2013 | A Balance Equation Determines a Switch in Neuronal Excitability |
The H1N1 influenza pandemic of 2009 has claimed over 18 , 000 lives . During this pandemic , development of drug resistance further complicated efforts to control and treat the widespread illness . This research utilizes traditional Chinese medicine Database@Taiwan ( TCM Database@Taiwan ) to screen for compounds that simultaneously target H1 and N1 to overcome current difficulties with virus mutations . The top three candidates were de novo derivatives of xylopine and rosmaricine . Bioactivity of the de novo derivatives against N1 were validated by multiple machine learning prediction models . Ability of the de novo compounds to maintain CoMFA/CoMSIA contour and form key interactions implied bioactivity within H1 as well . Addition of a pyridinium fragment was critical to form stable interactions in H1 and N1 as supported by molecular dynamics ( MD ) simulation . Results from MD , hydrophobic interactions , and torsion angles are consistent and support the findings of docking . Multiple anchors and lack of binding to residues prone to mutation suggest that the TCM de novo derivatives may be resistant to drug resistance and are advantageous over conventional H1N1 treatments such as oseltamivir . These results suggest that the TCM de novo derivatives may be suitable candidates of dual-targeting drugs for influenza .
The first global pandemic of the 21st century was announced by the World Health Organization ( WHO ) in 2009 due to the worldwide spread of influenza A subtype H1N1 ( H1N1/09 ) [1] . More than 214 countries have reported laboratory confirmed cases , and more than 18 , 449 deaths have been recorded [2] . Currently , the neuraminidase inhibitor Tamiflu® ( oseltamivir ) remains the primary drug prescribed to patients infected with H1N1/09 [3] . However , the emergence of drug resistant viral strains [4] and limited drug administration window [5] exemplifies the need for additional therapies . Important constituents of influenza surface membrane proteins include hemagglutinin , neuraminidase , and the matrix protein 2 ( M2 ) proton channel [6] , [7] . Hemagglutinin mediates the binding of viral particles to host cell surface sialic acid and the invasion of viruses into host cell [8]–[10] . Neuraminidase is responsible for the cleavage of sialic acid residues to promote the release of progeny viruses [11] , [12] . M2 proton channels are critical for viral mRNA incorporation into the virion and virus budding [13] . Over one hundred serological subtypes [14] have been identified through different combinations of the 16 hemagglutinin ( H1–H16 ) and nine neuraminidase groups ( N1–N9 ) currently known . The 3D-structure of M2 proton channels have recently been solved in both influenza A and B [15] , [16] , allowing more in depth studies regarding its biological function and action mechanism [17]–[19] . These proteins have been used as targets for rational attempts to design drugs for influenza [20]–[27] . The H1N1/09 virus strain is a triple reassortant that contains gene segments from avian , swine and human influenza viruses [28] . In addition to antigenic shift that can lead to fundamental changes in influenza surface antigens , antigenic drift could reduce binding affinity of host antibodies to antigens [29] , [30] . A major challenge in influenza vaccine development is the rapid evolution of influenza viruses , causing vaccines to be easily outdated and reformulation necessary each year [31]–[33] . Although the H1N1/09 virus is susceptible to neuraminidase inhibitors , cases regarding oseltamivir-resistant viruses with neuraminidase mutation ( such as H275Y ) have been reported [34] , [35] . Given that influenza viruses have RNA genomes that are prone to changes , it is imperative to devise new therapies . Much effort has been made to investigate the mechanism and devise alternative drugs against the drug-resistance issue of H1N1 [36]–[40] . Developing inhibitors that target both H1 and N1 antigens can reduce resistance issues resulting from the mutation of a single target antigen . Computational approaches have been widely applied to molecular biology and medicine [41]–[50] . Structure-based methods , including docking and MD simulation , are invaluable tools in drug discovery and design . Computational docking is important for investigating ligand-protein interactions and elucidating binding mechanisms [51]–[57] . Since publication of the pioneer paper in 1977 [58] , it has been established that low-frequency motions existing in proteins and DNA can help reveal dynamic mechanisms underlying fundamental biological functions [59]–[63] . NMR observation later confirmed such inferences and the findings were applied to medical treatments [64]–[67] . In recent years , application of molecular dynamics to investigate internal motions and biological functions of biomacromolecules has opened new frontiers . Vast amounts of information on molecular recognition and binding [68]–[71] , conformations or conformational changes [72]–[75] , molecular mechanisms of bioactivity and stability [76]–[79] , and drug discovery [80]–[84] have been found . To understand interaction of drugs with proteins or DNA , consideration should be given not only to the static structures but dynamical information obtained by simulation through a dynamic process . In this regard , both docking and MD simulation were utilized in this study to provide comprehensive analysis protein-ligand interactions under static and dynamic conditions . Much effort has been placed on developing new , effective influenza treatments , but most have focused on neuraminidase or M2 as the target protein [37] , [38] , [85]–[87] . To date , no hemagglutinin inhibitor is available . Traditional Chinese medicine ( TCM ) has been used extensively for finding effective drugs [88] , and we have successfully designed novel medicinal compounds and identified potential drug leads through traditional Chinese Medicine Database@Taiwan ( TCM@Taiwan ) [89] . Preliminary studies conducted in this lab show potential for TCM compounds to serve as neuramindase and hemagglutinin inhibitors individually [90]–[95] . In view of the current needs for drugs effective against native and mutant H1N1/09 and our promising preliminary results , the present study integrates the concept of “dual targeting” with the aforementioned computational tools and TCM in the attempt to identify dual-targeting inhibitors of H1N1 that may be useful for drug development .
The experimental procedures and screening results after each filtering step are summarized in Figure 1 . Among the 829 native TCM compounds , 81 docked into both H1 and N1 and were used for de novo evolution ( Table S1 ) . De novo compounds with dual binding capacities to H1 and N1 were ranked by combined DockScore and the top ten derivatives are listed in Table 1 . Nine of the ten top ranking de novo compounds were derived from Rosmaricine , a natural compound isolated from Rosemarinus officinalis [96] . The remaining de novo compound was based on Xylopine , which is naturally found in Guatteria amplifolia [97] . The top three derivatives , Xylopine_2 , Rosmaricine_14 and Rosmaricine_15 , have in common a pyridinium addition to their native structure ( Figure 2 ) . The pyridinium addition could be the main explanation for higher DockScores of these three derivatives compared to their native compounds and the other derivatives . Rosmaricine_14 and Rosmaricine_15 differed by the number of fused rings , but the slight difference in DockScore suggests that addition of an acyclic ring has little influence on binding affinity . Docking of the de novo compounds back to the receptor provides insights to modifications that can be made to modulate or enhance molecular properties and also highlights important protein-ligand interactions . When docked into the N1 protein binding site , Xylopine_2 interacts with Asp151 via a protonated amino group and has pi and hydrogen bond ( H-bond ) interactions with Trp179 and Glu228 , respectively ( Figure 3A ) . Rosmaricine_14 ( Figure 3B ) and Rosmaricine_15 ( Figure 3C ) , have interactions with Asp151 and Arg293 via the carbonyl group and Glu228 through the 2-aminopyridinium group . Tamiflu® forms H-bond interactions with Arg156 , Arg293 and Arg368 , but not with Asp151 or Glu228 ( Figure 3D ) . Both Asp151 and Glu228 have been reported as one of the major residues in the N1 ligand binding site [98] , [99] . The ability of the de novo derivatives to form interactions with both Asp151 and Glu228 may account for the higher DockScores . Binding of the top three de novo derivatives to H1 site is detailed elsewhere [92] . The ability to bind with important H1 residues Asp103 and Arg238 [100] indicates the dual targeting possibility of the candidates . The top ranking model generated by genetic function approximation ( GFA ) includes the following descriptors: ES_Sum_dssC , CHI_3_C , Kappa_1 , Jurs_PNSA_1 , and Jurs_RPCS . Utilizing these five descriptors , the MLR model established for the neuramindase inhibitors is: Correlation between the observed and predicted activities of the 27 ligands are shown in Figure 4A . All values were within the 95% prediction bands and the r2 value = 0 . 8043 . The SVM model was constructed using identical molecular descriptors and ligands as the MLR model . The r2 value of the SVM model was 0 . 8605 and the correlation between observed and predicted activities of 27 ligands are illustrated in Figure 4B . Table 2 summarizes the pIC50 values of Tamiflu® and the top three candidates as predicted by the generated MLR and SVM models . The predicted activity of Tamiflu® using the generated MLR model ( pIC50 = 7 . 613 ) is similar to observed bioactivity values reported in the literature ( pIC50 = 7 . 823 ) [101] . This indicates that the generated MLR model is a good prediction model . Predicted activity values using the SVM model indicate a lower pIC50 with regard to Tamiflu® . Nonetheless , both models indicate that all TCM de novo derivatives are good candidates with neuraminidase inhibitory activity . MLR and SVM models for predicting hemagglutinin inhibitory activity were not established due to the lack of available hemagglutinin inhibitor structures in the literature . To further investigate docking features , CoMFA and CoMSIA models were built and validated using 27 neuraminidase inhibitors listed in Table S2 . The PLS analyses results for CoMFA and CoMSIA models are shown in Table 4 . The CoMFA model was generated using both steric and electrostatic fields and yielded a non-cross validated r2 value of 0 . 924 and a cross validated q2 value of 0 . 524 with an optimal number of components as 5 . The optimal CoMSIA model ( r2 = 0 . 937 , q2 = 0 . 673 , ONC = 5 ) consisted of steric and hydrophobic fields , H-bond acceptors and donors . When compared against actual observed activities [102] , both CoMFA and CoMSIA models had good predictability , predicting pIC50 values that differed only marginally from the actual pIC50 values of 24 compounds ( Table 5 ) . The validated CoMFA and CoMSIA maps were used to assess ligand bioactivity . Contour of the de novo compounds at 20 ns MD simulation to the relative spatial positions of CoMFA and CoMSIA feature maps are shown in Figure 11 . In Xylopine_2 , Rosmaricine_14 and Rosmaricine_15 , the H-bond between the 2-aminopyridinium group and Glu228 matched the electropositive group feature of the CoMFA model ( Figure 11A , 11C , 11E ) and the H-bond donor feature in CoMSIA model ( Figure 11B , 11D , 11F ) . The hydrophobic benzene structures of Xylopine_2 matched the steric favoring region of the CoMFA map and the hydrophobic feature of the CoMSIA map . The carbonyl groups in Rosmaricine_14 and Rosmaricine_15 which formed H-bonds with Tyr402 satisfied the H-bond acceptor feature in the CoMSIA model . Tamiflu® also contours to both CoMFA and CoMSIA models . The 3-methoxypentane group close to Arg293 and Asn344 matched the steric favoring region of CoMFA ( Figure 11G ) and the hydrophobic feature of CoMSIA ( Figure 11H ) . This residue has similar characteristics to the 2-aminopyridinium group in the de novo derivatives . In addition , the N-methylacetamide group in Tamiflu® , which forms H-bond with Tyr402 , is located near the H-bond donor feature in CoMSIA . Though all compounds contoured to the N1 inhibitor features identified by CoMFA and CoMSIA , a critical difference was observed between Tamiflu® and the TCM de novo derivatives . All compounds except Tamiflu® formed H-bonds at Glu228 . As Glu228 is a primary binding site of N1 [99] , ability of the TCM de novo derivatives to maintain stable binding with Glu228 during MD simulation supports the potential of these compounds as drug alternatives to Tamiflu® . Due to the lack of reported H1 ligand bioactivities in the literature , direct assessment of bioactivity through construction of CoMFA and CoMSIA models was not possible . Alternatively , indirect support was provided by assessing the ability of de novo derivatives to maintain contour to the N1 CoMFA/CoMSIA maps while forming interactions at key residues in H1 , Glu83 and Asp103 [92] . As illustrated in Figure 12 the TCM de novo derivatives docked into the H1 binding site and formed critical interactions at Glu83 and Asp103 without losing contour to the CoMFA and CoMSIA maps . These results suggest that not only were the TCM de novo derivatives capable of docking into both H1 and N1 , but that biological activity was also predicted in both binding sites , thus it is possible to develop dual-targeting drugs from the selected de novo derivatives . Important features for potential H1 and N1 inhibitors are summarized in Figure 13 . For H1 , a salt bridge with Glu83 and H-bond donor and/or electrostatic interactions with Asp103 are important characteristics that should be met . Potential inhibitors for N1 should have salt bridge and/or H-bond formation at Glu228 and interactions with Asp293 . These features can be used to identify or design novel drugs for H1 and/or N1 . In the case of the TCM de novo derivatives from this study , each compound could structurally fulfill the requirements of both H1 ( Figure 13A , 13B , 13C ) and N1 ( Figure 13D , 13E , 13F ) binding sites , thus supporting their potential as dual-targeting compounds . In this research , we identified Xylopine_2 , Rosmaricine_14 , and Rosmaricine_15 as the top three de novo derivatives exhibiting binding affinity to H1 and N1 . Addition of a pyridinum residue to the native structures of xylopine and rosmaricine contributes to bond formation at key residues in both H1 ( Glu83 , Asp103 ) and N1 ( Glu228 , Arg292 ) . The de novo derivatives were predicted as active by the SVM and MLR models , and contoured well to the 3D-QSAR models . The TCM de novo derivatives were able to maintain contour while forming key binding interactions in H1 , thus providing indirect support for bioactivity in H1 . The results of this study indicate that the TCM de novo derivatives not only can bind to , but can also exhibit biological activities in both H1 and N1 . Key binding locations of the de novo derivatives include Glu83 and Asp103 for H1 , and Glu228 and Arg292 for N1 . Mutations currently attributed to oseltamivir resistance are located at H275 and N295S of the NA [103] . Since the key binding locations of the TCM derivatives do not overlap with those causing oseltamivir resistance , derivatives will be able to bind to viruses that are currently resistant to Tamiflu® . In addition , the de novo derivatives do not bind to amino acids in H1 or N1 that are prone to mutation ( Table 6 , Table 7 ) [40] , [104] , thus would likely be able to exert activity across a range of mutant H1N1 viruses . Last but not the least , multiple bond formations observed in MD provide additional insurance against possible mutations at key binding residues . In the case of a single point mutation , the de novo compounds will remain bound to the H1 and N1 sites through another key residue , therefore resisting the development of drug resistance in the virus . Based on the results and observations of this study , the TCM de novo derivatives may be attractive compounds for designing novel dual-target inhibitors for H1 and N1 .
Virtual screening , de novo derivative generation , and molecular dynamics ( MD ) simulation were performed using Discovery Studio Client v2 . 5 . 0 . 9164 ( DS2 . 5; Accelrys Inc . , San Diego , CA ) . The two-dimensional and three-dimensional structures of TCM compounds were generated using ChemBioOffice 2008 ( PerkinElmer Inc . , Cambridge , MA ) . Comparative molecular field analysis ( CoMFA ) and comparative molecular similarities indices analysis ( CoMSIA ) models were constructed using SYBYL© 8 . 3 package ( Tripos Inc . , St . Louis , MO ) . Compounds from the TCM Database@Taiwan were docked to H1 and N1 protein active sites reported in our previous study [91] . All procedures were completed under the forcefield of Chemistry at HARvard Molecular Mechanics ( CHARMm ) [105] . The virtual screening process was performed using LigandFit . The conformational search method was based on the Monte Carlo algorithm . Rigid body minimization following initial ligand placement was completed using Smart Minimizer . Scoring functions used by LigandFit were DockScore . TCM compounds that docked into both H1 and N1 proteins were selected and then ranked by the sum of their H1 and N1 DockScore . Tamiflu® was used as the control for N1 , and its N1 docking score was set as the minimum requirement . The top TCM compounds that passed the filtering were selected for de novo evolution . In de novo evolution , TCM compounds were placed into the H1 and N1 protein binding sites described previously , and Ludi-fragments were attached to the native structure . The new derivatives were generated in full evolution mode . Derivatives from de novo evolution were subjected to additional screening through Lipinski's rule [106] to rule out orally unstable or pharmacologically inapplicable compounds . As de novo products generated for H1 and N1 proteins differed , all de novo products were re-docked to H1 and N1 proteins to assess binding affinity . De novo products that docked into both H1 and N1 proteins were selected and ranked by the sum of their respective H1 and N1 DockScore . The top ten compounds with the highest DockScore were selected for further structure-based analysis . The 27 neuraminidase inhibitors used , including 24 training set compounds and 3 test set compounds , were adapted from Zhang's study [102] . Compounds were drawn using ChemBioOffice 2008 ( PerkinElmer Inc . , Cambridge , MA ) and modified to physiological ionization using the Prepare Ligand function in DS 2 . 5 . Bioactivity values ( IC50 ) were also obtained from Zhang's study though the original sources were not clarified , and converted to pIC50 ( log ( 1/IC50 ) ) . Molecular descriptors of the compounds were calculated using Calculate Molecular Properties in DS 2 . 5 and the GFA was used to select the best representative molecular descriptors [107] . Utilizing the best representative molecular descriptors identified through GFA , MLR and SVM models were constructed using MATLAB ( The Mathworks Inc . , Natick , MA ) and LibSVM [108] , respectively , and used to predict the bioactivity of TCM de novo compounds . The MD simulation was performed using the Molecular Dynamics package of DS 2 . 5 . The complexes were created with a 10 Å solvation shell of TIP3 water around the protein . Sodium cations were added to each system for neutralization . Minimization using Steepest Descent and Conjugate Gradient were performed at 500 cycles each . Each protein-ligand complex was gradually heated from 0K to 310K over 50 ps , followed by a 200 ps equilibration phase . The production stage was performed for 20 ns using NVT canonical ensemble and trajectory frames were saved every 20 ps . SHAKE algorithm was applied to immobilize all bonds involving hydrogen atoms throughout the MD simulation . Long-range electrostatics were treated with PME method . Time step was set to 2 fs for all MD stages . The temperature coupling decay time for the Berendsen thermal coupling method was 0 . 4 ps . Post processing of the trajectory was performed using Analyze Trajectory module . Torsion angles of each bond were also monitored through DS 2 . 5 . LIGPLOT [109] was used to generate schematic diagrams of protein-ligand interactions for each candidate and control in H1 and N1 . CoMFA and CoMSIA models were constructed through the partial least square ( PLS ) analysis using previously described neuraminidase inhibitors [102] . The optimal number of components was obtained from leave-one-out method to yield the highest r2 and q2 values in non-cross validation and cross-validation , respectively . Biological activities of the TCM de novo compounds were evaluated based on contour to the generated 3D-QSAR map . | The influenza A subtype H1N1 ( H1N1/09 ) pandemic raised public concerns due to drug resistance strains . Drug resistance occurs from conformational changes causing the original drug to lose binding ability and exhibit biological effects . The world's largest TCM Database@Taiwan was employed to screen for potential leads that simultaneously bind to H1 and N1 . Three de novo compounds derived from Rosemarinus officinalis and Guatteria amplifolia were identified as having dual binding properties to H1 and N1 . Structural analysis indicated that the candidates bind to multiple residues in both H1 and N1 . In addition , the de novo derivatives were predicted as bioactive using four different computational models . The compounds are validated as potent dual targeting influenza drug candidates through multiple validations . Key advantages of the candidates include ( 1 ) binding to H1 and N1 through multiple amino acids , and ( 2 ) not binding to known mutation residues in H1 or N1 . Such advantages can reduce drug resistance caused by single point mutations . On a broader context , features important for successful H1N1 drug development are discussed in hopes of providing starting templates for drug development and improvements . | [
"Abstract",
"Introduction",
"Results/Discussion",
"Materials",
"and",
"Methods"
] | [
"biotechnology",
"medicine",
"physics",
"computer",
"science",
"computer",
"modeling",
"drug",
"research",
"and",
"development",
"drugs",
"and",
"devices",
"pharmacology",
"biology",
"computational",
"biology",
"biophysics",
"simulations",
"biophysics",
"drug",
"discovery... | 2011 | Two Birds with One Stone? Possible Dual-Targeting H1N1 Inhibitors from Traditional Chinese Medicine |
Bluetongue virus ( BTV ) is the causative agent of a major disease of livestock ( bluetongue ) . For over two decades , it has been widely accepted that the 10 segments of the dsRNA genome of BTV encode for 7 structural and 3 non-structural proteins . The non-structural proteins ( NS1 , NS2 , NS3/NS3a ) play different key roles during the viral replication cycle . In this study we show that BTV expresses a fourth non-structural protein ( that we designated NS4 ) encoded by an open reading frame in segment 9 overlapping the open reading frame encoding VP6 . NS4 is 77–79 amino acid residues in length and highly conserved among several BTV serotypes/strains . NS4 was expressed early post-infection and localized in the nucleoli of BTV infected cells . By reverse genetics , we showed that NS4 is dispensable for BTV replication in vitro , both in mammalian and insect cells , and does not affect viral virulence in murine models of bluetongue infection . Interestingly , NS4 conferred a replication advantage to BTV-8 , but not to BTV-1 , in cells in an interferon ( IFN ) -induced antiviral state . However , the BTV-1 NS4 conferred a replication advantage both to a BTV-8 reassortant containing the entire segment 9 of BTV-1 and to a BTV-8 mutant with the NS4 identical to the homologous BTV-1 protein . Collectively , this study suggests that NS4 plays an important role in virus-host interaction and is one of the mechanisms played , at least by BTV-8 , to counteract the antiviral response of the host . In addition , the distinct nucleolar localization of NS4 , being expressed by a virus that replicates exclusively in the cytoplasm , offers new avenues to investigate the multiple roles played by the nucleolus in the biology of the cell .
Bluetongue is a major infectious disease of ruminants caused by an arbovirus ( Bluetongue virus , BTV ) transmitted by biting midges ( Culicoides spp . ) [1]–[3] . Historically , bluetongue has been endemic almost exclusively in temperate and tropical areas of the world where the climatic conditions favour both the spread of the susceptible insect vector population and the virus replication cycle within the vector [4] . However , in the last decade BTV has spread extensively in several geographical areas including Southern Europe and also , unexpectedly , in Northern Europe causing a serious burden to both animal health and the economy [5] , [6] . From a molecular and structural virology perspective BTV is one of the best understood animal viruses . BTV is a member of the Orbivirus genus , within the Reoviridae family , and possesses a double-stranded RNA genome formed by 10 segments ( Seg-1 to Seg-10 ) of approximately 19200 base pairs in total [1] , [3] . Until now , the BTV genome has been shown to encode for 7 structural and 3 non-structural proteins . The BTV genome is packaged within a triple layered icosahedral protein capsid of approximately 90 nm in diameter [1] , [7]–[10] . The outer capsid of the virion is composed by 60 trimers of VP2 and 120 trimers of VP5 [11] and differences within this outer capsid define the 26 BTV serotypes which have been described so far [12] , [13] . The outer capsid proteins , and VP2 in particular , stimulate virus neutralizing antibodies which in general protect only against the homologous serotype [14] . The internal core is formed by two layers , constituted by VP3 ( sub-core ) and the immunodominant VP7 ( intermediate layer ) [7] . Three minor enzymatic proteins , VP1 ( RNA dependent RNA polymerase ) , VP4 ( capping enzyme and transmethylase ) and VP6 ( RNA dependent ATPase and helicase ) are contained within the core that is transcriptionally active in infected cells [15]–[21] . The BTV genome encodes also 3 non-structural proteins: NS1 , NS2 and NS3/NS3a . NS1 and NS2 are highly expressed viral proteins and their multimers are morphological features of BTV-infected cells . Multimers of the NS1 protein form tubules ( approximately 50 nm in diameter and up to 1000 nm in length ) that appear to be linked to cellular cytopathogenicity [22] , while NS2 is the major component of the viral inclusion bodies . NS2 plays a key role in viral replication and assembly as it has a high affinity for single stranded RNA and possesses phosphohydrolase activity [23] . NS3/NS3a are glycosylated proteins involved in BTV exit . There are two isoforms of NS3: NS3 and NS3a with the latter lacking the N-terminal 13 amino acid residues [24]–[26] . Therefore , the segmented genome of BTV has been thought to be monocistronic ( i . e . ten genome segments encoding for 10 proteins ) for almost three decades [27] , [28] . Segment 9 however , contains the open reading frame ( ORF ) encoding VP6 but also a smaller coding sequence in the position +1 reading frame that is present in BTV and some related Orbiviruses such as African horse sickness virus and others [29] . Bioinformatic analysis predicts that the BTV “ORFX” encodes for a protein of 77–79 amino acid residues . This putative ORFX is subject to functional constraints at the amino acid level and its level of conservation is higher compared to that of the overlapping VP6 . In addition , the ORFX putative AUG initiation codon has a strong Kozak context suggesting that this protein might be translated by leaky scanning [29] . Alternative reading frames are expressed in a variety of RNA viruses and they can play fundamental roles in viral replication and virus-host interaction . In this study , we identified a previously unknown non-structural protein and characterized its biological properties .
All experimental procedures carried out in this study are included in protocol number 5182/2011 of the Istituto G . Caporale approved by the Italian Ministry of Health ( Ministero della Salute ) in accordance with Council Directive 86/609/EEC of the European Union and the Italian D . Igs 116/92 . BSR cells ( a clone of BHK21 , kindly provided by Karl K . Conzelmann ) were grown in Dulbecco's modified Eagle's medium ( DMEM ) supplemented with 10% fetal bovine serum ( FBS ) . Bovine foetal aorta endothelium ( BFAE ) cells were obtained from the Health Protection Agency ( HPA ) cell culture collection ( catalogue number 87022601 ) , and were grown in Ham's F12 medium supplemented with 20% FBS . CPT-Tert cells [30] are sheep choroid plexus cells immortalized with the simian virus 40 ( SV40 ) T antigen and human telomerase reverse transcriptase ( hTERT ) and were kindly provided by David Griffiths . CPT-Tert cells were grown in Iscove's modified Dulbecco's medium ( IMDM ) , supplemented with 10% FBS . Mammalian cell lines were cultured at 35 to 37°C , in a 5% CO2 humidified atmosphere . C6/36 cells are mosquitoes cells established from Aedes albopictus and were kindly provided by Richard Elliott . C6/36 cells were grown in Leibovitz's L-15 medium supplemented with 10% FBS and 10% tryptose phosphate broth . KC cells [31] ( established from Culicoides sonorensis larvae ) were grown in Schneider's insect medium and were supplemented with 10% FBS . Insect cells were incubated at 28°C . Initially , the open reading frame expressing ORFX ( NS4 ) was amplified by PCR from BTV-10 ( GenBank accession number D00509 ) and cloned into the pCI Mammalian Expression Vector ( Promega ) resulting into pCI-NS4 . The BTV-8 NS4 was cloned into the peGFP-N1 vector ( Clontech ) , resulting in plasmid pNS4-GFP . pNS47–77-GFP , pNS413–77-GFP and pNS419–77-GFP are mutants derived from pNS4-GFP expressing NS4 truncated of the amino terminal 6 , 12 and 18 amino acid residues , respectively . pNS47–77-GFP , pNS413–77-GFP and pNS419–77-GFP maintain the methionine and valine residues in position 1 and 2 of NS4 . Note that BTV-10 and BTV-1 NS4 are 100% identical at the amino acid level . While BTV-8 and BTV-1 NS4 differ for a single amino acid residue in position 6 . The set of BTV-1 and BTV-8 plasmids necessary to rescue these viruses in vitro by reverse genetics were obtained following the method recently published by Boyce and colleagues [26] . Briefly , total RNA was extracted from infected cells using Trizol ( Invitrogen ) according to the manufacturer's instructions . Each BTV genome segment was amplified by RT-PCR using the AccuScript PfuUltra II RT-PCR Kit ( Agilent ) from either BTV-1 or BTV-8 dsRNA preparations and the resulting PCR products were gel-purified ( Qiagen ) and cloned into either pUC57 ( Fermentas ) or pCI . Each BTV segment was cloned downstream of a T7 promoter and upstream of a BsaI or SapI restriction site . All of the mutants described in this study were obtained using the QuikChange II Site-Directed Mutagenesis Kit ( Stratagene ) , according to the manufacturer's instructions . All plasmids used in this study were completely sequenced before use . Sequences of PCR primers used in this study are available upon request . Antisera used in this study included polyclonal rabbit antisera raised against BTV VP7 , NS1 , NS2 , NS3 and ORFX ( NS4 ) expressed in bacteria as Glutathione S-transferase ( GST ) -tagged recombinant proteins ( Proteintech Group , Inc . ) . Antiserum against BTV-1 NS4 was raised against a recombinant GST fusion protein including the entire NS4 protein expressed in bacteria . Polyclonal rabbit antiserum against BTV VP6 was kindly provided by Polly Roy as previously described [32] . Antibodies against B23 and γ-tubulin were obtained commercially ( Sigma Aldrich ) . BTV-8 ( IAH reference collection number NET2006/04 ) was originally isolated from a naturally infected sheep during the 2006 outbreak in Northern Europe [33] . The virus was passaged once in KC cells and once in BHK21 cells . The reference strain of BTV-1 was originally isolated at the ARC – Onderstepoort Veterinary Institute ( IAH reference collection number RSArrrr/01 ) and was adapted to cell culture by passaging it twice in embryonated eggs and 9 times in BHK21 cells . Both viruses were kindly provided by Peter Mertens . Virus stocks were prepared by infecting BSR cells at a multiplicity of infection ( MOI ) of 0 . 01 and collecting the supernatant when obvious cytpopathic effect ( CPE ) was observed . The supernatants were clarified by centrifugation at 500 g for 5 min and the resulting virus suspensions aliquoted and stored at 4°C for short term usage and at −70°C for long term storage . Virus titres were determined by standard plaque assays using BSR or CPT-Tert cells [34] . 65 full-length segment 9 sequences representing 24 BTV serotypes were obtained from GenBank . Amino acid conservation plots and secondary structure predictions were obtained using the CLC Genomics Workbench ( CLC , Aarhus , Denmark ) software and bioinformatics tools available online ( the PSIPRED server [http://bioinf . cs . ucl . ac . uk/psipred] , and the Network Protein Sequence Analysis ( nps@ ) server , [http://npsa-pbil . ibcp . fr/] ) . Recombinant BTVs were rescued by reverse genetics as previously described [26] . Briefly , plasmids containing the genomic segments of BTV-1 or BTV-8 or resulting mutants were linearized with the appropriate restriction enzymes and then purified by phenol-chloroform extraction . Digested plasmids were used as a template for in vitro transcription using the mMESSAGE mMACHINE T7 Ultra Kit ( Ambion ) , according to the manufacturer's instructions . ssRNAs were purified sequentially by phenol/chloroform extraction and through Illustra Microspin G25 columns ( GE Healthcare Life Sciences ) , following the manufacturer's protocol . Monolayers of 95% confluent BSR cells grown in 12 well plates were transfected twice with BTV RNAs using Lipofectamine 2000 ( Invitrogen ) . Firstly , 0 . 5×1011 ( BTV-1 ) or 1×1011 ( BTV-8 ) molecules of each of the BTV segments encoding VP1 , VP3 , VP4 , NS1 , VP6 and NS2 were diluted in Opti-MEM I Reduced Serum Medium containing 0 . 5 U/µL of RNAsin plus ( Promega ) and then mixed with Lipofectamine 2000 diluted in Opti-MEM I Reduced Serum Medium . After 25 min of incubation at room temperature , the mixture was added to the cells . 16 to 18 h after the first transfection , the cells were transfected as before but with all 10 BTV segments . 3 to 4 h after the second transfection the cells were overlaid with 2 ml of minimal essential media containing 1 . 5% agarose type VII and 2% FBS , and monitored for development of plaques . Finally , individual BTV rescued clones were picked through the agarose overlay and used to infect fresh BSR cells in order to obtain a virus stock . Where necessary , BTV dsRNA was extracted from infected cells using Trizol ( Invitrogen ) . The ssRNA fraction was precipitated using lithium chloride , and the harvested dsRNA fraction was precipitated using isopropanol in the presence of sodium acetate . Growth curves of BTV recombinant viruses used in this study were derived in cells infected at a MOI of 0 . 05 and testing for the presence of infectious virus in supernatants collected at 8 , 24 , 48 , 72 and 96 h post-infection . Virus growth was also assessed in cells in the presence of 1000 antiviral units/ml ( AVU/ml ) of interferon Tau ( IFNT ) or universal type I interferon ( UIFN ) . Recombinant ovine IFNT was kindly provided by Tom Spencer . IFNT was produced in Pichia pastoris and purified as described previously [35] . Universal type I Interferon ( UIFN ) was obtained from PBL InterferonSource . BFAE cells and CPT-Tert cells were treated with 1000 AVU of IFN 20 h prior infection with BTV recombinants at a MOI of 0 . 1 ( BFAE and CPT-Tert ) , 0 . 01 or 0 . 001 ( CPT-Tert ) . Two hours after infection , the medium was replaced and the cells maintained in the presence of either IFNT or UIFN at the original concentration . Cell supernatants were collected at 24 , 48 and 72 h post-infection , centrifuged for 5 min at 500 g in order to pellet cell debris and virus infectivity was subsequently titrated by endpoint dilution analysis on BSR cells . Viral titers were calculated by the method of Reed & Muench and expressed as log10 TCID50/ml [36] . Each experiment was performed two to three times , each time in duplicate , using different stocks for each virus . CPT-Tert cells were plated in 24-well plates and treated for 20 h with 1000 AVU/ml of IFNT or UIFN and then infected with either BTV-1 or BTV-8 at different MOIs ( 0 . 1 , 0 . 01 and 0 . 001 ) . The medium was replaced 2 h after infection and the cells maintained in the presence of either IFNT or UIFN at the original concentration . At 72 h post-infection , the cells were washed once with phosphate buffered saline ( PBS; pH 7 . 4 ) and stained for 16 h using a 0 . 5% crystal violet/10% formaldehyde solution . We used Image-Pro Plus ( MediaCybernetics ) , in order to quantify in each well the percentage of the monolayer that was disrupted after BTV replication . Results were expressed as the percentage of destroyed monolayer by calculating for each well the following formula: ( number of pixels above background: total number of pixels times ) X 100 . BSR cells were transfected with 0 . 6–1 . 8 µg of either pCI-NS4 , pNS4-GFP or derived deletion mutants , using Lipofectamine 2000 ( Invitrogen ) according to the manufacturer's instructions . For western blot analyses of intracellular proteins , cells were lysed by standard techniques as described previously [37] . For viral pellet analysis , cell supernatants were collected and viral particles concentrated 200 times by ultracentrifugation as previously described [38] . Protein expression was assessed by sodium-dodecyl-sulfate polyacrylamide gel electrophoresis ( SDS-PAGE ) and western blotting using the various antisera as indicated above . Membranes were incubated with a horse radish peroxidase-conjugated secondary antibody ( GE Healthcare Life Sciences ) and developed by chemiluminescence using Amersham ECL Plus Western Blotting Detection Reagents ( GE Healthcare Life Sciences ) . Experiments were performed using BSR , BFAE , CPT-Tert or C6/36 cells cultured in two-well glass chamber slides ( Lab-Tek , Nalge Nunc International ) . Cells were either transfected with appropriate plasmids or infected with various BTV strains at a MOI between 0 . 01 and 1 . 5 . Cells were washed with PBS and fixed with 5% formaldehyde for 15 minutes . The fixed cells were then processed as described previously [39] and incubated with the appropriate antisera . Secondary antibodies were conjugated with Alexa Fluor 488 ( Invitrogen , Molecular Probes ) or Alexa Fluor 594 ( Invitrogen , Molecular Probes ) . Slides were mounted using VECTASHIELD Mounting Medium with DAPI ( 4′ , 6-diamidino-2-phenylindole , Vector Laboratories ) . Slides were analysed and images collected using a Leica TCS SP2 confocal microscope . BSR cells were infected with BTV-1 , BTV-8 or the corresponding deletion mutants at a MOI of 0 . 05 in 35 mm dishes . At 24 h post-infection , cells were fixed using cold 2 . 5% gluteraldehyde and 1% osmium tetroxide . Cells were subsequently pelleted through 1% SeaPlaque agarose ( Flowgen ) , dehydrated using a graded alcohol series and embedded in Epon 812 resin , followed by cutting and analysis in a Joel 1200 EX II electron microscope . Animal experiments were carried out at the “Istituto G . Caporale” ( Teramo , Italy ) following local and national approved protocols regulating animal experimental use . Study 1 . Litters of 3-day old NIH-Swiss mice ( n = 8–12 ) , were inoculated intra-cerebrally with 103 TCID50 of either BTV-1 , BTV-8 , BTV-1ΔNS4 or BTV-8ΔNS4 . Mock-infected controls included litters inoculated with tissue culture media . Mice were euthanized at three weeks p . i . or earlier if showing advanced clinical signs of encephalitis . For each virus , two litters were inoculated using two different virus preparations . Study 2 . Age-matched adult transgenic mice , deficient in the type I interferon ( IFN ) receptor [IFN Alpha Ro/o IFNAR ( −/− ) 129/Sv] , were inoculated intraperitoneally with 100 PFU of either BTV-1 , BTV-8 , BTV-1ΔNS4 or BTV-8ΔNS4 . For each virus , two groups ( n = 5 ) of mice were inoculated using two different virus preparations . Survival plots were constructed using data collected from two experimental groups ( n = 10 ) with the exception of a mock infected group that was constituted by a single group of 6 mice . Formalin-fixed and paraffin-embedded brains tissue sections from inoculated ( and mock inoculated ) mice were used in immunohistochemistry . Sections ( 4–6 µm ) were examined for the presence of BTV NS4 using a polyclonal NS4 antiserum and the EnVision ( DAKO ) detection system .
The BTV genome is formed by 10 segments . Segment 9 contains an open reading frame ( ORF ) between nucleotides 182 and 418 ( Figure 1A ) in position +1 with respect to the major ORF expressing VP6 [29] . In silico analysis showed that this extra ORF is highly conserved and encodes a putative protein of 77–79 amino acid residues . A stretch of 11 basic amino acid residues is present in the N terminal portion of the protein ( residues 3 to 20 ) . In addition , there are two putative alpha helices . Of note , the C terminal helix ( residues 34 to 75 ) contains a conserved leucine zipper domain , with leucine residues at positions 49 , 56 , 63 and 70 ( Figure 1A ) . We generated a polyclonal antiserum towards ORFX , in order to assess whether BTV expressed this previously uncharacterized protein . We detected ORFX in BSR cells infected with either BTV-8 or BTV-1 by western blotting ( Figure 1B ) . Controls included BSR cells transfected with plasmids expressing ORFX either in its native form , or with eGFP fused to its C terminus . These data confirm that BTV expresses a protein encoded by an alternative reading frame located in segment 9 . We subsequently investigated whether ORFX was a structural or non-structural protein . We infected BFAE cells with BTV-1 and analysed supernatants ( containing viral particles ) and total cellular protein extracts . Unusually for mammalian cell lines , BFAE cells show very little BTV induced cytopathic effect ( CPE ) , thus facilitating the efficient discrimination between all BTV proteins present in the cellular fraction and the structural proteins present in purified and concentrated viral particles released from infected cells . By western blotting , we detected NS1 and ORFX in the cellular fraction , while VP7 was abundantly present in the viral fraction ( concentrated by ultracentrifugation ) and barely visible in the cellular fraction ( Figure 1C ) . We obtained the same results by infecting C6/36 mosquito cells ( data not shown ) . We detected VP6 in both the cellular and the viral fraction ( Figure 1C ) . Interestingly , unlike VP7 , VP6 appeared to be relatively more abundant in cell lysates compared to the viral pellets , suggesting that there is an intracellular pool of this protein that is not incorporated in the BTV virions . The absence of ORFX in the viral pellet strongly suggested that this is a non-structural protein expressed by BTV . In light of these data , we designated this protein NS4 . NS4 was also expressed in vivo , as shown by immunohistochemistry of brain sections of mice inoculated intracerebrally with BTV ( Figure 1D ) . By confocal microscopy of cells transiently transfected with pCI-NS4 , we observed that NS4 localized mainly in the nucleus ( Figure 2A ) where it showed a strong co-localization with the nucleolar marker B23 [40] . Importantly , cells infected with either BTV-1 or BTV-8 also showed a strong nuclear co-localisation between NS4 and B23 [40] ( Figure 2B ) . We also observed NS4 to localise in the nucleus of the C6/36 insect cells ( Figure 2C ) . NS4 does not have a canonical nuclear localization signal ( NLS ) but possesses a stretch of basic amino acid residues , at the amino terminus portion of the protein , that could drive nuclear localization [41] ( Figure 2D ) . We constructed an NS4 expression plasmid ( pNS4-GFP ) and a series of deletion mutants ( pNS47–77-eGFP , pNS413–77-eGFP and pNS419–77-eGFP ) lacking the 6 , 12 and 18 amino terminal residues , respectively . pNS4-GFP and pNS47–77-eGFP transfected cells showed a strong nuclear localization of NS4 . On the other hand , NS4 showed a predominantly cytoplasmic localization in cells transfected with either pNS413–77-eGFP or pNS419–77-eGFP . These data suggest that the amino terminal basic domain of NS4 may play an important role in the nuclear localization of this protein . Interestingly , BFAE cells infected with BTV-1 revealed that NS4 expression was evident as early as 2 hours post infection , similar to that observed for other BTV structural and non-structural proteins ( Figure 3 ) . The data above clearly show that a previously uncharacterized BTV protein , here referred to as NS4 , is a non-structural protein that localises to the nucleolus of infected cells . Next , we generated by reverse genetics BTV NS4 deletion mutants in order to assess the requirement of this protein for viral replication . We generated a set of plasmids necessary for the rescue of BTV-1 and BTV-8 and engineered three mutations in the plasmids containing segment 9 of BTV-1 and BTV-8 such that the NS4 initiation codon was removed along with the introduction of two stop codons in the NS4 coding sequence . All the mutations introduced were designed in order to leave the VP6 amino acid sequence unaltered ( Figure 4A ) . As a negative control for BTV rescue , we designed a VP6 deletion mutant with a premature stop codon incorporated into the VP6 coding sequence ( position 79 ) . As shown in Figure 4B , viable BTV1-ΔNS4 and BTV8-ΔNS4 were rescued with similar efficiency to the respective wild-type ( wt ) viruses , upon transfection of RNA transcribed in vitro from the appropriate plasmids representing the genomic segments of wt or mutated BTV-1 and BTV-8 . As expected , BTV-1ΔVP6 and BTV8-ΔVP6 could not be rescued . We did not detect any variation in the migration pattern of dsRNA genomic segments extracted from all the wt or the NS4 deletion mutant viruses ( Figure 4C ) . The RNA profiles of both the wt and ΔNS4 rescued viruses were identical to the corresponding profile of the stock viruses from which the segments were originally cloned . For each virus , segment 9 was completely sequenced in order to confirm the presence of the introduced mutations . We confirmed , by western blotting and confocal microscopy , that the ΔNS4 mutants do not express NS4 but express levels of VP7 and NS2 comparable to the parental wild type viruses . It was also evident that in BSR cells BTV-8 expresses lower amounts of NS4 relative to BTV-1 ( Figure 4D and not shown ) . However , BTV-1 replicates better than BTV-8 in these cells and differences in the steady-state levels of VP7 between these two viruses were also observed ( Figure 4D ) . In cells infected by BTV1-ΔNS4 or BTV8-ΔNS4 , we found by electron microscopy all the ultrastructural features of BTV-infected cells ( e . g . viral inclusion bodies , NS1 tubules , viral particles ) ( Figure 4E ) . We next assessed the replication kinetics of the rescued viruses in a variety of mammalian and insect cell lines , including those corresponding to the natural hosts ( sheep and cattle ) and vector ( midges ) . All subsequent experiments were performed using the rescued versions of the wt viruses as they represent a more homogenous population and are therefore more directly comparable to the rescued ΔNS4 viruses . Cells were infected with a MOI of 0 . 05 and supernatants were collected at various times post-infection ( Figure 5 ) . No obvious difference was obtained in the replication of wt and ΔNS4 viruses , regardless of the cell lines used in the assay ( Figure 5 ) . Interestingly , the cell adapted BTV-1 viruses consistently grew more efficiently in vitro than the BTV-8 equivalent , including in cell lines derived from the natural host ( BFAE , CPT-Tert and KC ) . BTV , like most RNA viruses , is a strong inducer of interferon , both in vivo in its natural hosts and in vitro [42]–[44] . Given that other RNA viruses express proteins that counteract the innate immunity of the host , we hypothesised that NS4 might aid BTV replication in the presence of interferon ( IFN ) . We treated cells with two type I IFNs: IFN tau ( IFNT ) and universal IFN ( UIFN ) . IFNT is secreted by the ruminant conceptus and it is intimately linked to pregnancy recognition signalling and possesses antiviral activity [45] while UIFN is an alpha interferon hybrid constructed from recombinant Human IFNs alpha A and alpha D , and is known to stimulate an antiviral response in a wide variety of mammalian cells . CPT-Tert cells were pre-treated with IFNT or UIFN for 20 h prior to infection with BTV-1 or BTV-8 ( or mock infection ) with MOIs ranging from 0 . 001 to 0 . 1 . Both wt and the ΔNS4 mutants , destroyed 80 to 100% ( depending on the MOI used ) of the monolayer of infected cells in absence of IFN treatment ( Figure 6 ) . On the other hand , pre-treatment with both types of IFN significantly reduced BTV-induced CPE . Interestingly , in the presence of IFN , BTV-8 wt consistently induced a more pronounced CPE than BTV8-ΔNS4 . Conversely , only minor differences were observed in the CPE induced by both wt BTV-1 and BTV-1ΔNS4 in the presence of IFN ( Figure 6 ) . Subsequently , we performed multi-step virus growth curves in order to further assess the replication of BTV wt and ΔNS4 in the presence or absence of IFN . CPT-Tert cells were treated with interferon , as described above , and infected at a MOI of 0 . 01 with wt and mutant viruses . At 24 , 48 and 72 h post infection the cell supernatants were collected and the virus titrated in susceptible cells . BTV-8ΔNS4 consistently reached lower titres ( approximately 10 to 25 fold ) than wt BTV-8 in cells treated with 1000 AVU/ml of either IFNT or UIFN ( Figure 7 ) . Similar to what was observed in the IFN protection assays , there was no discernable difference in the replication growth of BTV-1 and BTV-1ΔNS4 after treatment with either IFNT or UIFN . Similar patterns with both BTV-1 and BTV-8 wt and the ΔNS4 mutant viruses where observed when the input viruses were used at a MOI of 0 . 1 and 0 . 001 in CPT-Tert ( data not shown ) , or in BFAE cells treated with UIFN and infected at a MOI of 0 . 1 ( data not shown ) . We next ruled out that the mutations inserted in segment 9 of BTV-8ΔNS4 had a negative effect on VP6 expression ( the other protein expressed by segment 9 ) . As shown in Figure 8A , BTV-8 wt and BTV-8ΔNS4 express similar amounts of VP6 , reinforcing the notion that the biological differences observed between these two viruses were indeed due to the expression of NS4 . Therefore , the data presented so far suggested that either the BTV-1 NS4 was somewhat defective or that the influence of this protein on viral replication in the presence of IFN varies from strain to strain . In order to discern between these two possibilities , we derived BTV-8 reassortants containing either segment 9 of BTV-1 ( BTV-8/1S9 ) or the ΔNS4 version ( BTV-8/1S9ΔNS4 ) . Similarly , we derived BTV-1 reassortants containing the wild type or mutated segment 9 of BTV-8 ( BTV-1/8S9 and BTV-1/8S9ΔNS4 ) . In addition , we obtained a BTV-8 recombinant ( BTV-8/1NS4 ) with a single amino acid residue mutated in the NS4 ( S to N in position 6 ) in order to render this protein identical to the homologous BTV-1 protein ( Figure 8B ) . Both cytopathic protection assays and multistep growth assays clearly showed that BTV-8/1S9 replicated more efficiently than BTV-8/1S9ΔNS4 in the presence of IFN ( Figure 8C , D ) . Similar results were obtained with BTV-8/1NS4 , which replicated more efficiently than BTV-8ΔNS4 in cells pre-treated with IFN , while no major differences were observed between BTV-1/8S9 and BTV-1/8S9ΔNS4 . Collectively , these data strongly indicate that the NS4 of BTV-1 is not defective and can function within the context of BTV-8 . Next , we assessed the virulence of ΔNS4 BTV mutants in two murine models of bluetongue infection [46] , [47] . 129sv IFNAR ( −/− ) mice , which are deficient in the type I IFN receptor , are susceptible to infection and disease induced by BTV inoculated by various routes [46] , [48] . Newborn NIH-Swiss mice inoculated intracerebrally are also susceptible to BTV infection [47] . These models have been previously used to assess BTV virulence [47] , [49] . In this study , we infected 129sv IFNAR ( −/− ) mice with either BTV-1 , BTV-8 or the corresponding ΔNS4 mutants . No major differences were observed in the virulence of wild type and ΔNS4 viruses; all viruses employed in this study killed 100% of the inoculated mice by day 8 post-infection ( Figure 9 ) . We also inoculated 3-day old NIH-Swiss mice intracerebrally with the same viruses as above . Once again , both wild type and ΔNS4 viruses were able to kill 100% of the inoculated mice with no major differences in the virulence observed ( Figure 9 ) .
In this study we have shown that BTV expresses a previously uncharacterised non- structural protein that favours viral replication in cells in an antiviral state . By constructing deletion mutants by reverse genetics , we showed that NS4 is dispensable for viral replication in vitro , both in mammalian and insect cells , and in vivo in murine experimental models . However , the coding sequence in the NS4 reading frame of segment 9 is highly conserved in BTV and in related Orbiviruses [29] , [50] , suggesting that it must be essential for the maintenance of BTV in nature . Indeed , we have found that NS4 confers a replication advantage to BTV-8 in cells pre-treated with type I IFN . We found NS4 to have strong nucleolar localization , although it may shuttle between the nucleolus and cytoplasm and possibly carry out its biological functions in the latter . The nucleolus is a dynamic sub-nuclear structure that plays crucial roles in ribosome subunit biogenesis , the response to cellular stress and cell growth [51] , [52] . Several examples of viral proteins targeting the nucleolus have been discovered in recent years [53] . The retroviral Rev and Rev-like proteins for example , shuttle between the nucleolus and cytoplasm , and function as post-transcriptional regulators of viral gene expression [54]–[57] . One of the main functions of these proteins is to facilitate the export of unspliced viral mRNA ( transcribed from the proviral DNA copy of the retroviral genome stably integrated in the cell genome ) by simultaneously binding an RNA structure in the viral RNA and the karyopherin export factor Crm1 ( chromosome region maintenance 1 ) [58] . Other RNA viruses ( including those that replicate exclusively in the cytoplasm ) have also been found to possess proteins that target the nucleoli . Examples include , among others , avian infectious bronchitis virus [59] , porcine reproductive and respiratory syndrome virus [60] , Newcastle disease virus [61] , Semliki forest virus [62] , dengue virus [63] , West Nile virus [64] , influenza virus [65] , avian reovirus [66] and encephalomyocarditis virus [67] , [68] The reasons for the nucleolar targeting of many of these proteins have not always been entirely clear . The avian reovirus σA protein is a structural protein and is a major component of the inner capsid shell . Although the σA protein localises mainly in viral factories in the cytoplasm of infected cells , it also localizes in the nucleoli [66] . σA has a strong affinity for dsRNA and it may provide protection against the IFN-induced and dsRNA dependent PKR response . Interestingly , σA mutants that do not bind dsRNA are also unable to reach the nucleoli , suggesting that dsRNA binding and nucleolar targeting may be strictly linked [69] . BTV NS4 may also bind nucleic acids but , unlike the reovirus σA , we show strong evidence that NS4 is not a structural protein . Indeed , by western blotting we did not detect NS4 in viral particles but only in lysates of BTV infected cells . In addition , by confocal microscopy we did not detect NS4 in viral inclusion bodies but predominantly in the nucleoli of viral infected cells . We cannot exclude completely that small amounts of NS4 , below the limits of detection of our western blotting analysis , are present in viral particles . The predicted structural features of NS4 resemble those of a transcription factor of the bZip family with a basic domain followed by a leucine zipper motif [70] . Thus , NS4 may function as a nucleic acid binding protein and either repress or enhance transcription of genes linked directly or indirectly to the IFN response of the cell . However , a BTV-8 recombinant virus ( BTV-8ΔLZNS4 ) expressing an NS4 with all the 4 leucine residues forming the putative leucine zipper mutated ( into either glutamine or serine ) replicated as efficiently as BTV-8 wt in cells pre-treated with IFN ( data not shown ) . Thus , more studies will be necessary to explore this possibility . The organization of VP6/NS4 ORFs in segment 9 of BTV mirrors that of NSP5/NSP6 in the rotavirus segment 11 [71] . The rotavirus NSP6 is not essential for virus replication but unlike the BTV NS4 , does not localize in the nucleus of infected cells [72] . To date , limited information is available on the interplay between BTV and the host innate immune system . BTV has been recognized as a potent inducer of type I IFN in sheep [42] , cattle [43] and mice [73] . However , limited data have been available on how BTV induces the IFN response of the cell and , more importantly , what counteracting measures the virus utilises to overcome this response . Our data suggest that BTV may use NS4 to defend itself from the innate immune response of the host given that replication of BTV8-ΔNS4 in cells treated with IFN is 10 to 25 fold less efficient compared to wild type BTV-8 . In addition , the cytopathic effect in cells treated with IFN is more pronounced when cells are infected by wild type BTV-8 compared to cells infected by BTV8-ΔNS4 . Viruses have evolved a variety of strategies to evade the host innate immunity [74] . Other dsRNA viruses such as rotaviruses , use different mechanisms ( which vary between strains and the type of infected cells ) to modulate the type I IFN response . For example , rotaviruses use NSP1 protein to promote the proteasome-dependent degradation of IRF proteins [75]–[77] and mediate repression of NF-kB , resulting in a reduction of IFN induction [78] . Rotaviruses also induce shut off of cellular protein synthesis resulting from the detection of dsRNA by PKR which , in turn is responsible for phosphorylation and consequent inhibition of the eukaryotic translation initiation factor eIF2α [79] . The blocking of host cell protein synthesis is another likely strategy used by some RNA viruses to counteract the IFN response [67] , [68] . BTV also blocks host cell protein synthesis early after infection , although the mechanisms underlying this phenomenon are not clear [27] . Interestingly , we found that BTV1-ΔNS4 replicated as efficiently as wild type BTV-1 , even in cells treated with IFN . However , the NS4 of BTV-1 appears to possess the same biological properties of the NS4 of BTV-8 . Indeed , a BTV-8 reassortant containing the entire segment 9 of BTV-1 ( BTV-8/1S9 ) or a recombinant BTV-8 expressing an NS4 100% identical to the homologous BTV-1 protein ( BTV-8/1NS4 ) , maintained the phenotype of wt BTV-8 . Thus , it is possible that the role played by NS4 in counteracting the IFN response of the host could vary between different virus strains . It is important to stress that the strain of BTV-8 that we used in this study has been passaged only a few times in culture ( once in KC cells and three times in BHK21 cells ) after isolation from blood of an infected animal . On the other hand , BTV-1 was derived from the “reference” South African strain passaged twice in embryonated eggs and 9 times in BHK21 . BTV-1 appears to grow slightly faster than BTV-8 in culture , especially at the early time points post infection . Thus , faster replication may help BTV1-ΔNS4 to escape the IFN response of the cell more efficiently , as already suggested for some strains of influenza , and this may render NS4 less critical in these in vitro assays [80] . More in vivo experiments will be needed in order to determine the role of NS4 in the interplay with the natural host of BTV infection . We observed no differences between wild type BTV-8 and BTV8-ΔNS4 in experimental mouse models , although it remains possible that differences could be identified in sheep . It is possible that NS4 is required for viral replication in insects , although we have established in this study that no differences are observed on the replication of the ΔNS4 mutants in insect cells in vitro . In conclusion , in the present study we have identified a previously uncharacterized non-structural protein of BTV . The identification of this highly conserved protein opens the way to understand finer details of virus-host interaction and pathogenesis . In addition , the distinct nucleolar localization , in a virus that replicates exclusively in the cytoplasm will offer new avenues to understand the various roles played by these organelles in the biology of the cell . | Bluetongue is a major infectious disease of ruminants caused by bluetongue virus ( BTV ) , an “arbovirus” transmitted from infected to susceptible hosts by biting midges . Historically , bluetongue has been endemic almost exclusively in temperate and tropical areas of the world . However , in the last decade BTV has spread extensively in several geographical areas causing a serious burden to both animal health and the economy . BTV possesses a double-stranded RNA segmented genome . For over two decades , it has been widely accepted that the 10 segments of BTV genome encode for 7 structural and 3 non-structural proteins . In this study we discovered that BTV expresses a previously uncharacterized non-structural protein that we designated NS4 . Although BTV replicates exclusively in the cytoplasm , we found NS4 to localize in the nucleoli of the infected cells . Our study shows that NS4 is not needed for viral replication both in mammalian and insect cells , and in mice . However , NS4 confers a replication advantage to BTV in cells in an antiviral state induced by interferon . In conclusion , we have elucidated a possible route by which BTV can counteract the defences of the host . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [
"veterinary",
"diseases",
"virology",
"biology",
"microbiology",
"veterinary",
"science"
] | 2011 | Identification and Characterization of a Novel Non-Structural Protein of Bluetongue Virus |
Non-coding RNA ( ncRNA ) play an important and varied role in cellular function . A significant amount of research has been devoted to computational prediction of these genes from genomic sequence , but the ability to do so has remained elusive due to a lack of apparent genomic features . In this work , thermodynamic stability of ncRNA structural elements , as summarized in a Z-score , is used to predict ncRNA in the yeast Saccharomyces cerevisiae . This analysis was coupled with comparative genomics to search for ncRNA genes on chromosome six of S . cerevisiae and S . bayanus . Sets of positive and negative control genes were evaluated to determine the efficacy of thermodynamic stability for discriminating ncRNA from background sequence . The effect of window sizes and step sizes on the sensitivity of ncRNA identification was also explored . Non-coding RNA gene candidates , common to both S . cerevisiae and S . bayanus , were verified using northern blot analysis , rapid amplification of cDNA ends ( RACE ) , and publicly available cDNA library data . Four ncRNA transcripts are well supported by experimental data ( RUF10 , RUF11 , RUF12 , RUF13 ) , while one additional putative ncRNA transcript is well supported but the data are not entirely conclusive . Six candidates appear to be structural elements in 5′ or 3′ untranslated regions of annotated protein-coding genes . This work shows that thermodynamic stability , coupled with comparative genomics , can be used to predict ncRNA with significant structural elements .
Non-coding RNA ( ncRNA ) are functional RNA transcripts that are not translated into protein ( i . e . , not messenger RNAs ) . Research , particularly over the last 10 years , has shown that they perform a wide range of functions in the cell [1]–[4] . Despite the growing body of knowledge about ncRNA , it is likely that many ncRNA remain undiscovered . Data from high-throughput experimental methods show that much of the intergenic DNA in eukaryotic genomes is transcribed and may be ncRNA [5]–[10] . Even in Saccharomyces cerevisiae , one of the most thoroughly studied model organisms , there is evidence that only a fraction of the ncRNA is known . Tiling arrays , large-scale cDNA libraries , and serial analysis of gene expression ( SAGE ) experiments have all shown transcription from many locations in the genome that appear to be unannotated ncRNA genes [11]–[14] . This along with recent identification of new protein coding genes such as YPR010C-A in 2006 shows that even in this best-studied Eukaryote , we still do not know the complete gene set [13] . Computational methods for accurate ncRNA gene prediction remain elusive . The development of such methods are crucial for identifying ncRNA that are difficult to detect experimentally such as those expressed at low levels or under unusual conditions . They are also needed to reduce the time and expense required to perform experimental methods , particularly when considering the large number of species of interest . The challenge of predicting ncRNA genes rests with the fact that they lack common primary sequence features and demonstrate poor cross-species sequence conservation [15] , [16] . They do not have start codons , stop codons or open reading frames which serve as key signposts for protein-coding genes and cannot be located using simple sequence searches . Some success with ncRNA gene prediction has been achieved by focusing on specific sub-classes of ncRNA that share common features . Examples include tRNAs , tmRNAs , snoRNAs ( C/D box and H/ACA box ) , and miRNAs [17]–[32] . In S . cerevisiae , computational screens for C/D box [19] and H/ACA box snoRNAs [20] have identified several new snoRNA genes . Additional ncRNA screens in S . cerevisiae have included searches for polymerase III promoters , searches in larger than average intergenic regions [33] and searches for ncRNA structural features using the QRNA program . The QRNA program was used to search pair-wise alignments for patterns of compensatory mutations consistent with base-paired secondary structure [34] . These regions were then tested experimentally to determine if they expressed a transcript likely to be ncRNA . Together , these three methods resulted in identification of 6 novel ncRNA that were supported by experimental evidence ( RNA170 , snR161 , snR82 , snR83 , snR84 , RUF5-1/2 ) . In another study , the S . cerevisiae genome was analyzed using the RNAZ program [35] . This program is based on the same principals as the QRNA program and uses multiple , cross-species sequence alignments to search for patterns of compensatory changes suggestive of secondary structure . RNAZ also includes thermodynamic analysis . A total of 572 candidate regions were identified as potentially containing unannotated ncRNA candidates using the RNAZ program [35] , [36] . Publicly available data sets were used to provide general support for these predictions but no detailed experimental analysis was performed on individual predictions . In this work ncRNA genes are predicted in S . cerevisiae based solely on the thermodynamic stability of ncRNA structures as proposal by Maizel in the late 1980's [37]–[39] . Maizel theorized that structural ncRNA are thermodynamically more stable than random sequences . An influential paper by Rivas & Eddy entitled “Secondary structure alone is generally not statistically significant for the detection of noncoding RNAs” suggested that Maizel's approach was generally not effective for structural ncRNA discovery [40] . Based on this conclusion , many investigators turned away from thermodynamic based approaches for ncRNA discovery to methods based on compensatory changes in cross-species alignments[31] . However , a growing body of evidence has been accumulating suggesting that thermodynamic stability is a discriminating feature of many classes of structural ncRNA [41]–[43] . In this work , we build on this result to not only evaluate the thermodynamic stability of known structural ncRNA but also to use it for structural ncRNA discovery . The work presented here demonstrates the value of thermodynamic structural stability , as summarized in a Z-score , for discovery of structural ncRNA . It also explores the impact of window size and step size on the sensitivity of ncRNA identification . Sets of positive and negative control genes were evaluated to determine the effectiveness of the approach . This approach was then applied to predict ncRNA genes on chromosome six of S . cerevisiae . The analysis was repeated independently in S . bayanus and the gene predictions common to both genomes comprised the final set of gene predictions . Experimental validation of these predictions show that four ncRNA transcripts are well supported by northern blot analysis , rapid amplification of cDNA ends ( RACE ) , and publicly available cDNA data . One additional ncRNA candidate is also supported by experimental data but the data is not entirely conclusive . Six of the predicted candidates appear to be structural elements in 5′ or 3′ untranslated regions ( UTRs ) of annotated protein-coding genes .
The thermodynamic stability of potential ncRNA candidates was evaluated using a Z-score based on the minimum folding energy ( MFE ) determined by RNAfold [44] . The Z-score represents the number of standard deviations that the MFE of a native sequence , x , deviates from the mean MFE of a set of shuffled sequences of x ( see Materials and Methods ) . A key variable in calculating the Z-score for ncRNA discovery ( as opposed to evaluating known structural ncRNA ) is the length of the sequence to be evaluated . As ncRNA vary in length and structure , no single window size is expected to be optimal for ncRNA gene identification . Short structural elements will probably only be detected with relatively short window sizes while longer structural elements will probably only be detected with relatively longer window sizes . To identify the window sizes most appropriate for ncRNA discovery , values ranging from 20 nt to 200 nt were investigated and incremented in steps of 5 nt ( window delta ) . A scanning approach was used to computationally search for potential structural elements within a test sequence . A starting minimum window size was selected and this window was used to scan the test sequence starting at the beginning of the sequence and moving each time by the amount of the step size ( our analysis used a step size of 5 nt ) . A Z-score was calculated for each window position . Once the entire test sequence was evaluated using this fixed window length , a new window length was selected by increasing window length by the amount of the window delta ( our analysis used a window delta of 5 nt ) . The test sequence was evaluated in the same manner using the new window size . This process was repeated until all window sizes had been evaluated . Since the same test sequence was evaluated using multiple window sizes , it was necessary to determine the impact of multiple hypothesis testing . In lieu of a Bonferroni correction , negative control sets were evaluated using the same number of window sizes and step sizes . Any windows producing a “significant” Z-score during the scanning process were considered candidate regions for structural ncRNA . The Z-score cutoff considered to be “significant” was determined by evaluating positive and negative test sets . It was sometimes the case that multiple , overlapping windows , of several lengths , produced “significant” Z-scores . In such cases , the region encompassed by all the overlapping windows constituted the candidate region . Once candidate regions were identified , primers were designed within these regions to determine whether they produced a transcript and to identify the transcript boundaries . The primers were designed as close as possible to the middle of the candidate regions . The exact position of the primer was dictated by the need to satisfy the fairly stringent requirements of the rapid amplification of cDNA ends ( RACE ) procedure ( See Materials and Methods ) . Positive and negative control sets were compiled to test if the Z-score could be used to distinguish known ncRNA from non-functional sequences as suggested by previous investigators [41]–[43] . The positive control set was drawn from the list of annotated ncRNA in the Saccharomyces Genome Database ( SGD ) [45] ( Table 1 ) . The tRNA and rRNA genes were not included in the positive control set as they can be identified with great accuracy using existing tools [17] and because tRNA are known to produce poor thermodynamic footprints [40] , [41] , [46] . The positive control set consisted of four snoRNA genes and all of the remaining known ncRNA ( Table 2 ) . Three negative control sets were created to cover the full range of negative control cases . The first negative control set consisted of 20 randomly generated sequences of 300 nt in length . This set was used because it was known not to contain any unannotated genes . The shortcoming of this control set is that it likely fails to capture the nuances of nucleotide distributions in S . cerevisiae . The randomly generated sequences had a GC content of ∼40% , ranging from 35 . 0% to 49 . 3% , reflecting the GC content of S . cerevisiae . A second negative control set was created by randomly shuffling the positive control set . Each sequence was shuffled preserving sequence length as well as its mono- and di-nucleotide composition using the “squid” utilities [47] . The third negative control set was generated by selecting six intergenic regions from the S . cerevisiae genome . Intergenic regions were chosen as a control instead of coding regions because the GC content in the S . cerevisiae genome differs between protein coding regions and non-protein coding regions . Since the ultimate goal was to search for ncRNA in intergenic regions , it was best to select a test set representative of these regions . The untranslated regions ( UTR ) of most genes in S . cerevisiae are not mapped so the actual intergenic regions are generally unknown . In order to minimize the possibility of choosing a region that contained an unannotated structural element , six intergenic regions were chosen that are flanked on one side by a gene with a known , short ( <40 nt ) 5′ UTR , unlikely to form a structure . A window of 300 nts from the 5′ end of the open reading frame ( ORF ) of each of these genes was used as a negative control test sequence ( Table S1 ) . Z-score values calculated for the 20 randomly generated negative control sequences revealed that large negative Z-scores are often generated when using window sizes of less than 65 nt . With these short window sizes , many shuffled sequences have a calculated minimum folding energy of zero or close to zero and the Z-score distribution of the shuffled sequences is narrow . This produces a small value for the standard deviation . If the MFE of the original , unshuffled sequence is even slightly above zero , it will be many standard deviations from the distribution mean and produce a large negative Z-score . When examining window sizes of 75 nt or greater , two ( Random9 and Random13 ) of the 20 randomly generated sequences produced a Z-score less than −3 . 5 ( Table S2 , Figures S1 and S2 ) . The total length of sequence producing a Z-score ≤−3 . 5 was 295 nt and represented 5 . 0% of the nucleotides in the entire randomly shuffled test set ( Table 3 ) . Z-score values calculated for the 6 intergenic sequences of the second negative control set produced a pattern very similar to that of the randomly generated sequences . For window sizes less than about 65 nt , large negative Z-scores were generated . Window sizes longer than 75 nt did not produce any Z-scores less than −3 . 5 with the exception of the intergenic sequence between genes PTP1 and SSB1 . The first 190 nt of this sequence produced Z-scores as low as −4 . 7 for various window sizes ( Table S2 ) . This may represent either a false positive or may suggest the presence of a structural feature ( ncRNA or long PTP1 5′ UTR structure ) . This 190 nt region represents approximately 10 . 5% of the total length of the intergenic negative control set . The final negative control set consisted of shuffled sequences of the positive control set ( Table 2 ) . Of these , portions of 5 out of 16 sequences ( 31% ) produced Z-scores less than −3 . 5 ( Table S2 and Figures S3 and S4 ) . The total sequence length included in these regions represented 8 . 1% of the total negative control set length . All of the sequences in the positive control set produced Z-scores less than −3 . 5 for multiple window sizes ( Table S3 , Figures S5 and S6 ) with the exception of three genes . These genes were snR76 , RNA170 , and SRG1 . The snR76 gene is a C/D box snoRNA and it is questionable whether structure plays a significant role in the function of this gene . The SnoScan program was written explicitly to predict C/D box snoRNA and has been used successfully to predict these genes in both D . melanogaster and S . cerevisiae [19] , [48] . Known C/D box snoRNA were used to identify features shared among this family of ncRNA . Only one of the six criteria identified is related to structure ( terminal stem base pairings ) . This base pairing consists of only 4–8 bps and is not always present [19] . This is in stark contrast to the snoGPS program used to identify H/ACA snoRNA [20] . The snoGPS program was trained using known H/ACA snoRNA examples and includes secondary structure as a key element in H/ACA box snoRNA detection . Results from these snoRNA gene identification efforts strongly suggest that structure is generally not a significant component of C/D box snoRNA genes . SRG1 is a ncRNA gene that has been shown to repress the expression of its neighboring gene SER3 [49] . Transcription of SRG1 interferes with the binding of SER3 activators in its promoter . This mechanism suggests that SRG1 fulfills its role as a transcriptional repressor through its transcription rather than through a significant structural component . The RNA170 gene was discovered through a genome-wide search of Polymerase III box A and B consensus sequences [33] . Its function and mechanism of action are unknown . It seems likely that this ncRNA does not require a significant structural component to perform its function . The total sequence length encompassed by a Z-score less than −3 . 5 in the positive control set represented 41% of the total sequence evaluated . If snR76 , SER3 and RNA170 are removed from the set , 46% of the positive control set produces a Z-score <−3 . 5 ( Table 3 ) . Window sizes of 75 nt to 85 nt were crucial for identifying the short ncRNA such as snR6 . To summarize , three negative control sets were used consisting of a set of randomly generated sequences , a set of intergenic sequences , and a set of shuffled positive controls . The percent of sequence producing a false positive indication ( i . e . , Z-score ≤−3 . 5 ) for each of these sets was 5 . 0% , 10 . 5% , and 8 . 1% , respectively ( Table 3 ) . We examined the regions producing Z-scores ≤−3 . 5 for unusual GC content that might explain the large negative Z-score but found nothing significant in these regions ( Table S4 ) . For the positive control set , 13 of the 16 genes produced a Z-score ≤−3 . 5 , encompassing 41% of the total sequence length of the set ( Table 3 ) . There is good reason to think that the three genes in this set failing to produce a Z-score ≤−3 . 5 do not contain structural features . Analysis of the positive and negative control sets provided the following conclusions , ( 1 ) Evaluating window sizes less than 65 nt produces many false positives , ( 2 ) A Z-score value of −3 . 5 is useful for discriminating known ncRNA from non-functional sequence , ( 3 ) The percent of false positive sequence was observed to be ∼5 . 0–10 . 5% when using a cut-off Z-score value of −3 . 5 . Evaluation of the positive and negative control sets showed that the Z-score was useful for discriminating known structural ncRNA from non-functional sequence . To apply the approach to de novo gene prediction it is necessary to scan through a large test sequence ( i . e . , a chromosome ) in search of regions that produce Z-score values indicative of structural ncRNA . To test the effectiveness of our approach for ncRNA discovery , and to determine the optimal parameters for the search , we performed two tests . We evaluated our ability to detect known ncRNA ( Table 1 ) , then we performed a detailed analysis of optimal search parameters using a small subset of ncRNA . First , each annotated , nuclear encoded ncRNA ( excluding rRNA ) , along with 200 nt upstream and downstream of the gene , was used as a test sequence . Z-scores were calculated on the ncRNA strand using the following parameters: window sizes = 75 to 200 nt , step size = 5 nt , window delta = 5 nt . The known ncRNA were considered detected if the center of the window ( s ) producing a Z-score ≤−3 . 5 overlapped the gene . 100% of the snRNA were detected , 72 . 4% of the H/ACA box snoRNA were detected , and 23 . 9% of the C/D box snoRNA genes were detected . Only 16% of the tRNA genes were detected . This result is consistent with previous reports of poor detection of tRNA based on a Z-score-type search criteria [40] , [41] , [43] . Clote et al [41] suggested that this may , in part , be due to the extensive post-transcriptional modifications that occur to tRNA that are not accounted for in the MFE calculation based on unmodified sequence . The percent of tRNA detected was a function of the tRNA length . 10 . 4% of the tRNA shorter than 75 nt ( 192 total ) were detected while 34 . 6% of tRNA greater than 75 nt ( 83 total ) were detected . This ncRNA data can also be used to show the impact of using a single window size or a large step size on ncRNA detection ( Table 4 ) . The table provides the percent of H/ACA box snoRNAs detected when only a single window size was used to perform the analysis . The impact of using different step sizes ( 5 nt , 25 nt and 50 nt ) is also presented . Using a single window size , as opposed to several sizes , reduces the number of snoRNA detected . The number of H/ACA snoRNA detected by evaluating all window sizes from 75 nt to 200 nt was 72 . 4% , which is greater than the number detected by using any single window size . The number of H/ACA snoRNA detected for a given window size decreases as the step size increases . These results can provide guidance for choosing a subset of window sizes to perform a ncRNA screen . Tradeoffs can be made between the percent of ncRNA detected and the computational investment required to perform the analysis . A second experiment was performed to further explore the question of optimal values for step size and window delta . Ten tRNA from the Rfam database [50] were embedded at random locations within 300 nt background sequences ( Table S5 ) . The selected tRNA ranged in length from 68 nt to 91 nt and generated large negative Z-scores ( <−4 . 0 ) when evaluated in isolation . The background sequences used were mRNA transcripts that had no significant Z-score along their length . A Z-score was calculated at each position along the total sequence ( step size = 1 ) for each window sizes from 60 to 95 nt ( window delta = 1 ) . In most cases it was possible to detect the tRNA in the embedded sequences using a step size of 5 and a window delta of 5 ( Figure 1 ) . However , in some cases the window size and window delta needed to be smaller than this to be certain of finding the transcript ( Figure 2 ) . Based on the above results , we chose to use a step size of 5 nt and a window delta of 5 nt for the remainder of our analysis . This provided a high probability of detecting most ncRNA while keeping computational time manageable . The ncRNA prediction method was applied to intergenic regions of S . cerevisiae chromosome VI using window sizes from 75 to 200 nt , a window delta size of 5 nt , and a step size of 5 nt . The UTRs of most genes in the S . cerevisiae genome are unknown so the term intergenic used here refers to the distance between ORFs of adjacent annotated genes . Genes classified as dubious in SGD [45] were ignored . The UTRs of the flanking genes are thus included in the intergenic region , and those containing structure [51] may be detected . The limited data available on S . cerevisiae 5′ and 3′ UTRs shows that most UTRs are short ( 3′ UTR median length 91 nt , 5′ UTR median length 68 nt ) [12] , [13] , suggesting that most of the structural signals detected should come from independent ncRNA rather than UTRs . Only intergenic regions greater than 90 nt in length were evaluated . Forward and reverse DNA strands were evaluated independently since the GU pairing in ncRNA confers different folding potential to the complementary strands . In an attempt to reduce the rate of false positives produced by the screen , the analysis was repeated in syntenic regions of S . bayanus ( MCYC623 ) [52] . For a region to be considered syntenic , it had to have the same flanking genes with the same orientation in both S . bayanus and S . cerevisiae . A total of 66 syntenic regions satisfying these criteria were identified . The percent identity between these regions in S . cerevisiae and S . bayanus varied between 18 . 0% and 76 . 5% with an average of 57 . 0% ( Table S6 ) . Predicted structural elements common to both species were taken as ncRNA candidates . There were no constraints placed on the relative position of the structural predictions in syntenic regions , only that they appeared between the same two flanking genes in both species . There were 23 intergenic regions in S . cerevisiae that produced Z-scores ≤−3 . 5 and 24 intergenic regions in S . bayanus that produced Z-scores ≤−3 . 5 . Fourteen of these regions were common to both S . cerevisiae and S . bayanus and resulted in a total of 16 high priority candidates ( two syntenic regions produced two separate candidates ) ( Table 5 ) . In many cases , a Z-score below the cutoff criterion was generated from both the Watson and Crick strand . For this reason , experimental testing was performed on both strands independently for all candidates . An example of the Z-score values generated by evaluating the Watson strand for each position in the intergenic region between SEC4 and VTC2 for all window sizes is provided in Table S7 . The position of windows producing Z-scores ≤−3 . 5 within selected intergenic regions are given in Figures S7 , S8 , S9 , and S10 . Northern blots and rapid amplification of cDNA ends ( RACE ) were used to test the validity of the ncRNA candidates . Since the environmental conditions required for expression of the ncRNA gene candidates were unknown , nine conditions were tested . Conditions were selected that have been shown to generate high overall transcript expression [53] , [54] . These nine conditions were: heat shock ( 25°C to 37°C ) , diamide treatment , growth in minimal media , saturated growth in minimal media , anaerobic growth , sporulation , schmooing , YPGlycerol ( non-fermentable carbon source ) , and YPD growth . RNA was isolated and northern blotting was performed ( see Materials and Methods ) . Strand specific blotting protocol was used for the northern blot analysis to identify the transcribed strand and to help rule out DNA contamination . Northern blotting confirmed expression of transcripts between SEC4 and VTC2 ( RUF20 ) on the Crick strand and between YFL051C and ALR2 on the Watson strand ( Figure S11 ) . The ACT1-YPT1 transcript showed strong expression on the Crick strand under all conditions but later proved to be part of the ACT1 5′ UTR ( data not shown ) . Rapid amplification of cDNA ends ( RACE ) was used to measure the 5′ or 3′ end of flanking genes as well as map candidate gene ends ( Table 5 , Table S8 , Table S9 ) . The cDNA was generated using a poly-T primer from RNA collected from anaerobic or heat shock conditions ( see Materials and Methods ) . The RACE analysis proved considerably more sensitive than northern blotting . In addition to this experimental data , several publicly available data sets were evaluated for their value in substantiating these ncRNA predictions . Tiling array data [11] , [12] has been used by several investigators to substantiate computational ncRNA predictions . However , we found this data quite noisy and difficult to interpret with a high degree of confidence . It also remains a point of debate whether all of the transcription measured by microarray tiling experiments represents true functional transcripts or whether some of it represents spurious transcription or experimental artifact [3] , [4] , [9] , [55]–[58] . The sequenced cDNA library data appears to be more useful in verification of ncRNA predictions [13] . The data included information on transcript ends and as such was likely to derive from a functional transcript . A summary of all the experimental data is provided in Table 6 . The candidates in Table 6 are listed in order of increasing experimental support . The top four ncRNA candidates have been assigned names RUF20 ( RNA of unknown function ) to RUF23 ( Figure 3 ) . The RUF name was chosen to follow the naming convention established by previous investigators [34] . These transcripts do not appear to be snoRNA or to encode an ORF ( see Materials and Methods ) . One of the candidates , RUF22 , overlaps with an autonomously replicating sequence , ARS607 . One other ncRNA candidate , IES1-YFL012W , partially overlaps ( 120 bp ) with the dubious ORF YFL012W-A which is on the opposite strand ( Watson ) . This dubious gene also partially overlaps ( 120 bp ) the IES1 gene . According to SGD , this dubious ORF is unlikely to encode a protein based on available experimental and comparative sequence data [45] . It is reasonable to question whether our computational screen provided an improved ability to identify ncRNA relative to simple random experimental searches . Previous investigators have shown that randomly probing intergenic regions of the S . cerevisiae is unlikely to reveal ncRNA . In the work by McCutcheon & Eddy , 20 intergenic regions were chosen randomly and probed by northern blot [34] . None of these regions produced a transcript . Olivas , Muhlrand and Parker also provided evidence that probing intergenic regions is unlikely to produce a transcript even though they were conducting a directed search for ncRNA [33] . They performed two different screens in an effort to discover ncRNA . In one case , they used a computational approach to identify 10 locations in the genome that contained potential RNA polymerase III binding motifs . When they probed the 10 regions , only one was found to express a transcript . In their second screen , they identified regions within the genome with large gaps between genes . They expected these regions to contain ncRNA transcripts because the high density of genes in the Saccharomyces genome suggested that any large gaps were likely to be occupied by unannotated genes . Probing 59 such regions revealed 15 potential transcripts . It is clear that even probing regions expected to contain ncRNA transcripts is often unsuccessful . Our experimental screen of 16 candidates produced 4 ncRNAs with strong support , 2 potential ncRNA with weaker support , and 6 UTRs likely to contain structure ( Table 6 ) . Thus , it appears that our computational method improves ncRNA identification over simple random searches . To further validate the SEC4-VTC2 ncRNA candidate , RACE was performed in syntenic regions of S . bayanus and the more distantly related hemiascomycete species Ashbya gossypii . This species diverged from S . cerevisiae prior to the S . cerevisiae whole genome duplication . However , A . gossypii still retains many syntenic regions with S . cerevisiae and , in the case of the SEC4-VTC2 gene candidate , gene order and orientation are preserved . RACE products were obtained from both S . bayanus and A . gossypii ( Figure 4 ) . The fact that the transcript is preserved over such a large evolutionary distance provides strong evidence that this is a bona fide ncRNA gene .
A computational screen for structural ncRNA in S . cerevisiae was performed using thermodynamic stability to discriminate structural ncRNA from background sequence . The method was tested on positive and negative control sets to determine its effectiveness for identifying known ncRNA and to develop optimal search parameters . These parameters were determined to be a Z-score <−3 . 5 , window sizes 75 nt to 200 nt , step size of 5 nt , and window delta of 5 nt . The parameters were then used to screen for novel ncRNA in the intergenic regions of S . cerevisiae chromosome VI . To reduce the number of false positive predictions , an independent analysis was performed on syntenic regions of S . bayanus . The set of predictions found in common in both species were subjected to further experimental verification . Like all computational ncRNA gene discovery approaches currently available , our method can only provide guidance on regions likely to contain structural elements . It cannot predict the exact location of the ncRNA gene or its precise ends . These must be determined experimentally . Northern blots , rapid amplification of cDNA ends ( RACE ) , and publicly available cDNA library data were used to test the predictions . Each of these methods was selected for specific reasons . The strength of northern blot analysis is that it does not rely on transcript amplification and hence avoids artifacts that can result from an amplification step . However , it is not as sensitive as other methods and this can be a significant limitation when testing for ncRNA that may be expressed at low levels . RACE provides greater sensitivity than northern blot analysis but may be subject to amplification artifacts . The potential for artifacts is reduced because the 5′ and 3′ ends of the transcript are captured . The presence of a cap and poly-A tail provides strong evidence that the transcript has been processed by the cellular machinery and is a legitimate functional transcript . This makes the approach superior to methods such as tiling arrays that provide information on transcription but for which it is difficult to distinguish transcriptional noise from genuine transcripts . The publicly available cDNA data used here also has the advantage of capturing the transcript 5′ and 3′ ends , providing strong evidence for a legitimate , processed transcript . The initial computational screen presented here produced sixteen ncRNA gene candidates on chromosome VI of S . cerevisiae . Four candidates are well supported by experimental data and have been given the names RUF20 to RUF23 ( Table 5 ) . The RUF20 candidate is also expressed in S . bayanus and in the more distantly related species A . gossypii ( Figure 4 ) . All of the transcripts were evaluated for the possibility that they might be snoRNA or encode a protein but this was shown to be unlikely ( see Materials and Methods ) . Two additional candidates are also supported by experimental evidence but further experimental testing is needed to confirm their legitimacy . Six of the candidates were found to be part of the 5′ or 3′ untranslated regions ( UTRs ) of annotated protein-coding genes . These structures are interesting because they may play a functional role in the UTRs of these genes ( Table 5 ) . Additional experimental analysis will be needed to determine the function of the structures as well as the function of the four new ncRNA , RUF20 to RUF23 . There are several possible explanations why experimental data could not be obtained to support three of the ncRNA predictions . These predictions may represent false positives , they may not be expressed under the conditions tested , or they may be expressed at such a low level that they could not be detected . It has been shown that transcript abundance in yeast varies over six orders of magnitude and that some important transcription factors are expressed at levels as low as one transcript per thousand cells [59] . It is also possible that these transcripts are not transcribed by RNA polymerase II , the method used in this study to generate cDNA is dependent on a poly-A tail in the RNA transcript . If the ncRNA candidates are transcribed by polymerase I or III , they would likely not be captured in the cDNA library . It should be noted that there were three genes in the positive control set ( Table 2 ) that did not generate a Z-score <−3 . 5 ( snR76 , SER3 , RNA170 ) . It is questionable whether these genes actually contain significant structural elements . One of them , snR76 , is a C/D box snoRNA and data from other investigators [19] shows that structural features are only present in a subset of these genes . It is not surprising that this category of ncRNA was not easily detected in this screen based on structural thermodynamic stability . It is clear that some classes of ncRNA will not be identified very well in structural screens . The other two genes in the positive control set were RNA170 ( unknown function ) and SER3 . The SER3 gene suppresses expression of its neighboring gene , SRG1 , by blocking access to the SRG1 promoter region via its transcription . SER3 and RNA170 are unlikely to contain significant structural features so the fact that they did not generate Z-scores less than −3 . 5 tends to validate the method . Two previous investigators have performed computational genome-wide screens for ncRNA in S . cerevisiae . McCutchen and Eddy , 2003 used the QRNA program to search for structural elements based on observed compensatory changes in pair-wise alignments of S . cerevisiae species . A fixed window size of 150 nts and a step size of 50 nt were used to perform the analysis . Two structural ncRNA candidates were found on chromosome VI . One prediction , between RIM15 and HAC1 ( 74738–74738 ) , was near one of the candidates predicted in this study between the same genes ( 74926–75006 ) . They were unable to obtain sufficient experimental support for expression of this transcript . This is consistent with our experimental results as well . The second McCutchen and Eddy prediction , between SMC1 and BLM10 , did not correspond to any predictions generated in this study . They obtained northern blot and RACE data to support expression of this second predicted gene . A second screen for ncRNA was performed by Steigele et al using the RNAZ program [35] . This program searches for compensatory changes in multiple sequence alignments as well as for thermodynamic stability cues indicative of structural elements . The relative contribution of these two factors in the prediction is not specified . A fixed window size of 120 nt and step size of 40 nt was used to perform the analysis . They reported a sensitivity ( true positives/total ) for identifying snoRNA of 47% ( pooling H/ACA box and C/D box snoRNA ) , sensitivity for identifying snRNA of 66% , and a sensitivity of 72% for tRNA . The screen generated a total of 18 novel intergenic structural predictions on chromosome VI . Of these , 8 were predicted to be on the Crick strand and 8 on the Watson strand . Five of these intergenic regions were shared by our predictions ( YFL051C-ALR2 , ACT1-YPT1 , TUB2-RPO41 , GYP8-STE2 and YFR017C-YFR018C ) . All 5 of the Steigele et al predictions were on the Watson strand in these regions . Two of the predictions overlapped with our predictions ( ACT1-YPT1 and YFR017C-YFR018C ) . Our experimental data suggested that the YFL051C-ALR2 region is transcriptionally complex and is likely to produce more than a single transcript . This could account for the fact that both studies predicted structural elements in this region . Our RACE analysis of the ACT1-YPT1 region showed that the predicted structural element was contained within the ACT1 UTR on the Crick strand . The Steigele et al prediction overlaps within the ACT1 UTR but is predicted to be on the opposite strand ( Watson ) . For the TUB2-RPO41 region , we experimentally confirmed a transcript on the Crick strand encompassing our predictions . This transcript overlaps with the Steigele et al prediction but is again on the opposite strand ( Watson ) . Our GYP8-STE2 prediction proved to be part of the GYP8 5′ UTR on the Crick strand . The Steigele et al prediction in this region was on the Watson strand and is beyond the region we measured for the GYP8 UTR ( although we were unable to map the end of this 5′ UTR ) . In the YFR017C-YFR018C region , we obtained RACE results that mapped our prediction to the Crick strand as part of the YFR018C 3′ UTR . The Steigele et al prediction , which largely overlaps our prediction , was for a gene on the Watson strand . Hence , while our predictions and those of Steigele et al are close to one another or overlapping in five regions , in all five cases they are on opposite strands . It is interesting that there is no overlap between the QRNA and the RNAZ predictions of chromosome VI since both programs consider compensatory changes within alignments to identify structural elements . The reason for this is unclear . There are two primary differences between the search for ncRNA presented here and the work of previous investigators . First , this method does not require sequence alignments in the analysis . Instead , it relies entirely on thermodynamic stability in unaligned syntenic regions of related species to predict ncRNA structure . The approach is capable of finding ncRNA that have moved out of register within syntenic regions and can be applied in situations where accurate alignments may be difficult to obtain . The second difference in this work is its examination of the impact of various window sizes and step sizes on ncRNA detection . The analysis shows that small step sizes are necessary to ensure that most ncRNA are identified . It also shows that more than one window size is needed when screening for ncRNA . Some ncRNA are detected only when using short window sizes while others are detected when using only long window sizes ( Table 7 ) . Limiting the search to a single window size , as has traditionally been done , is likely to bias the screen toward a subset of ncRNA for which that window size is optimal . The need for multiple window sizes and step sizes in the screening algorithm increases the computational investment necessary to perform the analysis . However , with the rapid increase in computer performance and the availability of computer clusters , these computations are not unreasonable . The increased computational investment will be rewarded by increased sensitivity . Our analysis suggests that a few carefully selected window sizes will be nearly as effective at detecting ncRNA as the entire set between 75 nt and 200 nt ( total of 26 window sizes ) . For example , when we used the entire set of window sizes from 75 nt to 200 nt , we detected 22 of the 29 known H/ACA snoRNA within embedded sequences ( Table 7 ) . If we had used only 4 window sizes ( 80 nt , 120 nt , 160 nt , 200 nt ) , we would have succeeded in identifying 90% of these H/ACA box snoRNA ( 20 of the 22 ) while reducing computational requirements by approximately 85% ( 4 of 26 window sizes ) . If these four window sizes were used with a step size of 25 nt , 77% ( 17 of 22 ) of the H/ACA box snoRNA would be detected ( Table S10 ) . This becomes 64% ( 14 of 22 ) if the step size is increased to 50 nt ( Table S11 ) . Tradeoffs between sensitivity and computational requirements should be evaluated when performing computational screens . We recommend using a range of four window sizes when screening for ncRNA in a genome ( one short , one long , and two intermediate values appears to be optimal ) . Our results suggest that the values of 80 , 120 , 160 and 200 should provide good results . A step size between 5 and 10 should also provide a good screen . These parameters should provide good ncRNA detection while keeping computational time manageable . The development of an efficient computational algorithm implementing the methodology presented here would also significantly reduce computational run time . This screen used a simple cutoff Z-score value ( ≤−3 . 5 ) to discriminate ncRNA . The sensitivity of the screen could probably be improved if a more sophisticated cutoff criteria were developed in which the Z-score cutoff was a function of window size . The number of aberrant negative Z-scores dropped as a function of window length in the negative control sets demonstrating that the likelihood of producing large negative Z-score drops with increasing window length . Developing a Z-score cut-off value as a function of window length would probably improve the sensitivity of the screen at longer window sizes . This work demonstrates that structural thermodynamic stability is an effective tool for predicting ncRNA genes . As examples of ncRNA are accumulated through computational screens such as this , it may become possible to determine ncRNA key features and gain insight into their biological function . Computational methods can complement experimental approaches in the effort to gain a deeper understanding of these genes .
S288C was used for all growth conditions except for sporulation ( SK1 ) and pheromone treatment ( BY4741 ) . Cells grown continuously at 25°C were collected by centrifugation , resuspended in an equal volume of 37°C medium , and returned to 37°C for an additional 20 minutes . The RNA was then isolated as described below . RNA was collected after twenty minutes as it has been shown to be the point of maximum RNA expression [53] . Pheromone treatment stimulates yeast cells to increase the expression of mating genes , arrest cell division in the G1 phase , and form polarizing mating projections directed toward the pheromone source [60] . Overnight yeast cultures grown in YPD at 30°C were treated with 50 nM α-factor ( GenScript Corporation ) . Cells were examined under a microscope to ensure schmooing was induced . Total RNA was extracted 75 minutes after pheromone treatment . A strong cellular response to diamide treatment has been shown previously [53] . It resembles a composite response to heat shock , H2O2 treatment and menadione treatment . It induces cellular redox genes and genes associated with defense against reactive oxygen species . Diamide ( Research Organics ) was added to cell cultures grown in YPD at 30°C in late log phase to a final concentration of 1 . 5 mM . Cells were returned to 30°C for growth for 30 minutes . RNA was then isolated as described above . This growth condition induces expression of genes involved in meiosis and spore morphogenesis . SK1 yeast cells were sporulated in a synchronous meiosis as described previously [61] . Briefly , yeast cultures were pre-grown in YPD to saturation at 30°C , diluted 200-fold into 100 ml of YPA ( 1% yeast extract , 2% Bacto-peptone , 2% acetate ) , and grown to early stationary phase ( about 5×107 cells/ml ) . Cells were then washed with water and resuspended into 100 ml of SPM ( sporulation media consisting of 0 . 3% potassium acetate and 0 . 02% raffinose ) . Sporulation was carried out at 30°C under conditions that allowed good aeration . Expression data suggested that metabolic , early , middle and late genes were active 11 hours after transfer to sporulation media so total RNA was collected at this time point [54] . Cells were inspected under a microscope to ensure that sporulation of at least some of the cells had taken place . RNA was then isolated as described below . S288C yeast cells were grown for approximately 55 hours in 100 ml of minimal media ( YNB ) in an anaerobic chamber using an anaerobic gas generating system ( Mitsubishi Gas Chemical Company , Inc . ) . Four minimal media plates were also streaked with S288C and grown anaerobically for the same time period . The anaerobic chamber was then opened and the cells growing on the plates were added to the cells in the liquid growth by washing . Total RNA from all of the cells was isolated immediately as described below . Saturated growth has been shown to activate gene expression , presumably allowing the cells to adapt to nutrient depleted conditions [53] . S288C cultures were grown to saturation ( OD 3 ) in minimal media ( YNB ) . They were also grown to logarithmic phase in rich media ( YPD ) and on a nonfermentable carbon source , YPGlycerol . All three cultures were grown at 30°C and aerated by shaking at 250–300 rpm . A phenol-chloroform extraction protocol was used as described previously [62] to extract total RNA from S . cerevisiae , S . bayanus and A . gossypii . All glassware used in the procedure was baked for 4 hours to destroy RNase activity . Reusable plasticware and laboratory bench surfaces were treated with RNAzap ( Biohit , Inc . ) . RNAse-free water was prepared by treating with Diethyl pyrocarbonate for one hour and then autoclaving . Cells were harvested from 50 ml cultures at an OD600 of 1–3 ( 1 OD = 3×107 cells/ml ) unless otherwise specified . The cells were collected via centrifugation ( except A . gossypii cells which were collected using a vacuum filter ) . The cell wall was disrupted by vortexing at high speed with acid-washed glass beads in a solution containing guanidine thiocyanate . Total RNA was purified using multiple washes with a combination of hot phenol and chloroform . The total RNA was treated with TURBO DNase ( Ambion ) and incubated at 37°C for 30 minutes prior to using for RACE or northern applications . The DNase activity was destroyed by heating to 70°C for 5 minutes per the standard protocol . RNA quality was assessed by measuring absorbance at a wavelength of 260 nm on a NanoDrop ( ND-1000 ) spectrometer . A 6% , 7 M urea , 1× TBE denaturing polyacrylamide gel was used to make a northern blot with total RNA as described previously [63] . Total RNA was treated with TURBO DNase ( Ambion ) and incubated at 37°C for 30 minutes prior to gel loading to ensure that no DNA was present . It was loaded onto the gel and run at 150 V for 3 hours . The total RNA was transferred from the gel to a nylon membrane using the OWL Scientific Panther Semi-Dry Electroblotter ( Model # HEP-1 ) with a current of 200 milliamperes for a period of 1 hr . The RNA was fixed to the blot with UV crosslinking . Radioactive strand-specific probes were produced using the Strip-EZ system with α-P32 dATP ( Ambion ) . Each probe was used to on a separate northern blot . This provided a check that the observed signal derived from only a single strand and was not the result of DNA contamination ( which would produce signal from both strands ) . The probes were detected by exposing the blot to BioMax XAR film ( Kodak ) at −80°C 24–48 hours . The SMART RACE cDNA Amplification Kit ( Clontech ) was used to map transcript ends . Total RNA was isolated from S288C under two different growth conditions: anaerobic growth and heat shock from 25°C to 37°C . It was treated with TURBO DNase ( Ambion ) prior to making the cDNA . To obtain RACE products for the ncRNA candidates , a RACE reaction and nested reaction were performed for both the Watson and Crick strand since it was uncertain which strand the transcript might be generated from . The temperature profiles developed to optimize the reaction are given in Appendix A . A hot start approach was used to minimize extraneous amplification by allowing the reaction tubes to reach a temperature of 94°C for 1 minute before adding the Ex Taq ( Takara ) polymerase . The RACE products were electrophoresed on a 1% agarose gel and the resulting bands were cut out of the gel . They were purified using one of two methods . The first was to use the QIAquick Gel Extraction Kit ( QIAGEN ) , according to the standard protocol . Alternatively , the gel slices were frozen at −20°C and then spun on a tabletop centrifuge at 1400 rpm for 20 minutes . The sample was then drawn from the top of the resulting liquid . This proved a quick and reliable method for obtaining purified product . The purified RACE products were sequenced using standard BigDye chemistry , version 1 . 3 , according to standard protocols ( Applied Biosystems ) . RACE primers were designed according to guidelines provided in the SMART RACE kit . They were 20–28 nt in length , had a GC content between 50–70% , a melting temperature ≥72°C , and had no more than 2 C's or G's in the last 5 nucleotides of the oligonucleotide . Each primer was confirmed to be unique in the genome using the “fuzznuc” program that is part of the EMBOSS utilities [64] . The Z-score compares the minimum folding energy ( MFE ) of a sequence , x , to the distribution of MFE generated by permuted versions of x having the same di-nucleotide composition . The di-nucleotide composition must be preserved because of the importance of stacked base-pairs in the MFE calculation [65] . The MFE of each sequence , x , was calculated using the RNAfold program [44] . Each sequence was then shuffled 500 times using the shuffle program provided in Sean Eddy's squid utilities [47] and a mean and standard deviation were calculated for the resulting distribution . The Z-score was then calculated using the equationwhere <·> and σ ( · ) denote the mean and the standard deviation of the MFEs of the sequences in xshuffled ( x ) . Hence , the Z-score represents the number of standard deviations that the sequence x deviates from the mean MFE of the shuffled sequences . The genome sequence data used for ncRNA prediction and subsequent evaluation of open reading frame coding potential is listed in Table 8 . It was important to investigate the possibility that the ncRNA candidates might be protein-coding genes . Comparative genomics was used to investigate this possibility for the four ncRNA gene candidates RUF20 , RUF21 , RUF22 and RUF23 ( Table 5 ) . This approach has been applied by other investigators with a high degree of success [66] . There are no conserved ORFs within the three candidates RUF21 , RUF22 and RUF23 among the closely related species S . cerevisiae , S . paradoxus , and S . bayanus ( sensu stricto ) . These transcripts are thus unlikely to be protein-coding genes . The RUF20 candidate contains one ORF consisting of 8 amino acids conserved among S . kudriavzevii , S . bayanus , S . paradoxus , and S . mikatae ( sensu stricto ) . However , the pattern of substitution among these species is not consistent with synonymous amino acid substitutions as would be expected for a protein-coding gene ( two mutations are in the 1st codon position , one mutation is in the 3rd position ) . The ORF is not conserved in Candida glabrata or A . gossypii . This is significant because our RACE data confirmed expression of the transcript in A . gossypii . In addition , the 8 amino acid ORF does not contain any splice signals suggesting that it is spliced to another exon . While a number of short ORFs have been identified in yeast [67] , there are none known to be as short as 8 amino acids . Taken together , this data strongly suggests that the short RUF20 ORF conserved among the sensu stricto does not encode a protein . The SnoScan [19] and SnoGPS [20] programs were used to test if the ncRNA candidates were likely to be snoRNA . The SnoScan program searches for features characteristic of C/D box snoRNA . None of the RUF20 to RUF23 candidate genes have features characteristic of C/D box snoRNA according to the program . The SnoGPS program searches for features characteristic of H/ACA box snoRNA . According to the program , RUF23 is unlikely to be a H/ACA box snoRNA . The program found some features of H/ACA box snoRNA evident in the RUF20 , RUF21 and RUF22 candidates , although their overall bit score was low ( 28 . 4 , 29 . 3 , and 29 . 9 respectively ) . A bit score value of 36 is recommended as the cutoff value when searching for new H/ACA snoRNA . To further evaluate the possibility that RUF20 , RUF21 and RUF22 might to be H/ACA snoRNA , sequence from two closely related species was used . The homologous gene sequences from S . paradoxus and S . bayanus were evaluated using the snoGPS program . The RUF20 candidates in these species were found to be unlikely to be a H/ACA snoRNA by the program . The RUF21 and RUF22 genes did generate possible H/ACA snoRNA candidates in the related species but there was no common rRNA target identified among the homologous sequences . Hence , the candidates appear to be unlikely H/ACA snoRNA genes . | Recent advances in DNA sequence technology have made it possible to sequence entire genomes . Once a genome is sequenced , it becomes necessary to identify the set of genes and other functional elements within the genome . This is particularly challenging as much of the genomic sequence does not appear to perform any function and is loosely referred to as “junk . ” Identifying functional elements among the “junk” is difficult . Experimental methods have been developed for this purpose but they are time-consuming , expensive , and often provide an incomplete picture . Thus , it is important to develop the ability to identify these functional elements using computational methods . Protein-coding genes are relatively easy to identify computationally , but other categories of functional elements present a significantly greater challenge . In this work , we used a computational approach to identify genes that do not encode for a protein but rather function as an RNA molecule . We then used experimental methods to verify our predictions and thereby validate the computational method . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"genetics",
"and",
"genomics/gene",
"discovery",
"computational",
"biology/genomics"
] | 2009 | Non-Coding RNA Prediction and Verification in Saccharomyces cerevisiae |
The mosquito immune system is involved in pathogen-elicited defense responses . The NF-κB factors REL1 and REL2 are downstream transcription activators of Toll and IMD immune pathways , respectively . We have used genome-wide microarray analyses to characterize fat-body-specific gene transcript repertoires activated by either REL1 or REL2 in two transgenic strains of the mosquito Aedes aegypti . Vitellogenin gene promoter was used in each transgenic strain to ectopically express either REL1 ( REL1+ ) or REL2 ( REL2+ ) in a sex , tissue , and stage specific manner . There was a significant change in the transcript abundance of 297 ( 79 up- and 218 down-regulated ) and 299 ( 123 up- and 176 down-regulated ) genes in fat bodies of REL1+ and REL2+ , respectively . Over half of the induced genes had predicted functions in immunity , and a large group of these was co-regulated by REL1 and REL2 . By generating a hybrid transgenic strain , which ectopically expresses both REL1 and REL2 , we have shown a synergistic action of these NF-κB factors in activating immune genes . The REL1+ immune transcriptome showed a significant overlap with that of cactus ( RNAi ) -depleted mosquitoes ( 50% ) . In contrast , the REL2+ -regulated transcriptome differed from the relatively small group of gene transcripts regulated by RNAi depletion of a putative inhibitor of the IMD pathway , caspar ( 35 up- and 140 down-regulated ) , suggesting that caspar contributes to regulation of a subset of IMD-pathway controlled genes . Infections of the wild type Ae . aegypti with Plasmodium gallinaceum elicited the transcription of a distinct subset of immune genes ( 76 up- and 25 down-regulated ) relative to that observed in REL1+ and REL2+ mosquitoes . Considerable overlap was observed between the fat body transcriptome of Plasmodium-infected mosquitoes and that of mosquitoes with transiently depleted PIAS , an inhibitor of the JAK-STAT pathway . PIAS gene silencing reduced Plasmodium proliferation in Ae . aegypti , indicating the involvement of the JAK-STAT pathway in anti-Plasmodium defense in this infection model .
Mosquito-borne diseases cause tremendous morbidity and mortality worldwide [1] . New approaches to control vector-borne diseases include interruption of the association between pathogens and vectors by genetic manipulation of vectors and the development of transmission-blocking vaccines . Potential success of these approaches requires in-depth knowledge of the molecular interactions between vectors' defense mechanisms and the evolutionary established ability of a pathogen to overcome these defenses . The yellow fever mosquito Aedes aegypti is the principal vector of Dengue fever and , due to a large body of knowledge amassed for this mosquito and readily available genetic and molecular tools , it also serves as an outstanding model for vector biology [2] . Sequencing and annotation of the genome of this mosquito have been critical in further advancing genomic and molecular approaches in studies of its immunity [3] , [4] . Despite being five times larger than the genome of the malaria mosquito Anopheles gambiae , the Ae . aegypti genome consists of a similar number of protein-encoding genes , around 17 , 700 [3] , [4] . Comparative genome analysis has indicated that 353 Aedes genes from 31 families are associated with immunity , compared with 285 and 338 immune genes in Drosophila melanogaster and An . gambiae , respectively , suggesting expansions of some immune gene groups in Ae . aegypti [4] . The key immune pathways are conserved between mosquitoes and the fruit fly; however , mosquitoes exhibit expansions of pattern recognition and effector molecules , likely due to their co-evolution with various pathogens [4] . Similar to Drosophila , mosquito Toll and IMD pathways constitute major immune pathways activating a battery of anti-microbial peptides and immune proteins in response to invasion by various microorganisms [4] , [5] . The activation of genes encoding these immune effector molecules is accomplished by the action of the NF-κB transcription factors REL1 , the orthologue of Drosophila Dorsal , and REL2 , the Relish orthologue , respectively [6] , [7] , [8] . Another important defense mechanism in Arthropods is melanization , which mediates wound healing and parasite encapsulation [9] . The key enzyme of melanization , phenoloxidase ( PO ) , is involved in the production of toxic melanin , which is deposited at the wound or around the parasite . A CLIP-domain serine protease cascade is responsible for amplification of signals , which are released upon infection , from wounded tissues or ruptured oenocytoids , and conversion of prophenoloxidase ( PPO ) into an active PO . ( Reviewer 1 , query 3 ) The activation of the melanization cascade is under strict regulation by serine protease inhibitors ( serpins ) . The importance of melanization cascades in mosquitoes is indicated by major expansions in their melanization pathway gene families ( 10 PPOs , 25 Serpins , and 79 CLIPs in Ae . aegypti ) [4] , [10] , [11] . The fat body of insects , such as Drosophila and mosquitoes , is the major metabolic tissue , and also serves as a powerful immune organ [5] , [12] . Although the role of the fat body in immunity has been demonstrated for the model insect Drosophila [5] , its precise function in immune responses in mosquitoes is still largely unknown . Deciphering the repertoire of immune genes expressed in the mosquito fat body is of particular importance because of the considerable expansion of immune-related genes in mosquitoes relative to that in Drosophila [4] , [10] , [11] . In previous studies , we have generated transgenic strains of Ae . aegypti , in which REL1 and REL2 were ectopically expressed under the control of the blood-meal-regulated promoter of the vitellogenin ( Vg ) gene in the fat body [13] , [14] . In this current work , we took advantage of the availability of these transgenic strains and performed transcriptome analyses to characterize repertoires of fat body-specific genes controlled by Toll and IMD pathways in this vector . Using microarray-based genome-wide transcriptional analyses , we have characterized gene repertoires in two transgenic Ae . aegypti mosquito strains that ectopically express either REL1 ( REL1+ strain ) or REL2 ( REL2+ strain ) . Moreover , we have shown a synergistic action of REL1 and REL2 in activating immune genes in the transgenic mosquito co-expressing both these NF-κB transcription factors . Infection of Ae . aegypti with Plasmodium gallinaceum resulted in the transcriptional modulation of a distinct subset of host immune genes . There was considerable overlap between the fat body transcriptome of Plasmodium-infected mosquitoes and the repertoire of genes regulated in mosquitoes transiently depleted of PIAS , an inhibitor of the JAK-STAT pathway . RNAi depletion of PIAS reduced Plasmodium proliferation in Ae . aegypti , indicating involvement of JAK-STAT in anti-parasite defense .
Previously generated transgenic strains of the mosquito Ae . aegypti ectopically expressing either REL1 or REL2 [13] , [14] have permitted us to decipher transcript repertoires of genes in the fat body controlled by the Toll and IMD pathways , respectively . We analyzed the transcriptional profiles of fat body-expressed genes using custom-made 60-mer oligonucleotide microarrays representing the approximately 17 , 700 Ae . aegypti genes [15] . The transgenic mosquitoes were constructed to ectopically express either recombinant REL1 or REL2 using the Vg promoter , which is a female- and fat-body-specific , blood meal-inducible gene [16] . The abundance of transcripts in REL1+ and REL2+ mosquitoes was compared with that in the non-transgenic wild type mosquitoes at 24 h post blood meal ( PBM ) , and genes uniquely regulated by REL1+ and REL2+ mosquitoes were further analyzed . The time point of 24 h PBM was chosen for transcriptome analyses because it is the expression peak for the Vg gene , whose upstream regulatory region was used to drive the expression of both REL1 and REL2 transgenes . The REL1 transgene is maximally expressed in the fat body of the REL1+ transgenic strain at this PBM time [13] . We reexamined REL2 transgene expression profile in the REL2+ strain , reported in [14] , by means of quantitative real time PCR ( qRT-PCR ) and found that its peak was at 24 h PBM ( Figure S1 ) . The fat body transcriptome of REL1+ transgenic mosquitoes contained 297 gene transcripts , 79 of which were up-regulated and 218 down regulated ( Figure 1A ) . Immune genes were the most predominant up-regulated group in the REL1+ fat body transcriptome , representing 66% of all up-regulated genes ( Figure 1A and Table S1 ) . Among the category of immune genes that were up-regulated in the REL1+ mosquito fat body were components of the Toll pathway , indicating the involvement of REL1 in the feedback regulation of its own pathway . These were genes encoding the Toll-specific pattern recognition receptor , the Gram Negative Binding Protein 1 ( GNBP1 ) , spätzle 3A , REL1 , and the negative regulator of the Toll pathway , cactus ( Table S1 ) . Activation of effector molecule transcripts–the anti-microbial peptides ( AMPs ) defensins A , C and D , and lysozymes C10 and C11–was high ( Table S1 ) . Defensins represent the major antimicrobial peptides ( AMP ) in mosquitoes [4] . Genes encoding opsonization factors , such as thio-ester proteins ( TEP ) –TEP2 , 3 , 20 , 21 , and 22–represented another predominant group of REL1-induced immune genes . Member of the TEP family have been identified in diverse animal species and play important roles in immune responses as components of the complement system [17] . In An . gambiae , hemocyte-specific TEP1 has been implicated as a key molecule involved in killing of midgut stages of Plasmodium [18] . It acts with two leucine-rich repeat ( LRR ) proteins , LRIM1 and APL1 , as a complement system in parasite killing [19] . However , functions of most TEPs in insects , including mosquitoes , remain to be elucidated . Transcripts of 6 genes , encoding galactose-specific C-type lectins were also elevated in the REL1+ transcriptome . Genes encoding proteolytic cascades and signaling modulators , CLIPs and serpins , were also represented in the immune repertoire of the REL1+-induced fat body transcriptome ( Table S1 ) . Previously , we have shown that some of these gene transcripts , TEP15 , TEP20 , defensin A , and CLIP13B , were activated by REL1 , thus , providing additional confidence to our genome-wide transcriptome data set [7] . Transcript of the gene encoding an orthologue of the Drosophila JAK/STAT pathway receptor Domeless ( Dome ) was among the most highly increased upon REL1 activation in the fat body ( AAEL012471 ) , indicating the involvement of Toll-REL1 pathway in regulating JAK-STAT pathway ( Table S1 ) . Dome is the Drosophila homolog of the vertebrate transmembrane cytokine class I receptor , which serves as a signal transducer and mediates activation of totA in the fat body [20] , [21] . Activation of Dome/JAK/STAT signaling requires hemocyte-specific cytokine Unpaired [21] . Fat body totA is also regulated by Relish , a Drosophila orthologue of mosquito REL2 . Here , we provide evidence on the involvement of REL1 in up-regulation of Dome , the gene encoding a key component of the JAK/STAT pathway in the mosquito fat body . Several recent studies in Drosophila have pointed out on a communication between immune tissues[21] , [22] , [23] , [24] . In addition to hemocyte-specific cytokine mediated activation of the Dome/JAK/STAT in the fat body , blood cells are also required for the immune activation of the fat body [22] , [24] . However , Rel proteins , Dif and Dorsal , also act in the fat body to produce factors that promote blood-cell number in Drosophila larvae [23] . The identity of these fat body factors remains undetermined . There is no evidence of possible involvement of Relish in similar fat body – blood cell communication . Utilization of fat body-specific , ectopically expressed REL1+ represents a unique opportunity to address the question about fat body – hemocyte communication in mosquitoes . Thus , although the ectopic expression of REL1 in the REL1+ transgenic mosquitoes is strictly fat body-specific , the fat body REL1-mediated production of blood cell stimulating factors could activate proliferation of blood cells adhered to fat body preparations . As a consequence , the overall transcriptome from REL+ mosquitoes could include genes from proliferating hemocytes . This aspect of fat body – blood cell communication will be studied further in the future . The REL1+ controlled transcriptome contained a large number of genes ( 230 ) attributed to non-immune biological processes; 26 . 6% of these gene transcripts ( REL1 ) were involved in ribosomal biogenesis , DNA replication and metabolism ( Figure 1A; Table S1 ) . 88% of these non-immune genes were down-regulated . Notably , 44 of down-regulated genes were related to ribosomal biogenesis and translation . This observation is in agreement with microarray analyses of ectopic expression of Rel proteins in Drosophila [25] . One of genes activated in the REL1+ fat body transcriptome encodes an orthologue of a vertebrate Grb2-associating protein ( Gasp , AAEL002492; Table S1 ) , which is a thymus-specific factor critical for T-cell differentiation [26] . Finding its function represents a potentially important aspect of immunity in the mosquito . REL1 also up-regulates an orthologue of the cytosolic sulfotrnasferase , SULT ( AAEL006334 , Table S1 ) . Members of SULT superfamily catalyze the sulfation of xenobiotics , hormones and neurotransmitters [27] . Considering multiple functions of these enzymes , it is difficult to predict the role of this REL1-dependent SULT in the mosquito fat body . An interesting glimpse in the gene functional conservation is also provided by the Rel-mediated up-regulation of an orthologue of a vertebrate major facilitator superfamily domain-containing protein ( Mfsd2a , AAEL009195; Table S1 ) , which is expressed in brown adipose tissue and liver ( the fat body is a functional analogue of these tissues combined ) [28] . In vertebrates , Mfsd2a is highly expressed during thermogenesis and have been found to be a tumor suppressor [28] , [29] . The fat body transcriptome of REL2+ transgenic mosquitoes contained 299 genes , 123 of which were up-regulated and 176 down regulated ( Figure 1A and Table S2 ) . Immune genes represented 44% of most highly up-regulated genes in the REL2+ fat body transcriptome . In particular , transcripts of AMPs and recognition molecules were enriched ( Table S2 ) . Genes encoding thio-ester proteins ( TEP ) –TEP20 , 21 , and 22–were also elevated among REL2-induced immune genes . Genes encoding factors of the IMD pathway–a peptidoglycan recognition protein ( PGRP-S1 ) and REL2 - were up-regulated ( Figure 1A and Table S2 ) . Two members of the APL1 family of leucine-rich ( LRR ) proteins , APL1B ( AAEL012086 ) and APL1C ( AAEL009520 ) , were up-regulated in the REL2+ fat body transcriptome but not in the REL1+ one ( Tables S1 and S2 ) . LRR proteins play an important role in the innate responses against pathogens in plants , insects , and mammals [30] , [31] , [32] . APL1 ( Anopheles Plasmodium-responsive leucine-rich repeat 1 ) was first identified in An . gambiae , in which it controls resistance to Plasmodium falciparum [33] . The APL1 family is comprised of paralogs APL1A , APL1B and APL1C [34] . APL1C is responsible for defense of An . gambiae against P . berghei , which is a rodent parasite . APL1C has been reported to function within the REL1-cactus immune signaling , which regulates APL1C at the transcriptional and translational levels [34] . However , further studies have revealed that protection of An . gambiae against its natural parasite P . falciparum is mediated by APL1A [35] . This protection correlates with the transcriptional control of APL1A by REL2 , suggesting that REL2 anti-parasite phenotype results partially from its control of APL1A [35] . APL1C has been implicated in a complement-like pathway that mediates parasite killing interacting with LRIM1 and TEP1 [36] , [37] . Our data indicate that Ae . aegypti APL1 proteins are controlled by REL2 . Up-regulation of a fibrinogen-related protein ( AAEL004150 ) was also observed in the REL2+ fat body transcriptome . The fibrinogen-related gene family belongs to pattern recognition receptors and involved in innate immunity in both invertebrates and vertebrates [38] , [39] , [40] . In An . gambiae , fibrinogen-related proteins interact with Gram-positive , Gram-negative bacteria and co-localized with both P . berghei and P . falciparum [40] . It has been suggested that fibrinogen-related proteins expand pattern recognition capacity , thus , enhancing innate immunity against various pathogens . 227 genes in the REL2+ fat body transcriptome belonged to genes encoded factors of non-immune biological processes . Transcript levels of some genes related to non-immune functional categories , most notably stress and metabolism , were predominantly repressed ( Figure 1A; Table S2 ) . Thirty-three genes ( 6 induced and 27 repressed ) in the REL2+ regulated transcriptome were related to proteolysis process . Interestingly , REL2+ transcriptome contained an up-regulated component of ribosome biogenesis , 20S rna accumulation protein 1 ( AAEL004493 ) , in contrast to overall down-regulation of genes related to ribosomal biogenesis and translation in the REL1 transcriptome . This difference points out on specificity of action of REL1 and REL2 not only in affecting immune , but also non-immune genes . We also compared the fat body transcriptomes of REL1+ and REL2+ transgenic mosquito with those of mosquitoes in which either the negative regulator cactus or caspar had been depleted by RNAi silencing . The latter two transcriptomes have been previously reported and are represented here for comparative purposes only [15] . Cactus is a repressor of Drosophila Dorsal/Dif and mosquito REL1 , which has been shown to directly interact with this NF-κB factor preventing the latter to translocate to the nucleus [5] , [15] , [41] , [42] . In mosquitoes , cactus silencing results in activation of REL1 and its underlying immune responses [7] , [10] , [11] , [15] , [41] . Hierarchical clustering confirmed the close relationship between the immune transcriptomes regulated by transgene REL1 overexpression and cactus depletion ( Figure 1C , Cluster I and Table S4 ) . However , REL1+ affected transcript abundance of fewer genes in diverse functional classes compared to cactus depletion . Our analysis revealed the presence of the same 53 genes in transcriptomes from REL1+ , REL2+ transgenic and cactus-depleted mosquitoes ( Figure 1B ) . 30 of them belonged to immunity category . In Drosophila and Anopheles , caspar has been shown to be an inhibitor of the IMD pathway , in which it has been suggested to prevent Dredd-dependent nuclear translocation of Relish and REL2 [43] , [44] . In Ae aegypti , RNA depletion of caspar triggered up-regulation of only a small number of genes when compared with REL2 transgene ectopic expression ( [15] and this report ) . Moreover , caspar-induced transcriptome only marginally overlapped with that of REL2+ ( Figure 1C and Table S4 ) . Further studies are required to clarify the role of caspar in the regulation of the IMD pathway . Importantly , 84 genes were present in both REL1+ and REL2+ fat body transcriptomes , suggesting co-regulation of these genes by the NF-kB factors REL1 and REL2 and their respective pathways ( Figure 1B and Figure S2 , Table S3 ) . The majority of highly enriched gene transcripts ( 50% ) , which were common for both REL1+ and REL2+ fat body transcriptomes , belonged to the immunity category . The AMP genes defensins A , C , D and lysozyme C displayed increased mRNA abundance in response to either REL1 or REL2 . However , REL2 appeared to be a more potent activator of these AMPs . A group of six galactose-specific C-type lectin transcripts was highly elevated in both transcriptomes . Gene transcripts encoding TEPs , TEP2 , TEP20 , TEP21 and TEP22 , appeared to be equally upregulated by REL1 and REL2 . Only 22% of all non-immune genes were present in both REL1+ and REL2+ -regulated fat body transcriptomes in contrast of 50% of immune ones . Among non-immune genes that were induced in both transgenic mosquitoes was juvenile hormone esterase ( JHE ) . Increased JHE activity has been linked with degradation of juvenile hormone during PBM development in Ae aegypti females [45] . However , modulation of juvenile hormone titer via immune factors has not been previously reported . The majority of non-immune down-regulated genes in both transcriptomes belonged to metabolism and cell cycle functional categories ( Table S3 ) . To decipher whether genes represented in both REL1+ and REL2+ fat body transcriptomes were synergistically regulated by these NF-κB transcription factors , we generated a hybrid REL+/REL2+ transgenic mosquito strain by crossing REL1+ and REL2+ strains . Unlike parental RE11+ and REL2+ strains , the hybrid REL1/REL2 mosquitoes carried both Vg-REL1 and Vg-REL2 transgenes , which were ectopically expressed after a blood meal specifically in female fat bodies ( Figure 2 ) . We analyzed transcript abundance of selected genes representing the functional group of 33 immune genes upregulated in both REL1+ and REL2+ transcriptomes . REL1+ , REL2+ and REL1+/REL2+ hybrid transgenic mosquitoes were blood fed , RNA was isolated from their fat bodies 24 h PBM and subjected to qRT-PCR analysis . This analysis revealed that the transcript levels of Defensin A , Defensin C , CLIPB39 , and TEP20 genes in the REL1+/REL2+ hybrid transgenic female mosquitoes was considerably higher as compared to those in either REL1+ or REL2+ strains ( Figure 2 ) . Defensin A and Defensin C were particularly elevated in the REL1+/REL2+ hybrid mosquitoes ( Figures 2A and 2B ) . A predominant concept in insect immunology is that Toll and IMD pathways act independently , with the Toll pathway responding to fungi and Gram-positive bacterium-derived Lys-type peptidoglycan and the IMD pathway to Gram-negative bacterium-derived diaminoacipimelic acid ( DAP ) -type peptidoglycan , with each pathway activating a separate set of effector genes [5] , [42] . However , the transcriptome analyses of double Drosophila mutants of the Toll and IMD pathways have revealed some co-regulated antimicrobial peptides [46] . Regulation of Drosophila antifungal AMP Drosomycin mainly depends on Toll pathway and receives a modest input from IMD during a systemic immune response; though , the IMD pathway solely activates Drosomycin and Diptericin , respective target genes of the Toll and IMD pathways , in the local immune response . [47] , [48] . Tanji et al . [49] have shown the synergistic action of Toll and IMD pathway in activating Drosomycin and Diptericin . Moreover , DIF and Relish form heterodimers to regulate antimicrobial peptides in Drosophila [50] . Our transgenic approach strongly suggests the synergistic action of REL1 ( an orthologue of Dorsal ) and REL2 ( an orthologue of Relish ) in activation of immune genes in the mosquito Ae . aegypti . The high level of up-regulation of immune genes in REL1+/REL2+ hybrid mosquitoes as co-expression of Vg-driven REL1 and REL2 clearly indicated the synergy of interaction between these NF-kB factors . Comparable levels of individual ectopic expression of each REL factor in a respective transgenic REL strain did not elicit similarly high up-regulation of immune factors , assayed in this experiment . Previously , we have shown that simultaneous ectopic expression of Defensin A and Cecropin A in Ae . aegypti leads to a complete elimination of the malaria parasite P . gallinaceum and interruption of its transmission in transgenic hybrid CecA/DefA mosquitoes [51] . Ectopic co-expression of REL1 and REL2 in a sex- , tissue- and stage-specific manner that elicits a strong synergistic effect on activation immune factors provides a potent method to study mosquito and pathogen interaction . Microarray-based transcriptome analysis was used to study the effect of immune signal transduction pathways on melanization-related gene expression in the Ae . aegypti fat body . At 24 h PBM , 21 melanization-related genes were significantly upregulated in the fat body of transgenic REL1+ female mosquitoes , while 24 ( 22 up- and 2 down- ) genes were controlled by REL2 ectopic expression in the same tissue ( Figure S2 , Table S5 ) . CLIP-domain serine proteases can be separated into five subfamilies , among which CLIPA , B , and C are implicated in the activation of melanization . The CLIPA subfamily is composed of the non-catalytic clip domain serine protease homologues , which contain imperative PPO activation cofactors . Mosquito CLIPA14 and CLIPA6 are homologous to the PPO activation cofactors , Manduca sexta SPH1 , SPH2 ( serine protease homologue ) , and Holotrichia diomphalia PPAF2 ( PPO activating factor ) [52] . CLIPA1 and CLIPA11 were up-regulated in both REL1+ and REL2+ . However , CLIPA5 , CLIPA6 , and CLIPA16 were induced only in the REL1+ , while CLIPA14 was enriched in the REL2+ . Two melanization proteases ( CLIPB39 and CLIP40 ) [11] and CLIPB79 were induced in both REL1+ and REL2+ mosquitoes . Of all the CLIPC , D , and E subfamilies , only a single gene , CLIPE8 , was induced in REL2+ ( Table S5 ) . REL1 and REL2 differently regulated transcription of genes encoding Serpins . Serpins-9 , −16 , −4B , −4C were up-regulated only in the REL1+ mosquitoes , while Serpins-2 , −11 , and −23 mRNAs in REL2+ mosquitoes ( Table S5 ) . Serpins-1 , −8 , and , −16 mRNAs were elevated in both REL1+ and REL2+ mosquitoes ( Table S5 ) . In Ae . aegypti , Serpin 1 is involved in control of immune melanization , while Serpin-2 in tissue melanization , exemplified by the formation of melanotic tumors after RNAi Serpin-2 depletion [11] . PPO gene transcripts were not detected in either REL1+ or REL2+ fat-body-specific transcriptomes . This is in agreement with previous data showing that PPO genes are expressed in hemocytes in both Drosophila and mosquitoes [5] , [53] . To assess the relationship of REL1+ and REL2+ transcriptomes with transcriptional responses induced by Plasmodium infection in the mosquito Ae . aegypti , we compared transcriptomes of mosquitoes fed on a Plasmodium-infected blood meal with those fed on a non-infectious blood meal . In both the midgut and fat body , the majority of genes regulated in the presence of Plasmodium belonged to diverse or unknown functional classes , defined as such because of insufficient information for assigning particular known functions . The number of genes that significantly changed their transcript levels after Plasmodium infection in the midgut was almost twice higher than that in the fat body; impressively , Plasmodium infection envoked the induction of 375 genes and repression of 724 genes in the midgut , while 513 genes were induced and 174 genes were repressed in the fat body ( Figure 3 , Tables S6 and S7 ) . The quantitative RT-PCR used to verify transcripts levels for 23 genes ( 12 from PgFB and 11 from PgMD ) showed a high degree of correlation ( best-fit linear-regression R2 = 0 . 77 ) with the microarray transcriptome data ( Figure S3 ) . The immune-related genes were the third most-represented functional gene group in both the midgut ( 124 genes ) and the fat body ( 99 genes ) transcriptomes in Plasmodium-infected mosquitoes ( Figure 3 ) . In the midgut , transcriptional responses affected by Plasmodium infection were marked by a significant down-regulation of immune gene transcripts ( 99 out of 124 regulated immune genes ) . Among these down-regulated immune genes were several serine proteases ( SPs ) , CLIP-domain serine proteases , Serpins ( serpins-4A , −7 , −11 , −17 , −20 and −21 ) , lysozyme C , various PRR molecules such as the PGRP proteins ( PGRP-LA , PGRP-LD , and PGRP-SC2 ) , GNBP1 , fibrinogen-related proteins ( FBN12 , FBN12 , FBN13 , FBN18 , FBN24 , and FBN27 ) , three AMPs ( defensins A and D , cecropin D ) ( Table S6 ) . Majority of genes putatively related to the melanization cascade were down-regulated in the midgut ( 26 down-regulated and only 4 up-regulated ) . The transcript abundance of CLIPA5 , CLIPA6 , CLIPB13A , CLIPB13B , CLIPB5 CLIPA3 , CLIPB1 , and Serpin-11 mRNAs were reduced in Plasmodium infected midgut ( Table S6 ) . CASPS18 was upregulated , while CASPS7 and CASPS20 were downregulated . 32 out of 40 genes related to oxidative stress were down-regulated , including eight cytochrome P450s , three carboxylesterase , and one glutathione peroxidase . Down-regulation of stress response gene expression suggested the existence of mechanism genes could potentially interfere with the proliferation of parasites in the midgut . Genes related to transport processes were differentially regulated in the Plasmodium-infected midgut transcriptome , with 38 repressed and 25 induced . A gene encoding oxidoreductase ( AAEL003312 ) was significantly up-regulated , suggesting Plasmodium-mediated elevated activity of this enzyme ( Table S6 ) . Among potential functions of oxidoreductase is detoxification of reactive oxygen species ( ROS ) , which play a pivotal role in anti-Plasmodium gut resistance [54] . A pronounced immune response was detected in the fat body of mosquitoes 24 h after a Plasmodium-infected blood meal ( Figure 3 and Table S7 ) . Significantly , Plasmodium infection resulted in the enrichment of 74 immune gene transcripts in the fat body ( out of 99 regulated immune genes in this tissue ) . REL1 ( AAEL007696 ) and REL2 ( AAEL007624 ) were induced in the fat body transcriptome of the Plasmodium-infected mosquitoes , suggesting that infection with the parasite activated both the Toll and IMD pathways ( Table S7 ) . Components of the Toll pathway - GNBP3 , TOLL8 , TOLL11 , and spätzle 6 - were also upregulated . The IMD receptors PGRP-LP and PGRP-S5 were elevated . However , IMD was down-regulated . Attacin C , Defensins C and D were among up-regulated AMPs . A distinct immune response of the Aedes fat body to Plasmodium infection was the activation of two Down-syndrome adhesion molecules ( Dscam , Table S7 ) . Dscam is a member of the immunoglobulin superfamily , its gene comprises of multiple exons , alternative splicing of which generates 19 , 000 different extracellular domains and provides [55] . An . gambiae orthologue of Dscam contains 101 exons that can produce over 31 , 000 alternative splice forms [56] . Hemocyte-specific Dscam isoforms have been associated with phagocytotic uptake of bacteria [55] , [57] . In An . gambiae , Dscam has been implicated in resistance to bacteria and Plasmodium [55] . Dscam is also expressed in Drosophila fat body , which is in agreement with our observation [55] . The role of fat body-specific Dscam isoforms remains to be elucidated . Multiple TEPs ( TEP13 , TEP15 , TEP20 , TEP22 , and TEP23 ) and leucine-rich ( LRR ) proteins were up-regulated in the fat body in response to Plasmodium infection as well; however , their functions are not clear . Plasmodium infection also caused changes in mRNA abundance in apoptosis related genes; IAP-2 ( an inhibitor of apoptosis ) , CASPS18 , and CASPS8 were induced , while CASPS19 was repressed in the fat body of the Plasmodium-infected mosquitoes . An interesting feature of Plasmodium-affected fat body transcriptome is the down-regulation of DOME , the JAK/STAT receptor , which was up-regulated in the REL1+ fat body transcriptome ( Table S7 ) . Another distinct feature of Plasmodium-affected fat body transcriptome was elevation transcriptional activity as evident by up-regulation of six zinc finger and forkhead transcription factors ( Table S7 ) . We found that the fat body transcriptome , the gene encoding dual oxidase ( DUOX , AAEL007563 ) was activated by Plasmodium infection ( Table S7 ) . DUOX enzyme is involved in production of reactive oxygen species ( ROS ) , which have been implicated in anti-microbial immunity [58] . ROS has been implicated in innate immune responses in the gut and anti-Plasmodium defenses [54] , [58] , [59] . The anti-Plasmodium effect of ROS is mediated by bacterial flora [60] . The adverse effect of ROS is modulated by antioxidants , including Gpx [54] . Our finding of DUOX in the Plasmodium-induced fat body transcriptome adds a new aspect in immune function of ROS . A possibility of activation of these enzymes in hemocytes attached to the fat body could not be ruled out . Availability of REL1+- and REL2+-induced fat body transcriptomes permitted us to conduct a comparative analysis with the P . gallinaceum fat body ( PgFB ) -responsive transcriptome . The comparison of fat body of Plasmodium infection-responsive gene transcript repertoire with those of REL1+ and REL2+ mosquitoes showed an overlap between these transcriptomes ( 7 induced , 5 repressed ) ( Figure 4 and Table S8 ) . Only three immune genes ( CLIPB13B , CLIPB15 , and Serpin-8 ) were found in the overlap of these transcriptomes , suggesting that immune responses elicited by the Plasmodium infection in the fat body were distinct from those regulated by either REL1+- and REL2+ ( Figure 4 , Figure S2 and Table S8 ) . Hierarchical clustering showed that cluster I consisted of a large group of down-regulated genes from PgFB-responsive and REL1+ and REL2+ gene repertoire , while cluster II represented genes , which were induced in by REL and repressed by Plasmodium challenge . Conversely , cluster IV was largely enriched by up-regulated immune genes ( 82% ) , which are putatively involved in melanization and signaling amplification ( Serpins , CLIPs ) and parasite recognition and killing ( TEPs , Lectins ) ( Table S8 ) . The core components of the JAK-STAT pathway are evolutionarily conserved from arthropods to mammals [61] . PIAS ( Protein Inhibitor of Activated STAT ) has been identified as a negative regulator of the JAK-STAT pathway in mammals , Drosophila and Aedes [62] , [63] . A multiple polypeptide sequence alignment showed that insect PIAS had a domain structure similar to that from vertebrates ( Figures S4A and S4B ) . The sequence of Aedes PIAS shared a high level of homology with Anopheles PIAS ( identity 61% , similarity 71% ) and Drosophila PIAS ( identity 47% , similarity 57% ) . It also had a lower degree of homology with human PIAS ( identity 32% , similarity 46% ) ; but the invertebrate PIAS group formed its own independent clade ( Figure S4C ) . Impressively , Ae aegypti PIAS gene contains ten alternative spliced isoforms , however , their respective roles are not known ( Figure S4 ) [3] . We compared the Plasmodium infection-responsive fat body transcriptome with that of PIAS-gene silenced mosquitoes , which was reported earlier [63] and found a significant overlap between these transcriptomes [63] ( 29 gene transcripts: 26 induced and 3 repressed ) ( Figure 3 ) . The transcript of the SOCS ( suppressor of cytokine signaling ) gene , which encodes another JAK-STAT pathway negative regulator , was highly enriched in the fat body in response to P . gallinaceum infection . Interestingly , not a single gene displaying transcript enrichment upon P . gallinaceum infection seemed to be co-regulated by cactus and PIAS silencing , suggesting that anti-Plasmodium responses mediated by Toll and JAK-STAT pathways were different . Hierarchical cluster analysis revealed several genes with mRNA abundance enriched in both PgFB , and PIAS-depleted fat body transcriptomes , but unaffected in REL1+ and REL2+ fat body transcriptomes ( Figure 4 , Cluster III; and Table S8 ) . As stated , cluster IV is mainly composed by immune genes , which are involved in melanization , pattern recognition , and signaling amplification ( Table S8 ) . Additionally , overlap between PIAS-depleted and PgFB transcriptome revealed numerous important immune genes: a putative LRR , whose gene family has been linked to Plasmodium killing in An . gambiae [33] , [64] , spätzle 6 , and two dengue virus restriction factors ( DVRF1 and DVRF2 ) [63] ( Figure S2B ) . We then evaluated the anti-P . gallinaceum activities of three major Ae aegypti immune pathways Toll , IMD , and JAK-STAT . One-day-old mosquitoes were injected with dsRNA for either one of the following negative regulator genes of these pathway: PIAS , caspar , cactus , or luc , as a control ( Figure S5 ) , and then fed on Plasmodium-infected blood 4 days later . Depletion of cactus , and hence activation of the Toll pathway REL1 , resulted in the highest level of resistance to P . gallinaceum ( Figure 5 ) . Knockdown of PIAS , which resulted in the activation of the JAK-STAT pathway-regulated immune response , also increased mosquito resistance to parasite infection in the midgut by a six-fold ( Figure 5 ) . However , we observed no anti-P . gallinaceum effect upon activation of the IMD pathway REL2 factor through depletion of caspar ( Figure 5 ) . Depletions of the negative regulators of Toll , IMD , and JAK-STAT pathways – cactus , caspar and PIAS – demonstrated differential patterns of resistance in different mosquito-Plasmodium infection models . REL1 activation by depletion of cactus resulted in the strongest anti– P . berghei and anti-P . gallinaceum effects in An . gambiae and Ae . aegypti , respectively [7] , [10] , [11] , [15] , [41] . Depletion of caspar has shown that the IMD pathway is most effective against the human pathogen P . falciparum in An . gambiae and other anopheline species [44] . In the present study we have not observed any effect of caspar depletion on the resistance of Ae . aegypti to P . gallinaceum , while our previous study based on overexpression of REL2 in transgenic Ae . aegypti , has clearly shown involvement of IMD pathway in defense against this pathogen [14] . This discrepancy , taken together with the findings from our transcriptome studies of REL2+ and caspar-depleted mosquitoes , may suggest that caspar is likely to regulate a branch of the IMD pathway , involving a subset of effector genes . Moreover , we have also demonstrated that simultaneous overexpression of two anti-microbial peptides , Cecropin A and Defensin A , which are under the dual control of Toll and IMD pathways , lead to a complete elimination of P . gallinaceum and termination of transmission [51] . In this current study , we have implicated the JAK-STAT pathway in anti-Plasmodium defense in Ae . aegypti . The STAT pathway is involved in late-phase immunity against P . berghei and P . falciparum in An . gambiae [65] . In conclusion , utilization of transgenic Ae . aegypti mosquitoes with altered immunity by means of ectopic expression of the NF-kB transcription factors REL1 and REL2 has permitted deciphering gene repertoires activated by Toll and IMD pathways in the fat body , the central tissue to mosquito immunity . Importantly , transgenic mosquitoes ectopically expressing both these factors exhibited strong synergistic activation of immune genes . A close correlation has been noted between REL1+ and cactus-depleted transcriptomes . In contrast , the REL2+ transcriptome was strikingly different from that of caspar-depleted mosquitoes , suggesting that caspar regulates a sub branch of the IMD pathway . Infections of the wild type Ae . aegypti with P . gallinaceum elicited enrichment of a distinct subset ( 76 up- and 25 down regulated ) of immune gene transcripts relative to that observed in REL1+ , REL2+ , and cactus-depleted mosquitoes . Considerable overlap was observed between the fat body transcriptome of Plasmodium-infected mosquitoes and that of mosquitoes depleted of PIAS , the inhibitor of the JAK-STAT pathway . PIAS gene silencing reduced Plasmodium proliferation in Ae . aegypti , indicating the involvement of the JAK-STAT pathway in anti-Plasmodium defense in this infection model in addition to Toll and IMD pathways .
This study was carried out in strict accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health . The protocol was approved by the University of California Riverside Institutional Animal Care and Use Committee ( IACUC #A20100016; 05/27/2010 ) and all efforts were made to minimize suffering . Wild type and transgenic Ae . aegypti mosquitoes , of the UGAL/Rockefeller strain , REL1+ [13] and REL2+ [14] , were maintained in laboratory culture under conditions of 27°C and 80% humidity . Female mosquitoes 3–5 days post-eclosion were fed on the blood of anesthetized white rats to initiate egg development . Hybrid REL1+/REL2+ mosquitoes were generated by crossing REL1+ and REL2+ . Selection of hybrids was performed as previously described [51] . To generate these hybrids , the two strains , REL1+ and REL2+ , were maintained as homozygous for four generations before crossing . The hybrid strain was established by crossing REL1 females with REL2 males; the F1 hybrid females were used for the experiments . Adult mosquitoes were maintained on 10% sucrose solution and water [66] . The avian malaria P . gallinaceum was maintained under the natural transmission cycle between the mosquito and chickens . To determine the number of parasite oocysts in the midgut dissected at 8 days post-infection , the tissue was stained with 1% mercurochrome and oocysts were counted under Nikon E400 light microscopy . All dissections ( fat body and midgut ) were performed in Aedes physiological solution ( APS ) [66] . Abdominal walls with adhering fat body tissue and free from other internal tissues ( thereafter called fat body ) were washed in APS to removed hemolymph and blood cells before freezing in liquid nitrogen . Midgut preparations included the anterior and posterior ( stomach ) without Malpighian tubules and hindgut . PIAS sequences from different metazoan species , retrieved from NCBI , Vectorbase , and Ensembl , were analyzed in PROSITE and SMART to confirm conserved domain structures . Multiple sequences were aligned in ClustlX2 . 0 ( Blosum matrixes , gap penalty 10 , and extension penalty 0 . 1 ) . A phylogenetic tree was constructed based on the neighbor-joining method and displayed by Treeview . Parasite oocyst data generated from three independent experiments were analyzed using the Kolmogorov-Smirnov goodness-of-fit test and pooled . The statistically significant difference between samples was calculated using the Mann-Whitney test ( Graphpad 5 . 0 ) . Total RNA was extracted from the fat body of eight mosquitoes using the Trizol method ( Invitrogen ) , according to the manufacturer's protocol . Total RNA ( 5 µg ) from each sample was separated on a formaldehyde gel , blotted and hybridized with the corresponding 32P-labeled DNA probe . Probes were generated using PCR and then following the High Prime ( Roche ) protocol . Actin was used as a loading control . For RT-PCR and Real-time PCR , cDNAs were synthesized from 2 µg total RNA using Omniscript Reverse Transcriptase kit ( Qiagen ) . RNA was treated with DNase I ( Invitrogen ) before cDNA synthesis . PCR was performed using the Platinum High Fidelity Supermix ( Invitrogen ) . Real-time PCR was performed on the iCycler iQ system ( Bio-Rad , Hercules , CA ) and we used an IQ SYBR green supermix ( Bio-Rad ) . Quantitative measurements were performed in triplicate and normalized to the internal control of S7 ribosomal protein mRNA for each sample . Primers and probes are listed in Table S9 . Real-time data were collected from the software iCycler v3 . 0 . Raw data were exported to EXCEL for analysis . Double-stranded RNA synthesis followed a method described previously [10] , [15] . In brief , double-stranded RNA ( dsRNA ) of specific gene template was synthesized using the MEGAscript kit ( Ambion ) . The luciferase gene was used to generate control iLuc dsRNA . After dsRNA synthesis , samples were treated by means of phenol/chloroform extraction and then ethanol precipitation . DsRNA was then suspended in Rnase-free water to reach a final concentration of 5 µg/µl . Naïve adult female mosquitoes were selected at 24 h post-emergence for dsRNA injection experiments . The Picospritzer II ( General Valve , Fairfield , NJ ) was used to introduce corresponding dsRNA into the thorax of CO2-anesthetized mosquito females , at one or two days post-emergence . DsRNA ( 300 nl ) was injected into the thorax of each adult Ae . aegypti female mosquito . Primers used for dsRNA knockdowns are listed in Table S9 . Transcription assays and analysis were conducted following standard protocols with a full genome Agilent-based microarray platform [15] . Relative mRNA abundance was compared between treated and control samples . In brief , 2–3 µg total RNA was used for probe synthesis of cy3- and cy5-labeled dCTP . Hybridizations were conducted with an Agilent Technologies In Situ Hybridization kit at 60°C , according to the manufacturer's instructions . Hybridization intensities were determined with an Axon GenePix 4200AL scanner , and images were analyzed with Gene Pix software . The expression data were processed and analyzed as described previously [15] . In brief , the background-subtracted median fluorescent values were normalized according to a LOWESS normalization method , and Cy5/Cy3 ratios from replicate assays were subjected to t-tests at a significance level of p<0 . 05 , using TIGR , MIDAS , and MeV software [67] . Expression data from all replicate assays were averaged with the GEPAS microarray preprocessing software prior to logarithm ( base 2 ) transformation . Self–self hybridizations were used to determine the cut-off value for the significance of gene regulation on these types of microarrays to 0 . 8 in log2 scale , which corresponds to 1 . 74-fold regulation [68] . For genes with p<0 . 01 , the average ratio was used as the final fold change; for genes with p>0 . 01 , the inconsistent replicates ( with distance to the median of replicate ratios larger than 0 . 8 ) were removed , and only the value from a gene with at least two replicates in the same direction of regulation were further averaged . Three independent biological replicate assays were performed . Numeric microarray gene expression data are presented in Tables S1 , S2 , S5 , S6 , S7; validation data by quantitative Real-time PCR- in Table S10 and Figure S3 . | Mosquito-borne diseases cause tremendous morbidity and mortality worldwide . New approaches to control vector-borne diseases include interruption of the association between pathogens and vectors by genetic manipulation of vectors and the development of transmission-blocking vaccines . Potential success of these approaches requires in-depth knowledge of the molecular interactions between vector defense mechanisms and the ability of a pathogen to overcome these defenses . A combination of the genome-wide microarray and transgenic approaches has permitted us to decipher repertoires of genes controlled by two major immune pathways , Toll and IMD , in the Dengue-fever mosquito vector Aedes aegypti . We have shown that these pathways interact to bring about a high level of immune genes by means of generating a transgenic strain , which ectopically expresses the NF-κB factors of TOLL and IMD . In Ae . aegypti , a malaria parasite Plasmodium gallinaceum elicited the transcription of a distinct subset of immune genes relative to those observed in transgenic mosquitoes with activated Toll or IMD pathways . However , a considerable overlap was observed between the fat body transcriptome of Plasmodium infected mosquitoes and that of mosquitoes with the activated JAK-STAT pathway . Plasmodium proliferation was reduced in the latter , indicating JAK-STAT involvement in anti-Plasmodium defense in this infection model . | [
"Abstract",
"Introduction",
"Results/Discussion",
"Materials",
"and",
"Methods"
] | [
"vector",
"biology",
"biology",
"microbiology"
] | 2011 | Transcriptome Analysis of Aedes aegypti Transgenic Mosquitoes with Altered Immunity |
Hepatitis B virus ( HBV ) persistence is facilitated by exhaustion of CD8 T cells that express the inhibitory receptor programmed cell death-1 ( PD-1 ) . Improvement of the HBV-specific T cell function has been obtained in vitro by inhibiting the PD-1/PD-ligand 1 ( PD-L1 ) interaction . In this study , we examined whether in vivo blockade of the PD-1 pathway enhances virus-specific T cell immunity and leads to the resolution of chronic hepadnaviral infection in the woodchuck model . The woodchuck PD-1 was first cloned , characterized , and its expression patterns on T cells from woodchucks with acute or chronic woodchuck hepatitis virus ( WHV ) infection were investigated . Woodchucks chronically infected with WHV received a combination therapy with nucleoside analogue entecavir ( ETV ) , therapeutic DNA vaccination and woodchuck PD-L1 antibody treatment . The gain of T cell function and the suppression of WHV replication by this therapy were evaluated . We could show that PD-1 expression on CD8 T cells was correlated with WHV viral loads during WHV infection . ETV treatment significantly decreased PD-1 expression on CD8 T cells in chronic carriers . In vivo blockade of PD-1/PD-L1 pathway on CD8 T cells , in combination with ETV treatment and DNA vaccination , potently enhanced the function of virus-specific T cells . Moreover , the combination therapy potently suppressed WHV replication , leading to sustained immunological control of viral infection , anti-WHs antibody development and complete viral clearance in some woodchucks . Our results provide a new approach to improve T cell function in chronic hepatitis B infection , which may be used to design new immunotherapeutic strategies in patients .
Hepatitis B virus ( HBV ) infection evolves into a chronic liver disease and leads to severe sequelae in about 5% of infected adults and in a larger proportion of children . It is estimated that approximately 400 million people are chronically infected with HBV worldwide . There are two types of antiviral therapies currently available for chronic HBV: treatment with pegylated interferon alpha ( PEG-IFNα ) and nucleot ( s ) ide analogues , such as entecavir ( ETV ) and tenofovir . However , treatment with PEG-IFNα leads to a sustained antiviral response in only about 30% patients and is associated with side effects . The introduction of PEG-IFNα in combination with nucleoside analogues did not significantly increase the rate of sustained responders [1] , [2] . Although treatment with nucleoside analogues improves the clinical condition of chronic HBV patients , it is hampered by emergence of drug resistance mutations , and rebounding viremia after cessation of antiviral therapy [3] , [4] . Therefore , alternative strategies to treat chronic HBV infection are urgently needed . Persistent HBV infection is associated with functional exhaustion of virus-specific CD8 T cells [5] . This defect in virus-specific T cells is one of the primary reasons for the inability of the host to eliminate the persisting pathogen . Therefore , therapeutic vaccination , which aims to enhance the patient's own antiviral cellular immune response , has been considered as an alternative therapy . However , the efficacy of such strategies in patients has so far been disappointing [6] , [7] , [8] . Recent work suggests that the high viral load at the time of vaccination might explain the inefficient responses to therapeutic vaccination [9] , [10] . Thus , it is important to develop a therapeutic vaccine strategy which could effectively boost endogenous T cell responses to control persistent viral infections . Recent studies in chronic virus infection models indicate that the interaction between the inhibitory receptor programmed death-1 ( PD-1 ) on lymphocytes and its ligands plays a critical role in T-cell exhaustion [11] , [12] , [13] , [14] . In various human chronic infections , including HBV , high PD-1 levels are expressed by virus-specific T cells , and improvement of the T-cell function has been obtained in vitro by inhibition of the PD-1/PD-ligand 1 ( PD-L1 ) interaction [15] , [16] , [17] , [18] , [19] , [20] , [21] . Moreover , in vivo blockade of PD-1/PD-L1 pathway has successfully been applied in mice persistently infected with lymphocytic choriomeningitis virus ( LCMV ) to restore the antiviral function of exhausted T cells , and hence improved the effect of the therapeutic vaccination [11] , [22] . We have previously demonstrated that therapeutic DNA vaccines in combination with an antiviral nucleoside analogue result in a prolonged suppression of WHV replication in chronically WHV infected woodchucks [23] . Recently , we have also demonstrated that in vitro blockade of the woodchuck PD-1/PD-L pathway could restore the T cell functions in chronic WHV infection [24] . In this study , we examined whether in vivo blockade of the PD-1 pathway in combination with antiviral nucleoside analogue treatment and therapeutic vaccination could enhance CD8 T cell immunity and lead to the resolution of chronic WHV infection in the woodchuck model . Persistently WHV-infected woodchucks were first treated with antiviral drug ETV to decrease the viral replication , and then received therapeutic vaccination and PD-L1 antibody treatment . This combinatorial therapeutic vaccination potently enhanced WHV-specific CD8 T cell responses , resulted in absence of WHV DNA in plasma and seroconversion to anti-WHs in two animals . However , residual WHV replication was still detectable in the liver of some animals .
The complete coding region of woodchuck PD-1 was obtained by reverse transcription polymerase chain reaction ( RT-PCR ) and subjected to sequence analysis . A comparison of the woodchuck PD-1 sequence revealed a high homology at the nucleotide ( nt ) and amino acid ( aa ) levels to the counterparts of mammalian species ( Table 1 ) . The length of putative woodchuck PD-1 protein is 288 aa residues , which shows the typical features of a membrane protein , an extracellular domain with a Ig-V-like region , a trans-membrane domain and a cytoplasmic domain ( Figure 1A ) . Importantly , the cytoplasmic domain of woodchuck PD-1 contains two highly conserved structural motifs , an immunoreceptor tyrosine-based inhibition motif ( ITIM ) and an immunoreceptor tyrosine-based switch motif ( ITSM ) ( Fig . 1B ) . The ITSM of PD-1 is believed to be essential for the inhibitory function of mouse PD-1 and human PD-1 [25] , [26] . Therefore , this result indicates that woodchuck PD-1 also behaves as an inhibitory molecule in the woodchuck immune system . It has been reported that PD-1 expression is up-regulated on HBV-specific CD8 T cells in the early phase of acute HBV infection [27] . Therefore , we first determined the kinetics of PD-1 expression on CD8 T cells ( CD3+ CD4− ) throughout the course of acute WHV infection in the woodchuck model . Four adult woodchucks were inoculated with 1×107 or 1×109 WHV genome equivalents and all of them went through an acute resolving infection . PD-1 expression on CD8 T cells was significantly up-regulated in acute WHV infection and reached its peak when the viremia started to decline . Following the successful viral clearance , PD-1 expression continuously decreased to the levels detected prior to WHV infection ( Figure 2 ) . One woodchuck ( A2 ) with very short and low viremia showed almost no increase of PD-1 expression on CD8 T cells ( Figure 2 ) . These data clearly indicate that the expression of PD-1 on CD8 T cells is associated with WHV viremia in acute resolving WHV infection . Next , we examined the level of PD-1 expression on PBMCs and T cells in woodchucks with chronic WHV infection . The PD-1 mRNA level of PBMCs and the percentage of PD-1+ CD8 T cells in chronic carriers were significantly higher than that of naïve woodchucks or resolvers ( Figure 3A ) . The mean fluorescence intensity ( MFI ) of PD-1 expression on CD8 T cells in chronic carriers was also significantly increased . Indeed , a distinct population of CD8 T cells which expressed extremely high levels of PD-1 ( PD-1hi ) was observed in the PBMCs of those woodchucks ( Figure 3B ) . We also compared PD-1 expression on PBMCs and CD8 T cells prior to antiviral treatment and at week 6 of antiviral treatment . Before initiation of ETV therapy , high levels of PD-1 expression were detected on both PBMCs and CD8 T cells in all subjects analysed . Antiviral treatment resulted in dramatic decline of detectable serum viral load , coincident with decrease of the PD-1 expression on PBMCs and CD8 T cells at both mRNA and protein levels ( Figure 3C ) . The MFI of PD-1 on CD8 T cells also decreased dramatically , and the PD-1hi CD8 T cells almost vanished after ETV treatment ( Figure 3D ) . These data indicate that high levels of antigen in chronic WHV infection may drive continuous high-level expression of PD-1 on CD8 T cells . The suppression of WHV replication resulted in a relatively lower level of PD-1 expression , which may facilitate the restoration of T cell function . The PD-1 expression on CD4 T cells was quite low in contrast to that on CD8 T cells in woodchucks with chronic WHV infection . Only a small number of CD4 T cells were found PD-1 positive ( Figure 3E ) . Antiviral ETV treatment showed no influence on PD-1 expression on CD4 T cells ( Figure 3F ) . To evaluate the effect of in vivo PD-L1 blockade during chronic WHV infection , a triple combination therapy strategy combined of antiviral treatment , therapeutic vaccination and PD-L1 antibody blocking was designed ( Figure 4 ) . The antiviral drug ETV was administered for 28 weeks to suppress the WHV replication . Starting from week 12 , animals received subsequently 12 intramuscular immunizations with DNA plasmids , expressing WHV core antigen ( WHcAg ) and surface antigen ( WHsAg ) . For PD-1/PD-L1 pathway blockade , woodchucks were treated with rabbit polyclonal PD-L1 blocking antibody ( αPD-L1 ) 3 times in week 24 . In total 12 chronically WHV-infected woodchucks were included in this experiment and were divided into four differently treated groups ( Figure 4 ) . Firstly , we longitudinally monitored individual woodchucks for WHcAg-specific and WHsAg-specific T cell responses by flow cytometric CD107a assay . Consistent with our previous study [28] , [29] , the WHcAg-specific and WHsAg-specific degranulation were not detectable in the chronic carriers without treatment or with only ETV treatment . Therapeutic vaccination in combination with antiviral treatment could induce a slight expansion of the WHcAg-specific CD8 T cell population in those animals after week 20 . Importantly , in vivo PD-L1 blockade in the vaccinated woodchucks resulted in a significant increase in the WHcAg-specific CD8 T cell response ( Figure 5A ) . The enhancement of WHcAg-specific CD8 T cell response in woodchucks of triple combination therapy group was observed immediately after αPD-L1 administration ( one week ) . Moreover , this effect was sustained even after cessation of the αPD-L1 treatment . At week 38 , which was 14 weeks after the cessation of the αPD-L1 treatment , strong WHcAg-specific T cell responses could still be detected in αPD-L1 treated woodchucks ( Figure 5B ) . Compared to the WHcAg-specific T cell response , the WHsAg-specific CD8 T cell response induced by DNA vaccination in ETV treated woodchucks was weak . Additional PD-L1 blockade did not result in any significant increase in the WHsAg-specific CD8 T cell response ( Figure S1 ) . We also longitudinally monitored individual woodchucks for WHcAg-specific responses by 2[3H]-adenine-based proliferation assay . In response to WHV core protein stimulation , no proliferation of PBMCs was observed in the chronic carriers without treatment or with only ETV treatment , and only weak proliferation of PBMCs was observed in woodchucks received DNA vaccination in combination with ETV treatment . In contrast to that , woodchucks received additional PD-L1 blockade showed strong proliferation of their PBMCs in response to WHV core protein stimulation after the αPD-L1 administration ( Figure 5C and 5D , Table S1 ) . To examine whether our triple combination therapy resulted in better control of viral infection , we monitored the WHV viremia , WHsAg levels in serum , development of anti-WHs , viral replication in the liver and liver transaminase levels of the treated woodchucks . The basal WHV DNA loads in the serum prior to ETV therapy in WHV chronic carriers enrolled in the experiment ranged from 4 . 27×106 GE/ml to 3 . 73×1011 GE/ml . Consistent with our previous study [30] , the treatment with 0 . 5 mg ETV/kg body weight led to a significant reduction of the serum WHV DNA concentrations up to >6 log ( Figure 6A ) . In woodchucks that received only ETV , the viral rebound was observed after week 24 , and WHV DNA concentrations returned to the pre-treatment levels shortly after cessation of ETV treatment . In woodchucks that received ETV treatment in combination with therapeutic vaccination , the relapse of viremia was delayed and the serum WHV DNA levels remained undetectable in all animals till week 28 . However , the viral rebound occurred in these woodchucks after cessation of ETV treatment , and WHV DNA concentrations returned to the pre-treatment levels at the end of the observation period . In contrast to that , the serum WHV DNA levels of woodchucks which received additional αPD-L1 treatment remained undetectable at almost all time points till the end of the observation period ( Figure 6A ) . There was a relapse of viremia in only one αPD-L1 treated woodchuck ( EDA3 ) after week 36 . This woodchuck had developed massive liver tumors during the ETV treatment ( Figure S2 ) and had to be sacrificed before the end of the observation period . The initial levels of WHsAg in the sera varied between the individual animals and were ranging from 45 . 8 µg/ml to 698 . 3 µg/ml . A significant decrease of WHsAg was observed in all woodchucks during the ETV treatment . In the woodchucks treated with only ETV or ETV in combination with DNA vaccination , WHsAg concentrations returned to the pre-treatment levels shortly after cessation of ETV treatment . In contrast to that , woodchucks that received additional αPD-L1 treatment showed no relapse of WHsAg in the sera after cessation of ETV treatment ( Figure 6B ) . Notably , woodchuck EDA1 which was negative for WHV DNA in both serum and liver was also WHsAg negative ( below limit of detection after week 28 ) . However , WHsAg of the other 2 woodchucks ( EDA2 and EDA3 ) with αPD-L1 treatment remained at very low levels till the end of our observation ( Figure S3 ) . No anti-WHs antibody was detectable in woodchucks of the ETV treatment only group and of the DNA vaccination group ( data not shown ) . In contrast , 2 woodchucks of the PD-L1 blockade group developed anti-WHs antibodies . These 2 woodchucks became anti-WHs antibody positive 2 weeks after the PD-L1 blockade . The titer of anti-WHs antibodies in woodchuck EDA1 was high , and remained positive till the end of the observation . The titer in the other woodchuck EDA3 was relatively lower , and dropped very quickly accomplished with a massive hepatocellular carcinoma ( HCC ) development . Woodchuck EDA2 did not develop a detectable level of anti-WHs antibody within the observation period up to week 42 ( Figure 6C ) . Anti-WHc antibodies in the sera of treated woodchucks were also examined by semi-quantitative ELISA . All chronic carriers showed fluctuant anti-WHc levels in the sera throughout the whole observation period . No significant change of serum anti-WHc levels was observed after PD-L1 antibodies administration in treated woodchucks ( Figure S4 ) . In addition to serum levels of WHV DNA , the WHV replication in the livers of treated woodchucks was evaluated by Southern blot . Woodchucks of the untreated group C , the ETV treated only group E and the ETV plus DNA vaccination group ED showed obvious WHV replication in the liver at the end of the observation period up to week 42 . In contrast , woodchuck EDA1 and EDA2 of the triple combination therapy group did not show any WHV replication in the liver at week 42 . Woodchuck EDA3 also demonstrated a very low level of WHV replication ( Figure 6D ) . In addition , we analyzed the presence of WHV DNA in the liver by sensitive PCR . Consistent with the results of Southern blot , woodchuck EDA1 was negative for WHV DNA in the liver . Woodchuck EDA2 and EDA3 also showed reduced WHV DNA loads in the liver compared to those of woodchucks without αPD-L1 treatment ( Figure 6D ) . Realtime PCR was also performed to precisely quantify the amounts of WHV DNA in the liver of all animals . The lower detection limit of this realtime PCR is 1000 GE/ml . The liver WHV DNA loads of animals without αPD-L1 treatment ranged from 1 . 34×1010 GE/ml to 4 . 33×1010 GE/ml . In contrast , in animals received additional αPD-L1 treatment , two woodchucks ( EDA2 and EDA3 ) showed 2–4 log reduction of the liver WHV DNA concentrations , and one woodchuck ( EDA1 ) remained undetectable for liver WHV DNA ( Figure 6D ) . Covalently closed circular DNA ( cccDNA ) is responsible for persistence of infection in the natural course of chronic HBV infection and during prolonged antiviral therapy . Therefore , the effect of the different treatment regimens on the cccDNA pool in the liver was examined . Consistent with the results of WHV DNA detection in the liver , woodchucks of the untreated group , the ETV treated only group and the ETV plus DNA vaccination group showed obvious WHV cccDNA presence in the liver at the end of the observation period up to week 42 . In contrast , WHV cccDNA remained undetectable in the liver of woodchuck EDA1 . Woodchuck EDA2 and EDA3 also showed reduced cccDNA loads in the liver compared to those of woodchucks without αPD-L1 treatment ( Figure 6E ) . Liver transaminase GOT levels were measured during ETV treatment , immunizations and antibody treatment . In most of the treated woodchucks , GOT flares were apparently associated with a reduction of WHV DNA loads . This is likely due to a partial restoration of immunological functions , similar to the reported cases in clinical trials of ETV . Although PD-L1 blockade strongly restored the T cell function in antibody treated woodchucks , GOT levels of these animals after antibody injection did not increase ( Figure 7 ) . GOT flare was observed after PD-L1 blockade in only one woodchuck ( EDA3 ) , but this elevation is more likely due to the development of massive HCC in this animal . We also evaluated the effect of our triple combination regimen on the induction of WHV-specific CD8 T cell response and the control of WHV replication in 1217 WHV transgenic ( tg ) mice carrying a 1 . 3 fold overlength WHV transgenome [30] . The antiviral drug ETV was daily administered for 10 days to suppress the WHV replication in mice . For DNA vaccination , mice received two intramuscular immunizations with DNA plasmid expressing WHcAg on day 7 and day 21 . For PD-L1 blockade , rat anti-mouse PD-L1 antibody was administered i . p . 5 times every 3 days , beginning on the day 14 . Four differently treated groups were included in this experiment ( Figure S5A ) . No increase of WHcAg-specific CD8 T cell response in the liver was observed in untreated , only ETV treated , and ETV plus PD-L1 blockade treated mice . In contrast , mice received ETV treatment in combination with DNA vaccination and PD-L1 blockade showed significantly increased WHcAg-specific CD8 T cell response in the liver ( Figure S5B ) . We also monitored the change of WHV DNA levels in the serum of all mice . As expected , no significant change in the WHV DNA load throughout the experiment was observed in the untreated mice . Decrease of WHV DNA load was observed in all mice which received ETV treatment after day 7 . On day 28 , WHV DNA concentrations in only ETV treated mice returned to the pre-treatment levels . In contrast , significantly reduced WHV DNA levels were measured in mice received ETV plus PD-L1 blockade treatment and mice received triple combination therapy ( Figure S5C ) . These results indicate that the triple combination regimen is able to break the immune tolerance against WHV antigens in WHV tg mice and leads to the suppression of viral replication .
The aim of therapeutic vaccination in chronic HBV infection is to elicit cellular immune responses specifically against HBV and thus reduce the viral burden . However , strategies investigated so far were often hampered by weak T cell responses observed after immunization , suggesting a strong need for alternative strategies to enhance T cell functions during chronic HBV infection . In this study , we evaluated the efficacy of antiviral treatment in combination with DNA vaccination and blockade of the inhibitory PD-1/PD-L1 pathway , which is a key factor of inducing T cell exhaustion during chronic viral infections in woodchucks chronically infected with WHV . Compared to nucleoside analogue treatment in combination with only therapeutic vaccine , additional PD-L1 blockade enhanced expansion and improved the function of WHcAg-specific CD8 T cells in woodchucks persistently infected with WHV . This triple combination therapy resulted in a prolonged suppression of WHV replication as compared to nucleoside analogue treatment alone or in combination with therapeutic vaccine . These results indicate that this may be a novel strategy for effective therapeutic vaccination of chronic HBV infection with subsequent complete recovery . The quantity of antigen to which the immune system is exposed can induce different degrees of functional impairment of antiviral T cells , up to physical T cell deletion [15] , [31] . This mechanism may play an important role in the HBV-specific T-cell hyporesponsiveness of chronic HBV infection , because high concentrations of antigens are constantly present at all stages of the infection . Studies have already confirmed that the functions of HBV-specific T cells are more profoundly inhibited in the presence of high viremia [16] , [32] , [33] . Therefore , when viral load is high and T cell dysfunction is severe , therapeutic vaccination alone can not induce a strong HBV-specific T cell response . Actually , poor responses and limited antiviral effects have been reported for the therapeutic vaccination of chronic HBV patients [7] , [34] , [35] , [36] . In contrast , T cell exhaustion may be less severe at lower viral loads , and subsequent therapeutic vaccination may be more effective [9] . Studies of SIV infection support this concept because therapeutic vaccination is more effective after antiviral therapy [37] , [38] . Therefore , the potent antiviral drug ETV was used in this study to strongly decrease both WHV DNA and WHsAg , which facilitated the induction of WHV-specific immune responses by DNA vaccination . Another benefit of effective antiviral treatment is that it decreases PD-1 expression on CD8 T cells , which makes the rescue of exhausted T cell by PD-1/PD-L1 blockade possible . It has been recently clarified that the proportion of CD8 T cells expressing PD-1 and the levels of PD-1 on virus-specific T cells are strongly correlated with viral load in the plasma [17] , [19] , [20] . Antiretroviral treatment resulted in the dramatic decline of plasma viral load , coincident with a decrease in the PD-1 expression level on virus-specific CD8 T cells [17] , [19] . In line with this , a better restoration of T cell functions upon in vitro anti-PD-L1 treatment was observed in chronic HBV patients with lower viremia [39] . In this study , we could not determine the PD-1 expression in WHV-specific T cells due to the lack of available woodchuck tetramers . Nevertheless , a significant positive correlation between the viral load and the PD-1 expression on total CD8 T cells in chronic WHV infection was observed . Both the proportion of PD-1+ CD8 T cells and the levels of PD-1 expression on CD8 T cells were significantly higher in the woodchucks with chronic WHV infection compared to naïve animals and resolvers . More importantly , during ETV treatment of those chronic carriers , a reduction of serum viral load was correlated with a dramatic decrease in the level of PD-1 expression on CD8 T cells . The PD-1hi population of CD8 T cells observed in the chronic carriers completely disappeared after ETV treatment . It was reported that PD-1hi CD8 T cells were not able to restore their function by PD-1/PD-L1 blockade in LCMV infection [40] . Therefore , the decrease of PD-1 expression on CD8 T cells caused by antiviral treatment may have increased the efficacy of in vivo PD-1/PD-L1 blockade on restoring CD8 T cell function . In contrast , treatment with PD-L1 antibody alone showed no effect on either enhancing T cell function or controlling WHV viral replication in chronic carriers ( Figure S6 ) . Due to the high costs of woodchucks and long experimental periods , we could not enroll enough woodchucks to establish exhaustive control groups ( such as empty plasmid vaccine , et al ) for this study . However , we did originally enroll one woodchuck as a control for examining irrelevant rabbit polyclonal antibody . This woodchuck received entecavir treatment , DNA vaccination and rabbit polyclonal IgG ( Europa Bioproducts Ltd , Cambridge , England ) injection . Compared to the woodchucks which received PD-L1 antibodies treatment , this woodchuck showed no significant increase in WHcAg-specific CD8 T cell responses in week one and two after antibody injection ( Figure S7 ) . Unfortunately , this woodchuck developed liver tumors during the experiment and died 4 weeks after antibody injection ( week 28 ) , which made us unable to tell the long term outcome of this animal . One major concern of this triple regimen is whether antiviral treatment in combination with PD-L1 blockade alone is already enough for inducing virus-specific CD8 T cell response and controlling viral replication . To address this question , we examined the impact of the ETV treatment in combination with PD-L1 blockade alone on WHcAg-specific CD8 T cell response and the WHV replication in WHV tg mice . In comparison to the triple treatment , the dual treatment ( ETV plus PD-L1 blockade ) failed to induce detectable WHcAg-specific CD8 T cell response in the liver . Nevertheless , the dual treatment exhibited better effect on suppressing WHV replication than ETV treatment alone ( Figure S5 ) . This result indicates that PD-L1 blockade alone can give additive effect to ETV treatment on suppressing WHV replication , but DNA vaccination seems to be essential for achieving a robust and persistent viral-specific CD8 T cell response in chronic hepatitis infection . DNA vaccination is an attractive method for the treatment of chronic infection and tumors . Clinical studies provided proof of concept that DNA vaccines could safely induce immune responses ( albeit low-level responses ) in humans . In this study , in total 12 DNA vaccinations were performed to elicit WHV-specific CD8+ T-cell responses to in WHV chronic carriers . Many improvements have been incorporated into the new DNA vaccine strategies , including optimization of antigen expression on a per cell basis and improved formulation to enhance and direct immune responses [41] . We have also showed very recently that a DNA prime-adenovirus ( AdV ) boost vaccination regimen could elicit strong and specific CD8+ T-cell responses to WHcAg in mice and woodchucks [30] . This strategy induces more robust T cell responses than the conventional DNA vaccine does with even less injections . Unfortunately , this new regimen was not available when we performed the presented PD-L1 blockade experiment . Therefore , enrolling more efficient DNA vaccine protocols such as the DNA prime-AdV boost protocol to our current combination therapy strategy may be an interesting option in future trials . In this study and our previous study [30] , we immunized ETV treated woodchucks with two DNA plasmids which express WHcAg and WHsAg respectively . In contrast to the WHcAg-specific T cell response , the WHsAg-specific CD8 T cell response induced by DNA vaccination in ETV treated woodchucks was weak . This could be due to 2 possible reasons: ( 1 ) The WHcAg expressing plasmid pCGWHc is more efficient at antigen expression than the WHsAg expressing plasmid pWHsIm [29] . Improved WHcAg expression may lead to the induction of a more T cell response in vivo . ( 2 ) The ongoing secretion of high amounts of WHsAg in the circulation of woodchucks may have hampered the induction of a vigorous WHsAg-specific CD8 T cell response . Nevertheless , two woodchucks of the PD-L1 blockade group developed anti-WHs antibodies , indicating that WHsAg vaccine may also contribute to the control of WHV infection . In addition , a vigorous T-cell response against HBcAg is crucial for the resolution of the infection but is predominantly absent in chronic hepadnaviral infections . Thus , using T cell vaccines targeting the core protein seems to be essential for a potent therapeutic strategy . The observation in this study suggests the conclusion that the T cell tolerance against WHcAg is easier to be broken than the tolerance against WHsAg in chronic carriers by DNA vaccines , and the induced robust WHcAg-specific CD8 T cell response contributes greatly to the control of WHV infection . A major concern of using PD-1/PD-L1 pathway blockade treatment of chronic HBV infection is that this treatment may lead to severe liver inflammation and may result in fulminant hepatitis . However , we did not observe any increase of liver inflammation in the animals after the PD-L1 antibody treatment by monitoring GOT levels . In most of the treated woodchucks , the GOT flares appeared at the stage of ETV treatment alone or the beginning of DNA vaccination . Consistent with our results , recent studies of clinic phase 1 trials have shown the safety of using anti-PD-L1 and anti-PD1 antibodies in cancer patients treatment [42] , [43] . The negative regulation of T-cell function involves numerous receptor and ligand interactions in separate cellular compartments at different phases of the immune response . Recent studies have suggested that multiple inhibitory receptors , such as CTLA-4 [44] , TIM-3 [45] and LAG-3 [46] , have played important roles in T-cell exhaustion during persistent HBV infection . These observations suggest that there may be a synergy between blockade of the various inhibitory pathways and encourage the examination of combinatorial strategies for treatment of woodchucks with chronic WHV or to later patients with chronic HBV in the future . Moreover , it has been shown very recently that activating Toll-like receptor 7 signaling can induce clearance of HBV-infected cells and prolonged suppression of HBV replication in chronically infected chimpanzees [47] . The mechanism of activating immune system and suppressing HBV by using TLR agonist significantly differs from that of PD-1/PD-L1 pathway blockade in T cells . Therefore , combining TLR7 agonist and PD-L1 antibody treatment may activate different arms of immune system and may achieve synergistic effects on control of chronic HBV infection . WHV and HBV show a marked similarity in the virion structure , genomic organization , mechanism of replication , and host immune responses during infection and recovery . Woodchucks chronically infected with WHV develop progressively severe hepatitis , which is remarkably similar to what is observed in chronic HBV patients . Likewise , the results of antiviral drug efficacy and toxicity studies in the woodchucks chronically infected with WHV are predictive for responses of chronic HBV patients . However , we are aware that we can not predict whether the triple regimen of treatment described in this paper will be successful in patients with chronic HBV . In conclusion , our study for the first time demonstrates that antiviral treatment in combination with therapeutic vaccination and PD-L1 blockade potently boosts virus-specific CD8 T cell responses and promotes viral control during chronic hepadnavirus infection . Further studies in a larger number of chronic carriers need to be conducted for investigating whether the rate of virus elimination could be significantly enhanced . Our results may lay a foundation for initiating a dual therapy ( combination of nucleotide analogue and PD-L1 blockade ) or a triple therapy ( combination of nucleotide analogue , DNA vaccine and PD-L1 blockade ) in chronic HBV patients .
All animal experiments were conducted in accordance with the Guide for the Care and Use of Laboratory Animals and were approved by the local Animal Care and Use Committee ( Animal Care Center , University of Duisburg-Essen , Essen , Germany and the district government of Düsseldorf , Germany; permission numbers G1303/12 and G1304/12 ) . The experiments were performed under ketamine-xylazine anesthesia , and all efforts were made to minimize suffering . The woodchucks ( Marmota monax ) were purchased from North Eastern Wildlife ( Harrison , ID ) . WHV chronic carriers were captive-born 1-year-old woodchucks neonatally infected with WHV . Persistence of WHV infection was based on the consecutive detection of WHV DNA and WHsAg in serum starting at 3 months of age . WHV transgenic mice lineage carrying WHV wild-type ( strain 1217 ) was created on C57BL/6 background ( genotype H-2b/b ) and previously characterized [30] . Animals were maintained according to the guidelines of the animal facility at the University Hospital Essen . Total RNA was extracted from fresh isolated woodchuck PBMCs using the TRIZOL reagent ( Invitrogen ) according to the manufacturer's instructions and subjected to RT-PCR for amplification of cDNAs of woodchuck PD-1 . The primers used for cloning woodchuck PD-1 cDNAs were: 5′-ATG CAG GGC CGG TGG-3′ , 5′-GCC TGG AAG CTG GCC T-3′ . The specific PCR fragments were cloned into pMD18-T TA vector ( TAKARA ) and subjected to DNA sequencing . The secondary structure of protein sequences data was predicted by online service of the Swiss Institute of Bioinformatics ( http://swissmodel . expasy . org ) . Sequences of PCR primer pairs for woodchuck PD-1 and β-actin are as follows: PD-1 ( forward , 5′-AGC CCC AGC AAG CAG AAC-3′; reverse , 5′-GCC CCG CAG AGG TAG AGG-3′ ) and β-actin ( forward: 5′-TGG AAT CCT GTG GCA TCC ATG AAA C-3′; reverse , 5′-TAA AAC GCA GCT CAG TAA CAG TCC G-3′ ) . The quantification of woodchuck PD-1 mRNAs was performed by real time RT-PCR using QuantiFast SYBR Green RT-PCR Kit ( Invitrogen , Karlsruhe , Germany ) on a Light Cycler . WHV DNA was quantified by real-time PCR as described previously [28] . Cell-surface and intracellular staining for flow cytometry analysis was performed using BD Biosciences or eBioscience reagents as described previously [28] , [48] . For woodchuck PD-1 and CD3 staining , an anti-mouse PD-1 FITC-conjugated antibody ( clone J43 , eBioscience ) and an anti-rat CD3 PE-conjugated antibody ( clones G4 . 18 , BD Biosciences ) were used . Data were acquired using a FACSCalibur flow cytometer ( BD Biosciences , Heidelberg , Germany ) and analyzed using FlowJo software ( Tree Star , Inc . , Ashland , Oregon ) . Cell debris and dead cells were excluded from the analysis based on scatter signals and 7-Amino-actinomycin D fluorescence . Chronically WHV-infected woodchucks were treated for 28 weeks with the nucleoside analogue entecavir ( ETV , Bristol-Myers Squibb , New York , NY ) . Initially , the drug was administered for 12 weeks by using osmotic pumps ( DURECT , Cupertino , CA ) which were implanted surgically under the skin of the animals . The pump releases subcutaneously approximately 0 . 2 mg of ETV per day . Pumps were exchanged every 4 weeks and overall 3 pumps were implanted for each woodchuck . From week 10 to 28 , subcutaneous injections of 1 . 5 mg of ETV were performed weekly . The protocol of the immunization of woodchucks was described previously [49] . The immunizations with previously constructed DNA plasmids expressing WHcAg and WHsAg [29] were performed by intramuscular injection . A week prior to the injection of plasmids , 500 µl of cardiotoxin ( Latoxan , Valence , France ) at a concentration of 10 µM in PBS was injected into M . tibialis cranialis of woodchucks . Woodchucks were vaccinated 12 times by intramuscular injection of 500 µl of plasmids into each M . tibialis cranialis at the indicated time points . For WHV transgenic mice immunization , ten to twelve weeks old sex-matched groups of mice were pretreated by intramuscular injection of 50 µl of cardiotoxin into Tibialis anterior muscle one week before the plasmid immunization . Animals were then intramuscularly vaccinated twice with 100 µg of pCGWHc ( 50 µg per muscle ) at two weeks interval . Polyclonal rabbit anti-woodchuck PD-L1 antibody was generated by our lab as described previously [24] . At week 24 of the ETV therapy , woodchuck PD-L1 antibody ( 25 mg/kg ) in PBS was intravenously injected to 3 woodchucks . Administration of rabbit isotype antibody to 1 woodchuck served as a control . Antibodies were injected every 2 days , and were overall injected 3 times . For mice PD-L1 blockade , 200 µg of rat anti-mouse PD-L1 antibody ( 10F:9G2 ) was administered i . p . 5 times every 3 days . CD107a degranulation assay was performed as described previously [28] . Briefly , woodchuck PBMCs were separated by Ficoll density gradient centrifugation and stimulated with 2 µg/ml WHcAg-derived epitope c96-110 ( KVRQSLWFHLSCLTF ) . Unstimulated cells and cells stimulated with 2 µg/ml of a control CMV-derived peptide ( YILEETSVM ) served as negative controls . After 3 days of cultivation , cells were restimulated with corresponding peptides and stained for CD107a for 5 hours . Data were acquired using a FACSCalibur flow cytometer . Antigen-specific proliferation of woodchuck PBMCs was determined by 2[3H]-adenine-based assay as described previously [49] . Briefly , 5×104 PBMCs were stimulated with 5 µg/ml purified WHcAg protein for 5 days . Unstimulated cells served as a negative control . Afterwards , cells were labelled with 1 µCi of 2[3H]-adenine ( Hartmann Analytic , Braunschweig , Germany ) for 16 h and collected using a cell harvester ( Perkin Elmer , Waltham , MA ) . Results for triplicate cultures are presented as a mean stimulation index ( SI ) . SI is calculated with the formula: ( stimulated cpm - blank cpm ) / ( unstimulated cpm - blank cpm ) . Woodchuck anti-WHs antibodies were detected by enzyme-linked immunosorbent assay ( ELISA ) as described previously [49] , [50] . WHV DNA was quantified by real-time PCR using Platinum SYBR Green Kit ( Invitrogen ) as described previously [28] . Total DNA from liver samples of chronically WHV-infected woodchucks was extracted using the QIAamp Tissue Kit ( Qiagen , Hilden , Germany ) according to the manufacturer's instructions . WHV replication intermediates were analyzed by Southern blot hybridization as described previously [51] , [52] . WHV DNA was determined by both qualitative PCR and quantitative real-time PCR using Platinum SYBR Green Kit ( Invitrogen ) . WHV cccDNA was determined by PCR as described previously [53] . Serum WHsAg concentration was determined by electroimmunodiffusion as described previously [23] . The glutamic oxaloacetic transaminase ( GOT; also known as aspartate transaminase , AST ) level was quantified according to the standard diagnostic procedure at the Central Laboratory of University Hospital Essen . The values above 50 IU ( international units ) per millilitre were considered as elevated . Statistics comparing two groups were done using the nonparametric t test . When more than two groups were compared , a one-way ANOVA was used with a Tukey posttest ( GraphPad Prism software; GraphPad , San Diego , CA ) . | Chronic hepatitis B virus ( HBV ) infection is still one of the major public health problems . Two billion people worldwide have been infected with HBV , of whom more than 360 million developed chronic infection . Every year , approximately one million of these individuals will die from HBV-associated liver diseases such as cirrhosis and hepatocellular carcinoma ( HCC ) . Treatment of chronic hepatitis B remains a clinical challenge , and alternative strategies to treat chronic HBV infection are urgently needed . Here , we designed a new combination strategy to enhance the patient's own antiviral immune response and to achieve long-term viral suppression . The therapeutic effect of our combination therapy strategy for chronic hepadnaviral infection was tested in the woodchuck model . We demonstrated that our novel combination therapy could elicit potent antiviral immune response and achieved a strong antiviral effect , leading to sustained immunological control of chronic hepadnaviral infection and complete viral clearance in treated woodchucks . The results of this study may have an impact on clinical trials of the immunotherapy in chronically HBV-infected patients . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [] | 2014 | Enhancing Virus-Specific Immunity In Vivo by Combining Therapeutic Vaccination and PD-L1 Blockade in Chronic Hepadnaviral Infection |
Gammaherpesvirus cyclins have expanded biochemical features relative to mammalian cyclins , and promote infection and pathogenesis including acute lung infection , viral persistence , and reactivation from latency . To define the essential features of the viral cyclin , we generated a panel of knock-in viruses expressing various viral or mammalian cyclins from the murine gammaherpesvirus 68 cyclin locus . Viral cyclins of both gammaherpesvirus 68 and Kaposi's sarcoma-associated herpesvirus supported all cyclin-dependent stages of infection , indicating functional conservation . Although mammalian cyclins could not restore lung replication , they did promote viral persistence and reactivation . Strikingly , distinct and non-overlapping mammalian cyclins complemented persistence ( cyclin A , E ) or reactivation from latency ( cyclin D3 ) . Based on these data , unique biochemical features of viral cyclins ( e . g . enhanced kinase activation ) are not essential to mediate specific processes during infection . What is essential for , and unique to , the viral cyclins is the integration of the activities of several different mammalian cyclins , which allows viral cyclins to mediate multiple , discrete stages of infection . These studies also demonstrated that closely related stages of infection , that are cyclin-dependent , are in fact genetically distinct , and thus predict that cyclin requirements may be used to tailor potential therapies for virus-associated diseases .
Gammaherpesviruses are oncogenic viruses that establish lifelong infection of the host . Primary gammaherpesvirus infection of healthy adult hosts results in an acute stage of lytic virus replication which is then cleared , with lifelong latent infection established primarily in B lymphocytes . A transient mononucleosis-like stage is associated with establishment of latent infection with Epstein Barr virus ( EBV ) and the murine gammaherpesvirus 68 ( gHV68 ) . The latent stage of infection is controlled by an active immune response , and immune deficient hosts suffer increased virus reactivation from latency and persistent infection ( evidenced by ongoing production of infectious virus ) , both of which are associated with disease . Viral cyclin genes are conserved among gamma-2-herpesviruses , including the human Kaposi's sarcoma-associated herpesvirus ( KSHV ) , and Epstein Barr virus ( EBV ) , a closely related human gammaherpesvirus , uses positional homologs to up regulate expression of host D-type cyclins . Cyclins are the regulatory partners of the catalytic cyclin dependent kinases ( cdks ) , which together regulate cellular DNA replication and cell division . Viral cyclins share the greatest sequence similarity to one another and to mammalian D-type cyclins , yet are functionally most similar to mammalian cyclins A and E [1]–[4] . Relative to mammalian cyclins , the viral cyclins confer increased kinase activity and demonstrate broader cdk binding and substrate specificity , as well as increased resistance to cyclin-dependent kinase inhibitors [5]–[10] . The viral cyclin ( v-cyclin ) protein of the mouse model gHV68 is abundantly expressed in lytic virus replication and in reactivation from latency [11] , and v-cyclin transcript is also detected in latently infected B cells [12] . The first gammaherpesvirus viral cyclin gene was described in 1992 [13] , since which time numerous activities of the viral cyclins have been discovered and proposed as important in gammaherpesvirus pathogenesis . However , to date , no study has addressed whether the unique biochemical features of the v-cyclin are essential to promote infection or if mammalian cyclins , with more restricted activities , are capable of promoting infection . This issue is particularly important given the increasing evidence that mammalian cyclins have an unexpected degree of plasticity and redundancy in promoting cell cycle progression [14] [15] , yet specific cyclins are required for cell- or tissue-specific functions [16] , [17] . The emerging picture of the mammalian cyclins in cell cycle , development and cancer present a compelling case for understanding the specific activities of the unique viral cyclins . While extensive biochemical characterization of viral cyclins revealed multiple unique characteristics of viral cyclins relative to mammalian cyclins when expressed in isolation , the precise function of the viral cyclins in the context of virus infection has only more recently been elucidated . In transgenic studies in which the viral cyclins are constitutively expressed in mice , both the gHV68 and KSHV viral cyclins are tumorigenic [11] , [18] . This observation , coupled with the known cell cycle promoting effects of the viral cyclins and viral cyclin expression in some gammaherpesvirus associated tumors , initially lead to a focus on the oncogenic effects of the viral cyclin during infection . However , given that gammaherpesvirus infection in healthy individuals rarely induces malignancy , the viral cyclin is very likely to have roles in promoting viral infection and pathogenesis . To rigorously assess the genetic contribution of the viral cyclin in the context of virus infection and gammaherpesvirus pathogenesis , we have made extensive use of the gHV68 mouse model and have now shown that the v-cyclin of gHV68 plays a critical role in several distinct aspects of virus infection . We demonstrated that virus production in acute pulmonary infection is dependent on the v-cyclin [19] . Additionally , we noted a dramatic decrease in the survival of persistently infected endothelial cells upon infection with v-cyclin-deficient virus [20] . Finally , we and others observed a profound defect in viral reactivation from latency in the absence of the v-cyclin [21] , [22] . The requirement for the v-cyclin is manifested in many disease states , that is , the v-cyclin-deficient virus is attenuated in lethal pneumonia [19] , arteritis [23] and chronic pulmonary disease [24] , chronic mortality in immune deficient mice [25] , and in atypical lymphoid hyperplasia [26] found in immune deficient mice and pathologically similar to EBV-induced post-transplant lymphoproliferative disease . In contrast , the v-cyclin is dispensable for viral replication , the establishment of latency [22] and the development of pulmonary lymphoma in immunodeficient mice [27] . To rigorously dissect the essential cyclin feature ( s ) required for the v-cyclin during virus infection , we generated a panel of recombinant viruses in which the v-cyclin of gHV68 was precisely replaced with the viral cyclin of KSHV ( k-cyclin ) or with multiple different mammalian cyclins . By testing the capacity of different viral and mammalian cyclins to substitute for the function ( s ) of the endogenous v-cyclin of gHV68 in known v-cyclin dependent parameters , we determined that the viral cyclins of gHV68 and KSHV were able to interchangeably fulfill all v-cyclin dependent parameters of infection . On the other hand , analysis of viral recombinants expressing mammalian cyclins revealed varying capacity to support v-cyclin dependent stages of infection . Unexpectedly , distinct and non-overlapping cyclins were capable of functioning in different stages of infection , an observation which allowed us to genetically separate reactivation from latency and viral persistence . In total , these studies demonstrate that the viral cyclins are uniquely multifunctional and mediate their complete function by possessing properties of multiple mammalian cyclins .
We generated a complete panel of recombinant viruses to genetically test cyclin requirements in promoting gammaherpesvirus infection ( Figure 1A and S1 , Table S1 ) . Using bacterial artificial chromosome mediate mutagenesis , we generated six viral recombinants in which different viral or mammalian cyclins precisely replaced the endogenous cyclin gene . This method placed different cyclins under control of the endogenous v-cyclin promoter and viral polyA signal to faithfully recapitulate the transcriptional regulation of this gene . To facilitate uniform and sensitive detection of cyclin expression among these recombinant viruses , a 3x-FLAG epitope tag was fused to the amino terminus of each cyclin [28]–[30] . The cyclins included in this recombinant panel were based on similarity in either sequence or function to the v-cyclin: 1 ) the gHV68 v-cyclin , 2 ) the viral cyclin of KSHV ( k-cyclin ) , 3 ) the mammalian cyclins D2 and D3 , based on sequence similarity [6] , [10] , [31] and their predominant expression in lymphocytes [32] , the major reservoir for gHV68 latency , and 4 ) the mammalian E and A cyclins based on structural and functional similarity [33] . Quantitative analysis of virally expressed cyclin mRNAs , via the shared 3x-FLAG sequence , demonstrated similar RNA expression of all 3x-FLAG tagged cyclins during virus infection at 12 and 48 hours post-infection . Further , all 3x-FLAG-cyclin RNAs were expressed at low levels 12 hours post-infection , and were abundant by 48 hours post-infection ( Figure 1B ) , consistent with the early-late gene kinetics previously established for the gHV68 v-cyclin [11] . These data demonstrated that the FLAG-tagged cyclins are equivalently transcribed during virus infection . Viral cyclin protein expression during infection was detectable by immunoflourescence at 12 and 24 hours post-infection ( Figure S2C ) , and demonstrated a similar and primarily nuclear/perinuclear pattern . Specificity of cyclin protein expression from each recombinant virus was verified by western analysis of independent duplicate infections using both FLAG- and cyclin-specific antibodies ( Figure 1C ) , and abundant expression of each cyclin protein was demonstrated during infection of fibroblasts ( Figure 1D ) and endothelial cells ( Figure S2A ) . Because viral and mammalian cyclins are presumed to function via binding of cellular cdks , we performed kinase interaction analyses for each of these viruses at 24 hours post-infection of two different cell types . Infection of both fibroblasts and endothelial cells demonstrated the expected interaction partners , with the viral cyclins binding cdks more efficiently than their cellular counterparts ( Figure S2B ) . Notably , the cdks associated with the viral cyclins were distinct from each other , and neither of the viral cyclins shared a common interaction profile with any of the mammalian cyclins during virus infection . Variation in relative protein abundance and in kinase binding of the 3x-FLAG cyclins is consistent with known differences in protein stability and partner preferences , and notably did not correspond to function during virus infection in subsequent studies . Finally , we previously showed that the v-cyclin is not required for virus replication in vitro [22] . To ensure that insertion of other cyclin genes did not alter viral replication , we compared virus replication of WT virus with the panel of cyclin recombinants in multiple cycles of replication and found indistinguishable replication among these viruses ( Figure 1E ) . Thus , replacement of the v-cyclin of γHV68 with other cyclin genes does not alter virus replication in vitro . As we recently reported , infection of immunodeficient mice with gHV-cycKO virus resulted in a significant defect in acute virus production in the lung and failed to cause the lethal pneumonia that results from WT infection and WT levels of acute virus production [19] . Therefore , to investigate the cyclin requirements for acute virus production and lung pathology , we infected IFN-g-/- mice with the panel of recombinant cyclin viruses . We tested gHV-cycK , the cyclin with the greatest overall similarity to the v-cyclin , gHV-cycD3 and D2 for sequence similarity and cell type relevance , and gHV-cycA and E for functional similarity . We infected IFN-g-/- mice with the recombinant cyclin viruses for 8 days , previously identified as the time at which virus titer and lung pathology differed most between WT and gHV-cycKO infection ( Figure 2A ) [19] . The severity of pathology , or acute pneumonia , was most profound following infection with the gHV-cycV , with similar morphology in the gHV-cycK-infected lungs ( interstitial and airway edema , hypercellularity , tissue condensation and severe inflammatory infiltrates marked by neutrophils; Figure 2B ) , and mice in these groups demonstrated hunched posture and ruffled fur at time of sacrifice . Less severe pathology was observed in lungs infected with viruses expressing mammalian cyclins ( inflammatory cell infiltrates and edema primarily surrounding vessels; Figure 2B ) and these mice did not show physical symptoms . Similarly , virus production in acute lung infection was fully restored by the viral cyclins , whereas all viruses expressing mammalian cyclins were impaired relative to those expressing the viral cyclins ( all statistically significant relative to gHV-cycV , p≤0 . 05 ) , but did partially restore acute lung titers relative to gHV-cycKO ( Figure 2C ) . These data suggest that mammalian cyclins have only a modest ability to function in viral infection and that the viral cyclins are unique in their ability to facilitate viral pathogenesis in lungs at early times post-infection . We previously reported that endothelial cells are able to support persistent gHV68 infection , a process that is dependent on the v-cyclin [20] . gHV68 infection results in a characteristic alteration in endothelial cell morphology , marked by modified gene expression and adherence-independent growth . These persistently infected endothelial cells remain viable for an extended time and are not lysed by virus infection , yet are productively infected and release abundant infectious virus . We next sought to determine the capacity of the recombinant cyclin viruses to promote survival and persistent infection in endothelial cells . Growth and survival of non-adherent surviving endothelial cells was measured at nine days post-infection ( Figure 3A ) . Cell survival ( percent annexin V- and PI-negative cells ) following infection with the recombinant cyclin viruses is shown in Figure 3B . As in acute pulmonary infection , the KSHV k-cyclin and the gHV68 v-cyclin were both fully functional in endothelial cell persistent infection; however , no mammalian cyclins showed modest or intermediate capacities . Instead , mammalian cyclins A and E , which bear functional similarity to the viral cyclins , were fully functional ( ≥50% viability ) in promoting persistent endothelial cell infection . In contrast , cyclins D3 and D2 , which share the most sequence similarity to the viral cyclins , conferred no advantage over a cyclin deficient virus ( Figure 3C ) . These data demonstrated that the viruses expressing D-type cyclins were completely defective in promoting endothelial cell survival ( Figure 3C ) , equivalent to the defect observed with a virus completely deficient for the v-cyclin . In contrast , not only both viral cyclins , but also mammalian cyclins E and A , were able to fully restore endothelial cell persistence ( Table 1 ) to the level conferred by the gHV68 v-cyclin . Latent infection with gammaherpesviruses is a complex process that is normally established in vivo , and is best measured after primary lytic infection has resolved . We previously used ex vivo analysis of cells infected with wild-type or cyclin deficient virus to show that the v-cyclin is critical to reactivation from latency [22] in both healthy and immune deficient mice [22] , [25] . This requirement for the v-cyclin is surprising , given the presumed expression of the homologous host cyclins during infection . And while other viral genes are also required for reactivation from latent infection , to date , no other single gene has been found to play an equivalent role . To determine the required cyclin function in reactivation during infection of mice , we first established that the recombinant cyclin viruses behaved as expected in vivo; that is , no mortality was observed during six weeks of infection , the relative cellularity of splenic and peritoneal cells was consistent with WT infection at both 16 and 42 days post-infection ( data not shown ) , and while infected cells are scarce , the FLAG-tagged v-cyclin can be detected in vivo ( Figure S3A ) during the peak of infection . Furthermore , we verified that reactivation of the FLAG-tagged recombinant virus was equivalent to that of the original WT virus ( Figure S3B ) , and based on our previous demonstrations that the viral cyclin is not required for the establishment of latency [22] , [25] , we used a subset of these viruses ( representing both those that do and do not complement reactivation from latency ) to show that , as expected , latency was established normally ( Figure S3C; no significant differences found in the frequency of latently infected cells ) . We previously demonstrated that the v-cyclin is required for both reactivation from latency and for persistent infection in immune deficient mice . Therefore we hypothesized that cyclin requirements for reactivation might be synonymous with those for persistence . This would predict complementation in reactivation by mammalian cyclins E and A , that is , that gHV-cycE and gHV-cycA would be significantly increased over gHV-cycKO . Infected peritoneal exudates cells ( PECs ) were plated on highly permissive MEF indicator cells for measurement of viral cytopathic effect ( CPE ) ( Figure 4A ) . Reactivation analyses of the full panel of recombinant cyclin viruses are shown in Figure 4B , with cyclin recombinant viruses that restored v-cyclin function in reactivation shown in Figure 4C and those that fail to complement shown in Figure 4D . To our surprise , gHV-cycE and gHV-cycA failed to support reactivation , as did gHV-cycD2 , with reactivation less than or equal to the cyclin deficient virus for each of these viruses ( Figure 4D ) . Reactivation frequencies of the gHV-cycKO and the non-complementing viruses were from extrapolated values , as the cells reactivating fell short of 63% even at the highest concentration . We found that the viral cyclins of both gHV68 and of KSHV complemented v-cyclin function in reactivation , and that gHV-cycV and gHV-cycK infections resulted in reactivation frequencies that did not statistically differ from each other . In addition , gHV-cycD3 was the only mammalian cyclin virus that differed significantly from the gHV-cycKO virus in supporting reactivation from latency ( Figure 4C ) . As expected from previous studies , the frequency of latently infected cells , or latency establishment , was similar between PECs infected with complementing versus non-complementing recombinant viruses ( Figure S2D ) . These data demonstrate that reactivation from latency is supported by a cyclin activity common to the viral cyclins and mammalian cyclin D3 ( Table 1 ) , but that is not shared with mammalian cyclins E , A or D2 .
Gammaherpesviruses establish lifelong infections in their host and are considered to be etiological agents for a variety of disease states , ranging from inflammatory conditions to malignancies , particularly in immunosuppressed individuals [34] , [35] . While the precise mechanisms by which these viruses establish a chronic infection remains an ongoing area of investigation , one gene that clearly influences chronic infection is the viral cyclin , encoded by the human virus Kaposi's sarcoma associated herpesvirus and murine gammaherpesvirus 68 . In KSHV , the k-cyclin is expressed during latency and reactivation from latency , and the k-cyclin regulates latency in KS cell lines [36] , [37] . By using murine gHV68 infection of mice to assess the role of the v-cyclin in multiple stages of infection , we and others have found that the v-cyclin is necessary for multiple facets of chronic infection and pathogenesis , including acute virus production in the lung during immune deficiency [19] , endothelial cell survival and viral persistence [20] , and reactivation from latency [22] . Notably , the gHV68 v-cyclin is also required for chronic pathogenesis in immunosuppressed individuals , including the induction of chronic inflammatory conditions ( e . g . in IFNgRKO; [25] , [38] ) and tumorigenesis ( e . g . in BALB/b2M KO mice; [26] ) . Given the varied roles that this gene has in promoting optimal gammaherpesvirus chronic infection and pathogenesis , there is a pressing need to understand the molecular mechanisms by which the v-cyclin mediates these diverse outcomes . While numerous reports have identified biochemical differences in the viral cyclins relative to host cyclins , to date there have been no studies to define which of these enhanced features of the viral cyclin are critical for gammaherpesvirus infection and pathogenesis . In fact , as reported here , analyses of kinase binding by viral and mammalian cyclins expressed under identical conditions during virus infection indicated multiple distinct patterns that do not correspond to subsequent functional studies . Additionally , kinase binding and activation may well differ in particular cell types and infection states in vivo , many of which are not readily amenable to biochemical analysis . Recently , mouse knock-outs and knock-ins have led to major advances in our understanding of cyclins and cdks , such that cyclins are now implicated not only in cell cycle progression , but in development , tissue specificity , tumorigenesis , and DNA damage and transcription , in the presence or absence of cdk partners . Using this successful approach , in this report , we tested the capacity of both viral and mammalian cyclins to function in multiple v-cyclin dependent stages of infection . Based on the enhanced biochemical features of the viral cyclins relative to mammalian cyclins , and the fact that host cyclins are present within the virus infected cell , we hypothesized that the viral cyclins of gHV68 and KSHV might be uniquely capable of functioning during virus infection . Indeed , when we first analyzed the ability of the various recombinant viruses to undergo replication in the lungs of immunosuppressed mice , we found that only the viral cyclins of either gHV68 or KSHV , and not mammalian cyclins , were capable of conferring wild-type levels of virus production and consequent increased pneumonia in infected lungs . These observations are consistent with the idea that the viral cyclins mediate their functions during infection through unique biochemical features , such as kinase binding , not present in mammalian cyclins . Notably , the viral cyclins of both gHV68 and KSHV were interchangeable in these tests of genetic complementation , despite this and previous reports identifying potential differences in their cdk binding partners and substrates [37] , [39]–[41] . These data identify a genetically conserved mechanism of the gammaherpesvirus cyclins for in vivo infection and pathogenesis . While only the viral cyclins provided optimal virus production in the immunosuppressed lung , further investigation of v-cyclin dependent stages of infection revealed a surprising ability of mammalian cyclins to mediate different stages of infection ( model represented in Figure 5 ) . On the one hand , host cyclins E and A was capable of fully promoting persistent infection of endothelial cells , while host cyclin D3 was capable of promoting reactivation from latency . Strikingly , the ability of host cyclins to mediate these distinct processes was non-overlapping , such that host cyclins capable of functioning in viral persistence were not capable of functioning in reactivation and vice versa . These data clearly demonstrate that , when expressed in the correct spatiotemporal manner ( by insertion in the endogenous v-cyclin locus ) , host cyclins are able to mediate distinct subsets of v-cyclin dependent functions in vivo . The existence of distinct genetic complementation groups of mammalian cyclins for optimal infection strongly suggests that these processes are mediated by distinct molecular mechanisms . Based on the ability of host cyclins E and A , but not D-type cyclins , to promote viral persistence it is worth asking what this complementation pattern might tell us about how the v-cyclin promotes persistence . What unique features do cyclins E and A have that differ from the D-type cyclins ? First , cyclins E and A ( and the v-cyclins ) differ in kinase partners , but all confer stronger kinase activation and longer half-lives than the D-type cyclins . While cyclins E and A may function by promoting cell cycle progression , herpesvirus infection is also associated with cell cycle arrest [42] . A second possible explanation for cyclins E and A in promoting endothelial cell persistence might be the fact that a cellular DNA damage response is important in promoting early herpesvirus DNA replication [43] , and these cyclins are important in the DNA damage response [44] , induction of which correlates to strength of kinase activation [45] . It is also worth noting that viral persistence is dependent on host autophagy machinery and an ability to survive substrate detachment as well [46] . These studies clearly demonstrate that the requirement for cyclin function in endothelial cell persistent infection corresponds to capacity for kinase activation . Based on this , we hypothesize that viral persistence may be particularly sensitive to therapeutic kinase inhibitors . Reactivation of virus replication from latently infected cells is a critical function of the v-cyclin in vivo , and correlates well with pathologies of chronic infection [25] , [26] , [38] . Because latent infection over time is concentrated in quiescent memory B cells , one straightforward possibility is that the v-cyclin is required in reactivation simply to stimulate quiescent cells into cycle . However , past reports have indicated that induction of the cell cycle by immunoglobulin cross linking or Toll-like receptor stimulation is insufficient to overcome the defect in reactivation that is observed with the v-cyclin-deficient virus [21] , [47] . If cell cycle progression via cdk activation is the sole requirement for reactivation , then redundancy in cell cycle function [14] predicts that proper expression of any cyclin should substitute for v-cyclin in reactivation . Instead we found that only the viral k-cyclin and mammalian cyclin D3 were able to genetically function in promoting reactivation from latency . These data compellingly indicate that reactivation from latency is dependent on a highly restricted cyclin activity possessed by cyclin D3 that does not correspond to kinase requirement in cell cycle progression , in which cyclins E and A can substitute for D-type cyclins and cyclin D/cdk4 or cdk6 complexes are not required for cell cycle [16] . This observation is consistent with the demonstration of wild-type reactivation following infection with mutant v-cyclin viruses that are impaired in cdk binding in vitro [48] , and with reports of cdk-independent functions of D-type cyclins [49]–[51] . The inability of cyclin D2 to compensate in reactivation is not likely a feature of poor protein stability [52] , because this is a shared feature of the D-type cyclins . Instead , our data suggest that the ability of mammalian cyclin D3 to function in reactivation is based on a unique role for cyclin D3 , such as transcription regulation [53]–[55] , activation of unconventional kinase partners [56] , [57] , or cell type-specific function . It is worth noting that a unique role for cyclin D3 in promoting virus infection has also been observed in promoting herpes simplex virus reactivation [58] . Given that B lymphocytes are the major latency reservoir of the gammaherpesviruses , it is notable that cyclin D3 is specifically required in lymphocyte development [59] , and particularly in germinal center B cells , a prominent early reservoir for viral latency [60] , [61] . Beyond specific insights into the mechanisms by which the v-cyclin promotes chronic infection , this study also revealed a fundamental new insight in gammaherpesvirus infection , by demonstrating that viral reactivation from latency and viral persistence are genetically separate processes . To date , these processes are frequently intertwined spatially and temporally , making it difficult to discern their interrelationship . These distinct cyclin functions suggest a new explanation for the partial complementation of the mammalian cyclins during the acute phase of replication in the lung . While the general assumption that primary lytic virus replication is cleared and then followed by latent infection , Flano , et al . provided evidence that lytic and latent infection occur simultaneously and early in the lungs , and that latently infected cells are apparent as early as three days post-infection [62] . Our study further supports this finding , and provides genetic evidence that reactivation from latency , generally considered only in later stages of infection , contributes to virus production during the early stages of primary infection . Previously , persistent infection ( as defined by detection of infectious virus late in infection ) and reactivation were both increased in immune-deficient mice , consistent with increased reactivation from latency resulting in increased persistent infection , or vice versa [38] . Two viral homologs of host genes , the viral cyclin and the viral bcl-2 , are both required in persistence and in reactivation . The first indication that these are separable processes was identified by recent analysis of the viral bcl-2 homolog , in which different v-bcl-2 mutants were capable of supporting either persistence or reactivation [63] . Further , we demonstrated that both the v-cyclin and the v-bcl-2 are required for optimal virus production and lethal pneumonia in immune-deficient hosts [19] . First , it is remarkable that the genetic requirements for both the v-cyclin and the v-bcl-2 are separable in these distinct aspects of virus infection . Second , these data illustrate that in immune-deficient mice , acute virus production cannot be solely attributed to primary virus replication , but may be the sum of replication , persistence , and reactivation . Genetic separation of these functions raises the potential that chronic disease previously associated with both persistence and reactivation may be dependent on one or the other . The cyclin recombinant viruses now provide a mechanism to determine the relative contribution of reactivation and persistence in various disease processes , and may provide insight for therapeutic interventions specifically tailored to the cyclin susceptibility of each . In total , our findings demonstrate that the multifunctional nature of the viral cyclins described in in vitro biochemical studies corresponds to genetically distinct and required functions during virus infection , and that both the gHV68 and KSHV viral cyclins share this multifunctional capacity in infection . Additionally , this study revealed distinct genetic complementation groups of the mammalian cyclins , demonstrating that mammalian cyclins can fulfill the biochemical features of the v-cyclin in infection . These studies reveal that the unusual biochemical features of viral cyclins , such as broad substrate specificity and increased kinase activity , are not absolutely required to mediate specific processes within viral infection . And yet , cyclin D3 restored v-cyclin dependent reactivation less effectively than did the viral cyclins , suggesting that unique biochemical feature ( s ) of viral cyclins may be required to facilitate robust activity in reactivation . These data also indicate that v-cyclin features , such as resistance to cell cycle inhibitors or enhanced kinase activity , are necessary for optimal gammaherpesvirus pathogenesis . The unique capacity of the viral cyclins to encompass functions of multiple mammalian cyclins probably explains the evolutionary advantage of encoding viral cyclins within the viral genome . Whereas only the viral cyclins can perform all v-cyclin dependent parameters of infection , our data also suggest that expression of endogenous host cyclins could complement v-cyclin-dependent functions in vivo . This idea is consistent with our observations that neither persistent infection nor reactivation from latency is completely abrogated in absence of the v-cyclin . Since mammalian cyclins can genetically replace the v-cyclin in distinct stages of infection , we hypothesize that methods of interfering with mammalian cyclin-mediated processes may also be effective at inhibiting specific functions of the v-cyclin . The ultimate test of this idea will be specific chemical inhibition of specific v-cyclin functions , and whether such inhibition indeed decreases persistent infection and reactivation levels below that of v-cyclin deficient viruses . Finally , the distinct cyclin requirements in different v-cyclin stages of infection provide potential for specific treatment of different gammaherpesvirus pathologies using existing therapeutic inhibitors specific to certain host cyclins and cdks [64] . Because cyclins and cdks are well-conserved and are host proteins , this strategy circumvents potential virus escape and may also prove useful for treatment of herpesviruses that do not encode cyclins within their genomes .
This study was carried out in strict accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health . All animal studies were conducted in accordance with the University of Colorado Denver Institutional Animal Use and Care Committee under the Animal Welfare Assurance of Compliance # A3269-01 . All surgery was performed under isoflurane anesthesia , and all efforts were made to minimize suffering . gHV68 clone WUMS ( WT; ATCC VR1465 ) , gHV68 containing a stop codon within ORF 72 ( cycKO ) , and parental and epitope-tagged recombinant viruses ( Figure S1 ) were passaged and grown , and the titer was determined as previously described [22] , [31] . NIH 3T12 ( ATCC CCL-164 ) , Vero-Cre cells ( Dr . David Leib , Dartmouth Medical School of Medicine ) and mouse endothelial cell lines MB114 [65] were grown in Dulbecco's modified Eagle medium ( DMEM ) supplemented with 5% fetal bovine serum ( FBS ) , 100 U/ml penicillin , 10 ug/ml streptomycin sulfate , and 2 mM L-glutamine . Mouse embryonic fibroblasts ( MEFs ) were isolated from C57/BL6 mice as previously described [66] and cultured in DMEM supplemented with 10% FBS , 2 mM L-glutamine , 10 U/mL penicillin , 10 µg/mL streptomycin sulfate , and 250 ng/mL amphotericin B . Infection of MB114 endothelial cells was carried out at a multiplicity of infection ( MOI ) of 5 plaque forming units ( PFU ) per cell , as previously described [20] . The inoculum was removed after one hour of infection at 37°C , and the cell monolayers were cultured in complete media after rinsing with PBS . Intact and non-adherent cells were collected at six days post-infection , at which time cells and media were collected [20] . C57BL/6 and IFN-g−/− mice on a BALB/c background ( strain C . 129S7 ( B6 ) -Ifngtm1Ts/J ) were purchased from The Jackson Laboratory ( Bar Harbor , ME ) . Eight to ten week old mice were infected by intraperitoneally ( i . p . ) with 1×106 PFU of virus in 0 . 5 ml of DMEM/5% FBS for reactivation studies and intranasally ( i . n . ) with 4×105 PFU of virus in 40 µl of DMEM/5% FMS for acute infection studies . Upon sacrifice , lungs for which virus titers were to be determined were placed in 1 ml of DMEM/5% FBS on ice and frozen at 80°C [19] . Peritoneal cells ( PECs ) were harvested by peritoneal lavage with 10 ml of DMEM/1% FBS [22] . Viral DNA was generated by infection of 3T12 cells at an MOI of 0 . 05 for each recombinant virus . Infected-cell cultures were harvested at 50% CPE , and DNA was prepared as previously described [31] . 5–10 ug of viral DNA was restriction enzyme-digested for four hours . Digests were electrophoresed on 0 . 8% agarose gels with biotinylated DNA ladders ( New England Biolabs , Ipswich , MA ) . The DNA was alkaline transferred to Zeta-Probe membrane ( Bio-Rad Laboratories , Hercules , CA ) using the Turboblot apparatus ( Schliecher & Schuell , Keene , NH ) , according to the manufacturer's recommendations . Probes were from gHV68 genome coordinates 101656 to 105385 ( cyclin region probe ) or 11100 to 16328 region ( left end probe ) . Each probe was biotinylated and quantitated , according to manufacturer's instructions for the KPL Detector HRP chemiluminescent blotting kit ( KPL , Inc . , Gaithersburg , MD ) . gHV68 clone WUMS ( ATCC VR-1465 ) and recombinant viruses were passaged and grown , and titer determined as previously described [31] . Plaque assays were performed on 3T12 cells as previously described [19] , [25] . Lung homogenates were serially diluted , and plated onto NIH 3T12 cells in 12 well plates in triplicate . The limit of detection of the assay is 50 PFU . Viral replication in vitro was determined by infection of 3T12 cells at a MOI of 0 . 05 PFU per cell to measure multiple cycle replication . Cells and supernatants were collected at various times post-infection and frozen at 80°C . Samples were subjected to four cycles of freezing and thawing prior to quantitation by plaque assay . Plasmids and BAC DNA were introduced into cells using the calcium phosphate method . 293T , Vero-Cre or 3T12 cells were plated in 6-well plates and transfected at 50–80% confluency . The DNA mixture for each well , which consisted of the 1–10 ug DNA , 2 M CaCl2 , and sterile H20 , was combined with 2× HEPES balanced saline buffer ( 0 . 3 M NaCl , 0 . 05 M HEPES , 0 . 003 M Na2HP04 , H20; pH 7 . 05–7 . 15 ) and added to each well dropwise while gently swirling plates . At 16 hours post transfection , cells were washed with 1× phosphate buffered saline , the media were replaced with DMEM/5% FBS , and cells were examined by fluorescence microscopy at various times post-transfection to monitor transfection efficiency or BAC-GFP deletion . Total RNA was isolated from infected 3T12 cells using TRIzol Reagent ( Invitrogen ) and then purified using the RNeasy Micro Kit ( Qiagen ) . An ABI Prism 7900 sequence detector ( Applied Biosystems , Foster City , CA , USA ) was used for measurement of the fluorescence spectra in a thermal cycler during PCR amplification ( University of Colorado Cancer Center Quantitative PCR Core Facility ) . Forward and reverse primers and probe ( Applied Biosystems ) specific to the 3x-FLAG-CMV 7 . 1 epitope were designed according to the recommendations of the TaqMan PCR chemistry design and optimized using the Primer Express software ( Applied Biosystems ) . Primer and probe sequences used were 3x-FLAG7 . 1FWD-CTACAAAGACCATGACGGTGATTATAA; 3x-FLAG7 . 1REV . NEW-TCGCGGCCGCAAGC; 3x-FLAG7 . 1PROBE-6-carboxyfluorescein-CATGACATCGATTACAAGGATGACGATGAC-6-carboxy-tetramethylrhodamine . Amplification reactions and thermal cycling conditions were performed as per the manufacturer's recommendations . A standard curve was created using the fluorescence data from 10-fold serial dilutions of a 24 hour 3x-FLAG-v-cyclin infection . The 24 hour 3x-FLAG data were normalized to the 12 hour 3x-FLAG data , and is represented as the ratio of 3x-FLAG RNA to the total amount of 18S rRNA per sample . The following antibodies were used: mouse anti-Flag ( M2 , Sigma-Aldrich ) , rabbit anti-Flag ( Cell Signaling Technology , Inc , Danvers , MA ) , rabbit anti-v-cyclin[11] ) , rabbit anti-k-cyclin [gift from Sibylle Mittnacht [67] , rabbit anti-cdk1/cdc2 , goat-anti-cdk2 , goat anti-cdk4 , rabbit anti-cdk6 , rabbit-anti-cyclin A , rabbit anti-cyclin D3 , rabbit anti-cyclin E ( Santa Cruz Biotechnology , Inc . , Santa Cruz , CA ) , mouse anti-beta-actin ( Sigma ) , and donkey anti-rabbit-HRP , donkey anti-mouse-HRP , donkey anti-goat-HRP ( Jackson ImmunoResearch Laboratories , Inc , West Grove , PA ) . Protein expression was detected by lysing cells in ELB buffer ( 50 mM HEPES pH 7 . 2 , 250 mM NaCl , 2 mM EDTA , 0 . 1% NP-40 ) for 20 minutes on ice , and boiling for 10 minutes . Equal cell equivalents or equal amounts of protein , based on RC-DC protein assay ( Bio-Rad ) were loaded per lane . Samples were separated by electrophoresis on 7 . 5%–15% denaturing polyacrylamide gels and transferred to Immobilon-P membranes ( Millipore Corp . , Bedford , MA ) by semi dry protein transfer ( Panther Semi Dry Electroblotter , Thermo Fisher Scientific , Inc . , Portsmouth , NH ) , and analyzed ECL Plus western blotting detection reagents ( GE Healthcare , Piscataway , NJ ) . 10% of each cell lysate was set aside for lysate loading controls in immunoprecipitations . Remaining lysates in ELB containing protease inhibitors ( 1 mM DTT , 10 mM NaF , 50 ug/mL PMSF , 1 ug/mL aprotinin , 1 ug/mL leupeptin ) were precleared with protein A sepharose CL-4B beads ( GE Healthcare ) for one hour at 4°C with agitation . Lysates were then clarifed , and incubated for one hour at 4°C with anti-Flag Ab ( Sigma ) prior to the addition of sepharose beads and overnight incubation . Beads were washed four times in cold ELB with inhibitors , boiled for 10 minutes in Laemmli buffer ( 0 . 25 M Tris-HCl pH 6 . 8 , 2% SDS , 10% glycerol , 5% b-mercaptoethanol , 0 . 002% bromophenol blue ) and subjected to SDS-PAGE . Cells infected for immunohistochemical detection of 3x-FLAG cyclins were fixed using 3∶1 methanol: glacial acetic acid . 20 ug/ml of mouse anti-FLAG ( Sigma ) was added to the cover slips and visualized with goat anti-mouse Alexa Fluor 568 ( 1∶1000; Invitrogen ) . For ex vivo FLAG detection , four-six µm sections were deparaffinized and before antigen retrieval using 10 mM citrate buffer . Tissues were blocked using 10% 2 . 4G2 and 5% goat serum in PBS prior to staining with rabbit anti-FLAG at 1∶500 ( Cell Signaling ) followed by biotin goat anti-rabbit at 1∶50 ( BD Pharmingen ) and streptavidin-RPE ( Invitrogen ) at 1∶100 . All slides were mounted with ProLong Gold antifade reagent with 4′-6-diamidino-2-phenylindole ( DAPI , Invitrogen ) and images were obtained using an Olympus IX81 inverted motorized scope with spinning disk ( Olympus , Center Valley , PA ) , a Hamamatsu ORCA IIER monochromatic CCD camera ( Hamamatsu , Bridgewater , NJ ) and Intelligent Imaging Slidebook v . 4 . 067 ( Intelligent Imaging Innovations , Denver , CO ) . The frequency of cells containing viral DNA was determined by a limiting-dilution nested-PCR assay that amplifies gHV68 gene 50 sequences with approximately single-copy sensitivity , as described previously [22] , [68] . Briefly , peritoneal cells ( PECs ) were harvested from latently infected mice and plated as a limiting-dilution series of cells . The cells were lysed prior to PCR amplification , and the first-round PCR product served as a template for the second round of PCR amplification . Control reactions of uninfected cells ( negative control ) or plasmid DNA ( pBamHIN ) of known copy number ( positive control ) were included in each experiment [25] . Quantitation of gHV68 reactivation from latency was performed as previously described [22] , [68] , [69] . Briefly , PECs were harvested from infected mice at day 42–50 post-infection , and single-cell suspensions were generated . Two-fold serial dilutions of infected cells were plated onto MEFs and scored for CPE after 21 days of co-culture . To detect preformed infectious virus , parallel samples were mechanically disrupted as previously described [25] . Two parameter viability studies using propidium iodine ( PI ) and annexin V were performed as previously described [20] and analyzed by FlowJo ( Treestar , Ashland , OR ) . For histologic examination , lungs were fixed in 10% formalin , paraffin embedded , sectioned ( 4–6 µm ) and stained with H&E for analysis using a Zeiss Axiocam HR camera and KS 300 Imaging System 3 . 0 software [27] . Pulmonary disease was evaluated by board certified pathologist , Dr . Carlyne Cool . All data was analyzed by using GraphPad Prism software ( GraphPad Software , San Diego , CA ) . Viral titers were statistically analyzed with a one-way ANOVA test . Differences in endothelial cell survival were statistically analyzed by unpaired t-test . The frequencies of reactivation and genome-positive cells were statistically analyzed by paired t-test . Frequencies of latently infected and reactivating cells were obtained from the cell number at which 63% of the wells scored positive for either reactivating virus or the presence of the viral genome based on the Poisson distribution . Data were subjected to nonlinear-regression analysis to obtain the single-cell frequency for each limiting-dilution analysis . Genbank accession numbers for proteins studied within this manuscript: gHV68 cyclin AAB66456; KSHV cyclin ADQ57958; human cyclin A2 AAM54042; cyclin E1 AAH35498; cyclin D3 AAA52137; cyclin D2 AAH89384 . | Many viruses encode homologs of human oncogenes , including the gammaherpesvirus viral cyclin genes . These viruses cause lifelong infection associated with chronic diseases , including malignancies , which are exacerbated in immune deficiency . The conserved viral cyclins were first recognized nearly two decades ago , and despite extensive interest and study , their essential features for virus infection and disease have been elusive . We used a mouse model of these viruses to make recombinant viruses with viral or human cyclins knocked into the endogenous locus . We then determined the requirements for cyclins by genetic complementation in three distinct viral cyclin dependent aspects of infection . We report that the viral cyclins of different gammaherpesviruses are able to support all three stages of infection . However , none of the human cyclins can , and instead comprise distinct complementation groups that are functional in non-overlapping aspects of infection . We showed that gammaherpesvirus encoded cyclins are functionally conserved , and that their essential unique property is the assimilation of the functions of distinct mammalian cyclins within a single multifunctional gene . Finally , in dissecting the requirements for viral cyclins during gammaherpesvirus infection , we demonstrated that related stages of infection are genetically separable and therefore may be susceptible to specific therapeutic manipulation . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"molecular",
"cell",
"biology",
"cell",
"division",
"cell",
"biology",
"virology",
"biology",
"microbiology"
] | 2012 | Viral Cyclins Mediate Separate Phases of Infection by Integrating Functions of Distinct Mammalian Cyclins |
The Epstein-Barr virus ( EBV ) infects and transforms B-lymphocytes with high efficiency . This process requires expression of the viral latent proteins and of the 3 miR-BHRF1 microRNAs . Here we show that B-cells infected by a virus that lacks these non-coding RNAs ( Δ123 ) grew more slowly between day 5 and day 20 , relative to wild type controls . This effect could be ascribed to a reduced S phase entry combined with a moderately increased apoptosis rate . Whilst the first phenotypic trait was consistent with an enhanced PTEN expression in B-cells infected with Δ123 , the second could be explained by very low BHRF1 protein and RNA levels in the same cells . Indeed , B-cells infected either by a recombinant virus that lacks the BHRF1 protein , a viral bcl-2 homolog , or by Δ123 underwent a similar degree of apoptosis , whereas knockouts of both BHRF1 microRNAs and protein proved transformation-incompetent . We find that that the miR-BHRF1-3 seed regions , and to a lesser extent those of miR-BHRF1-2 mediate these stimulatory effects . After this critical period , B-cells infected with the Δ123 mutant recovered a normal growth rate and became more resistant to provoked apoptosis . This resulted from an enhanced BHRF1 protein expression relative to cells infected with wild type viruses and correlated with decreased p27 expression , two pro-oncogenic events . The upregulation of BHRF1 can be explained by the observation that large BHRF1 mRNAs are the source of BHRF1 protein but are destroyed following BHRF1 microRNA processing , in particular of miR-BHRF1-2 . The BHRF1 microRNAs are unlikely to directly target p27 but their absence may facilitate the selection of B-cells that express low levels of this protein . Thus , the BHRF1 microRNAs allowed a time-restricted expression of the BHRF1 protein to innocuously expand the virus B-cell reservoir during the first weeks post-infection without increasing long-term immune pressure .
The Epstein-Barr virus ( EBV ) is the first discovered tumor human virus and is etiologically associated with approximately 2% of all tumors worldwide [1 , 2] . These tumors are largely diverse in terms of lineage and include multiple types of lymphomas and carcinomas [3] . Immune deficiency , e . g . caused by immunosuppressive regimen is a strong risk factor for the development of EBV-associated lymphomas [2] . These tumors are thought , at least to some extent , to reflect EBV’s ability to transform primary B-cells [2] . This process can be easily observed in vitro as it leads to the establishment of lymphoblastoid cell lines ( LCLs ) and requires the simultaneous expression of some members of the viral latent gene family [2] . In recent years , it has become clear that the BHRF1 microRNAs ( miRNAs ) encoded by the virus markedly potentiate this process . Recombinant viruses that lack one or several of these three miR-BHRF1s are less transforming than their wild type counterparts and the effect is cumulative [4–6] . One study has ascribed this property to the ability of the BHRF1 miRNAs to prevent massive apoptosis in the first days of infection [4] . Furthermore , viruses that lack the three BHRF1 miRNAs ( Δ123 ) grow more slowly and display abnormalities of the cell cycle [4 , 5] . Humanized NSG mice infected by Δ123 eventually develop B-cell proliferations that are indistinguishable from those caused by wild type infection , but cell growth induced by the mutant is delayed by several weeks , confirming that the BHRF1 miRNAs are particularly required in the early phases of infection [7] . The BHRF1 protein , around which the BHRF1 microRNAs are located , has also been implicated in EBV-mediated B-cell transformation , although its role appears to be more difficult to define . BHRF1 is a bcl-2 homolog that shares its anti-apoptotic properties [8 , 9] . Although its expression level is hardly detectable in LCLs , it is strongly expressed in the Wp-restricted Burkitt’s lymphoma ( BL ) cells , a subset of Burkitt’s lymphomas that are infected by EBVs that carry a deletion of the EBNA2 gene and whose latent genes are driven by the Wp promoter [10 , 11] . A recombinant virus that lacks the BHRF1 protein retains full transformation abilities , suggesting that this protein is dispensable for transformation [12] . However , its enhanced expression in Wp-restricted BLs leads to a markedly enhanced resistance to apoptosis induced by ionomycin [11 , 13] . Thus , both the BHRF1 protein and the BHRF1 miRNAs have been implicated in the regulation of apoptosis . The prominent role played by some of the EBV latent genes in B-cell transformation raises the question of a possible interaction of the BHRF1 miRNAs with the latent genes . Although these have not been directly identified in a search for the miR-BHRF1 targets , we have clearly identified the EBNA-LP latent gene as a , probably indirect , target of the BHRF1 miRNAs [5] . The expression of this protein is usually downregulated in LCLs after several weeks of growth in culture but this process is largely delayed after infection with Δ123 [5] . The other latent genes were also upregulated in LCLs infected by Δ123 relative to wild type counterparts but the effect was much weaker and inconstant . This raised the question whether the BHRF1 open reading frame is also a target of the BHRF1 miRNAs but its transcription level was not affected in cells infected by the mutant [4 , 5] . In this paper we examine the role played by the complete BHRF1 locus during EBV infection . We found that the BHRF1 miRNA cluster controls the temporal expression of the BHRF1 protein but also downregulates PTEN . The deletion of this cluster also led to the frequent emergence of transformed B-cells with a downregulation of p27 .
EBV infection of B-cells induces permanent cell division that gives rise to the establishment of lymphoblastoid cell lines ( LCLs ) . Therefore , we began our investigations by monitoring cell growth and cell vitality over the first four weeks after infection with the Δ123 mutant or with wild type EBV controls . This was achieved by directly counting mitoses in the samples or staining cells with phospho-histone H3 , a marker of cells undergoing mitosis . Both methods showed no evidence of cell division before day 3 , as expected [14] . Cells infected with the wild type control then began dividing , reaching a peak at around day 10 and then maintained a constant mitotic rate between 1 and 2% ( Fig 1a and 1b ) . The same B-cells infected in parallel with Δ123 differed from wild type controls in that their mitotic rate was 2–3 fold lower . However , after 25 days , both mutants and controls hardly showed any difference . These data suggested transient differences in cell cycle regulation between both groups of cells . Therefore , we performed a BrdU incorporation assay at day 13 post-infection ( p . i . ) . This experiment showed a decrease in the fraction of cells that entered the S phase , as well as a relative increase in the number of cells present in G2/M in Δ123-infected cells ( Fig 1c ) . This resulted in a statistically significant difference in the G2/M to S ratio between B-cells infected with Δ123 or with wild type virus . We then stained the same infected cells with an antibody specific to cleaved pro-caspase 3 that detects the form of the protein activated during apoptosis , coupled to a TUNEL assay that detects double strand DNA breaks . These assays are well suited for the detection of apoptosis at the single cell level and the results are summarized in Fig 1d and 1e . They showed that between day 1 and day 5 , B-cells infected with wild type viruses or with Δ123 behaved identically . However , from day 8 on , the apoptosis rate grew larger in the cells infected with the miRNA triple mutant and became twice as high at day 15 . The apoptotic rate decreased in these cells after day 18 to reach those evinced by wild type LCLs at day 34 . Altogether , we conclude that cells infected by the Δ123 mutant do not enter the cell cycle as efficiently and undergo more apoptosis than the controls after initiation of cell division between day 8 and 20 after infection . As described in the sequel , transformation of additional B-cell samples revealed that the amplitude of the difference between B-cells infected by wild type or mutant viruses in terms of apoptotic rate and mitotic growth can vary . However , the general picture remained similar . The BHRF1 miRNA cluster is located around the BHRF1 gene , whose protein product is endowed with anti-apoptotic properties [9 , 11] . Therefore , we monitored BHRF1 protein expression by western blot around the critical period , between 1 and 20 days p . i . We used an EBV-negative clone of the Burkitt’s lymphoma cell line Elijah and the Wp-restricted cell line Oku as a negative and a positive control , respectively . This assay , shown in Fig 2a , revealed that the BHRF1 protein is transiently produced in cells infected by the wild type controls at levels in the range of those observed in Oku . Oku is known to express BHRF1 at much higher levels than established LCLs [11] . Expression began at day 1 , became fully visible at day 3 and reached a peak at day 5 , after which it decreased again to nearly disappear at day 18 . In stark contrast , B-cells infected with Δ123 produced hardly detectable levels of the protein . We then performed northern blots at the peak of BHRF1 protein expression at day 5 in B-cells infected with EBV wild type or Δ123 ( Fig 2b ) . This assay revealed that B-cells infected with wild type viruses produce multiple and abundant transcripts ranging from 1 . 3 to larger than 10 kb . In comparison , cells infected with Δ123 showed only large transcripts that were altogether much less abundant than in the wild type-infected LCLs . Thus , the reduced BHRF1 protein expression correlates with reduced BHRF1-specific transcription . We left the infected cells grow for another 55 days and repeated the experiment . The blots revealed that by that time the 1 . 3 kb band had become prominent in the LCL infected by wild type viruses with larger bands becoming much fainter . Interestingly , the LCL infected with Δ123 retained large BHRF1 transcripts , albeit slightly smaller than at day 5 . Consequently , the large BHRF1 transcripts were more abundant in LCLs infected with Δ123 than in those generated with wild type controls . These results indicate that the wild type BHRF1 locus is transcribed at a different rate at an early and late stage of infection . However , whilst transcription markedly diminished with time in the wild type LCL , it remained nearly constant in cells infected with the mutant . We attempted to confirm these investigations by qPCR . We interrogated several BHRF1 transcripts , as indicated in Fig 2c , as well as those driven by the Wp promoter that dominates transcription at day 5 . We found that LCLs infected with the wild type virus express much higher levels of BHRF1 transcripts at day 5 than at day 12 . The same pattern was visible for W2-BHRF1 spliced transcripts and for Wp-driven transcripts . In the LCL infected with the Δ123 mutant , the BHRF1 transcripts at day 5 were expressed at a much lower level ( 10 to 20% of wild type level ) . They then increased to become more abundant than in the wild type controls , to fall again at approximately wild type levels after one month of infection . Thus , the usual pattern of Wp transcription , i . e . high levels shortly after infection that rapidly decrease after a few days , is translated to the right and reduced in its initial peak in cells infected with the mutant . We conclude from these findings that the BHRF1 locus undergoes very dynamic changes in terms of transcription over time and that BHRF1 transcription and translation is abnormally low at an early time point in LCLs infected by Δ123 . The latter results also suggested that the low abundance of the BHRF1 protein might be responsible for the observed increased apoptosis rate . We addressed this issue by studying the phenotype of a virus that carries an inactivating point mutation of the BHRF1 start codon ( ΔBHRF1 , see S1 Fig ) and of a virus that lacks both the BHRF1 miRNAs and the BHRF1 open reading frame ( Δ123ΔBHRF1 ) . B-cells exposed to these viruses or to a wild type control showed little differences in apoptosis and cell growth until day 5 ( Fig 3a , 3b and 3c ) . From day 6 onwards , apoptosis remained low in B-cells infected with wild type EBV but increased steadily in B-cells infected with the Δ123ΔBHRF1 double mutant to reach 100% between day 15 and day 30 , depending on the blood samples tested . B-cells transformed by wild type viruses retained a mitotic rate of about 1 . 5% of the total cells after staining with PH3 . In contrast , LCLs generated with the Δ123ΔBHRF1 double mutant exhibited a much lower mitotic rate that never exceeded 0 . 5% . LCLs transformed with ΔBHRF1 or Δ123 showed an intermediate profile between these 2 extreme phenotypes . In some of the 5 studied cases , the ΔBHRF1 LCLs and the Δ123-infected LCLs did not differ markedly from wild type-infected LCLs , with a good mitotic rate and a low level of apoptosis . In others , their behavior was closer to those of cells infected with the double knockout virus . In all cases , the LCLs generated with the ΔBHRF1 virus displayed higher mitotic rates than their Δ123 counterparts . We addressed this issue in more detail and performed a BrdU incorporation assay early after infection with wild type EBV , Δ123 and ΔBHRF1 . The results of this experiment are depicted in Fig 3d and show that the percentage of cells in S phase is lowest in B-cells transformed with Δ123 , highest in those transformed with wild type EBV and intermediate in cells transformed with ΔBHRF1 . These data suggest that the mild increase in apoptosis and some of the cell cycle abnormalities observed in B-cells infected with Δ123 could be largely explained by the reduction in BHRF1 protein levels and that the low BHRF1 expression level , the only difference between B-cells infected with Δ123 and those infected with Δ123ΔBHRF1 , becomes indispensable for survival of cells infected by a Δ123 virus . We then performed transformation assays at low cell density and low MOI on feeder cells in 96-well cluster plates ( Fig 3e ) . Feeder cells have previously been shown to reduce apoptosis in EBV-infected B-cells [15] . This assay showed that the B-cell transformation rate was highest in cells infected with wild type EBV , whilst B-cells infected with the Δ123ΔBHRF1 double knockout did not show any signs of outgrowth . The transformation assays performed with ΔBHRF1 or Δ123 showed again intermediate results . However , the transformation rate was much higher after infection with ΔBHRF1 than after infection with Δ123 . Thus , the absence of BHRF1 but not those of the BHRF1 miRNAs can be largely compensated by feeder cells . Therefore , the Δ123 phenotype is not limited to a reduction in BHRF1 levels . This experiment also shows that ΔBHRF1 viruses are mildly less transforming than wild type viruses , an observation consistent with the moderate reduction in S phase entry observed previously . We tested whether it is possible at all to generate LCLs with the Δ123ΔBHRF1 double mutant . To this end , we infected B-cells from 4 different donors at high cell density ( 104 cells per well of a 96-well cluster plate ) at an MOI of 10 infectious units per cell and kept the cells on a feeder cell layer for 1 . 5 months . This led to the establishment of LCLs in 2 out of 4 cases . The results obtained with the Δ123 virus showed that the BHRF1 miRNAs are involved in the control of the BHRF1 protein production . We determined which of these miRNAs is implicated in this process by infecting primary B-cells with viruses that lack one of the BHRF1 miRNAs . Western blot analysis with a BHRF1-specific antibody revealed that infection of primary B-cells with a virus that lacks miR-BHRF1-1 or with the wild type control gave rise to the same level of BHRF1 protein production at day 5 ( Fig 4a ) . In contrast , BHRF1 expression in B-cells infected with single and double miRNA mutants that lack miR-BHRF1-2 , miR-BHRF1-3 or both ( Δ2 , Δ3 , Δ23 ) was markedly reduced relative to B-cells infected with wild type viruses ( Fig 4b and S1 Fig ) . We wished to confirm these findings by infecting B-cells with a virus that carries a seed mutation in miR-BHRF1-3 ( 3SM ) ( S1 Fig ) . This mutant expresses miR-BHRF1-1 and miR-BHRF1-2 at wild type levels ( S2 Fig ) . Primary B-cells were infected in parallel with wild type virus , Δ123 and 3SM and harvested at day 5 post-infection . We performed an immunoblot with an anti-BHRF1 antibody that confirmed a clearly decreased BHRF1 protein expression after infection with 3SM , relative to wild type levels , although the amplitude of the effect was not as pronounced as after infection with the Δ3 virus ( Fig 4c ) . This might point towards a role for miR-BHRF1-3* in this process . However , the low expression of this miRNA in LCLs argues against its role in the regulation of BHRF1 protein expression [16 , 17] . We also performed this experiment with a mutant that carries mutations of both seed regions encoded by pre-miR-BHRF1-2 ( 2/2*DSM ) ( S1 Fig ) . Indeed , we previously showed that the Δ2 virus also evinces a reduced miR-BHRF1-3 expression [6] . 2/2*DSM expresses normal miR-BHRF1-3 levels and is therefore suitable to study the contribution of miR-BHRF1-2 to the regulation of BHRF1 protein expression ( S2 Fig ) . B-cells infected with this mutant displayed no altered phenotype and expressed the BHRF1 protein at day 5 post-infection at approximately 60% of the levels observed in cells infected with wild type viruses ( Fig 4d ) . Altogether , this set of experiments identify miR-BHRF1-3 , and to a lesser extent miR-BHRF1-2 , seed regions as positive modulators of BHRF1 protein expression early after infection . We then gauged the expression of the BHRF1 transcripts at the same early time point using qPCR ( Fig 4e ) . This assay confirmed the reduced Wp-driven transcription in B-cells infected with Δ123 , but also revealed that this effect was also visible in B-cells infected with Δ3 , 2/2*DSM and 3SM . The BHRF1 transcripts were also less abundant in all these samples , although the Δ3 mutant showed more drastic effects than the 2 seed mutants , thereby confirming the data gathered at the protein level . The BHRF1 protein can be produced either from a lytic or from a latent promoter . To determine which of these forms is produced at an early stage of infection we infected B-cells with a virus that lacks the BZLF1 gene ( ΔZ ) that encodes a transactivator indispensable for the onset of lytic replication in B-cells ( Fig 4f ) . This experiment showed that BHRF1 protein production is only slightly reduced at day 5 in B-cells infected with the ΔZ mutant , relative to wild type virus . Thus , it is the latent form of the BHRF1 protein that is predominantly produced at an early time point as previously suggested [11] . The data gathered so far confirmed that the B-cells infected with Δ123 have a reduction in cell cycle entry , even on feeder cells under conditions in which apoptosis is limited . We knew from previous work that a virus that lacks miR-BHRF1-3 displays similar , if less pronounced , abnormalities [6] . We looked for a gene implicated in the cell cycle control that would be regulated by miR-BHRF1-3 . PTEN has previously been identified by a PAR-CLIP method as a potential target of miR-BHRF1-3 [17] ( Fig 5a ) . Therefore , we tested expression of this protein at different time points after infection with Δ123 and wild type controls and found that it increased in intensity regularly from day 1 to day 5 . After day 5 it became obvious that B-cells infected with the Δ123 virus expressed more PTEN than wild type counterparts ( Fig 5b ) . We wished to confirm that this effect was due to the absence of miR-BHRF1-3 and assessed expression of PTEN in cells infected with the Δ3 or the 3SM virus . This experiment confirmed that cells infected with either of these single miRNA mutants evinced a stronger PTEN expression relative to wild type ( Fig 5c and 5d ) . We also quantified PTEN expression in LCLs generated with ΔBHRF1 ( S3 Fig ) . This assay could not reveal any difference in PTEN expression in these cells , relative to wild type controls . We went on to perform a luciferase reporter assay in which the luciferase gene is fused with part of the 3’UTR of PTEN that contains the putative miR-BHRF1-3 binding site . We also included a negative control in which the putative miR-BHRF1-3 binding site had been mutated ( seed-match mutant ) . Cotransfection of either of these constructs together with a miR-BHRF1-3 expression plasmid revealed a modest but statistically significant decrease in relative luciferase activity in the wild type PTEN 3’UTR fusion that was not visible in the seed-match mutant control ( Fig 5e ) . This weak effect can at least in part be ascribed to the intrinsic low miR-BHRF1-3 expression level [6] . We then treated 2 14-days old EBV wild type-infected LCLs with wortmannin , an inhibitor of PI3K that mimics an activation of PTEN [18] . We found that the treatment of these cells with wortmannin decreased entry in S phase by one third and increased the G2/M to S phase ratio to a value close to the one we observed in the LCLs infected by Δ123 ( Fig 5f ) . Thus , wortmannin treatment reproduced the cell cycle abnormalities observed after excision of the BHRF1 miRNAs . This supports the idea that the relative excess of PTEN seen in LCLs generated with Δ123 is responsible , partly or entirely for the observed abnormalities in cell cycle entry . We then turned our attention to infected cells that survived the critical day 5 to day 20 period and measured the expression of BHRF1 in cells established for more than 20 days . We found that , in accord with previous observations [11] , established cell lines generated with wild type controls , hardly express BHRF1 ( Fig 6a ) . In contrast , LCLs generated with Δ123 showed a clear expression of the protein . We repeated this experiment for 3 additional B-cell donors and obtained similar results that are given in Fig 6a . These results suggested that the BHRF1 mRNA that is used for translation of the protein could also be used as a template for miRNA processing . Therefore , we performed a northern blot on polyadenylated RNAs with a probe specific for the 3’ end of the BHRF1 gene ( Fig 6b ) . This assay showed that the 0 . 5 kb transcript , that can only be generated by processing of the polyadenylated BHRF1 mRNA transcript at miR-BHRF1-2 or miR-BHRF1-3 ( Fig 6c ) , was present in B-cells infected with wild type EBV but not in those infected with Δ123 , confirming that the BHRF1 mRNA is cut by miRNA processing . To find out which of the BHRF1 miRNAs is responsible for this process , we infected B-cells with the Δ1 , Δ2 and Δ3 mutants as well as with double mutants and measured BHRF1 expression in these LCLs . We found that only the LCLs infected by a virus that lacks miR-BHRF1-2 expressed higher levels of BHRF1 protein , demonstrating that processing of miR-BHRF1-2 cuts the potentially translated BHRF1 mRNA ( Fig 6d ) . We then performed a northern blot with the same cells and could confirm that the BHRF1 3’UTR is not cut in the absence of miR-BHRF1-2 . Thus , this miRNA plays an important role in the control of BHRF1 protein expression ( Fig 6e ) . In the absence of miR-BHRF1-3 , the 3’UTR was cleaved at miR-BHRF1-2 . This resulted in a slightly larger signal in the northern blot . Thus , miR-BHRF1-2 and miR-BHRF1-3 are sequentially processed resulting in a cleavage of the primary BHRF1 transcript . This is consistent with our previous observation that efficient miR-BHRF1-3 processing requires the presence of miR-BHRF1-2 [6] . The set of BHRF1 miRNA mutants gave us the opportunity to investigate the mechanisms used by the virus to express the BHRF1 protein and more generally the transcription of the BHRF1 locus in more detail . To this end , we used again northern blots to assess the nature of the transcripts that entail the BHRF1 gene with probes specific to the BHRF1 open reading frame , its intron , or to the origin of lytic replication ( oriLyt ) which is located directly 5’ of the BHRF1 locus ( S4 Fig ) . This analysis led us to propose a model in which large unspliced polyadenylated transcripts encompassing the oriLyt region and the BHRF1 ORF are a source of BHRF1 protein . These transcripts also yield BHRF1 miRNAs , whose processing give rise to smaller and smaller RNA products . Hybridization of total RNA from wild type-infected LCLs with the BHRF1 ORF probe revealed a prominent 1 . 3 kb transcript that contains the BHRF1 intron but neither oriLyt not the BHRF1 3’UTR . It also evidenced the existence of larger transcripts of very faint intensity , some of which are larger than 8 kb . This pattern reproduces what we previously observed with other wild type LCLs at 60 days and suggests that the difference in the pattern of BHRF1 transcripts observed at day 5 post-infection might be related to the fact that BHRF1 miRNAs are not fully expressed at that time , or at least not proportionally to total BHRF1 transcription ( Fig 2b ) . Indeed , in mutants that lack miR-BHRF1-1 or miR-BHRF1-2 , the 1 . 3 kb transcript shifted to transcripts that were larger than 8 kb in size and of variable intensity , depending on the mutants . In mutants that lack miR-BHRF1-2 , there was in addition a band at approximately 2 . 3 kb . Importantly , miR-BHRF1-3 is not required for the generation of the 1 . 3 kb transcript . Thus , the 1 . 3 kb transcript is generated from larger fragments through the processing of miR-BHRF1-1 and miR-BHRF1-2 , entails the BHRF1 intron but not oriLyt , and has the size of the BHRF1 RNA fragment between these 2 miRNAs . In all probability , it is identical to the transcript generated by miRNA processing that was previously identified [19 , 20] . This 1 . 3 kb transcript was already visible at day 5 , but , in contrast to what we saw in established cell lines , larger signals of equal or even stronger intensity were also present ( Fig 2b ) . However , it is important to note that the total BHRF1 transcription at 1 . 5 to 2 months post-infection is much reduced compared to the first days post-infection , suggesting that processing of the BHRF1 miRNAs leads to a disappearance of high molecular weight BHRF1 RNAs , except if the BHRF1 transcripts are in massive excess . The investigation of poly A+ mRNAs revealed that LCLs produced additional RNAs . LCLs infected with wild type controls contain one clear signal at 2 . 2 kb as well as larger , much fainter signals . The 2 . 2 kb fragment contains the BHRF1 intron , the BHRF1 ORF but not the oriLyt fragment . It is absent in LCLs infected with mutants that lack miR-BHRF1-1 or -3 but is very abundant and slightly larger in cells infected with mutants that lack miR-BHRF1-2 . The LCLs infected by viruses that lacked miR-BHRF1-1 or -2 , including the Δ123 mutant , also displayed an accumulation of transcripts ranging from 4 to larger than 10 kb , in line with the assumption that large BHRF1 transcripts serve as a source of these miRNAs . The dominant role of miR-BHRF1-2 processing was also visible in the LCLs generated with Δ13 in that they also produced the large transcripts , but at a lower expression level than those infected with viruses that lack this miRNA . These large transcripts contain the BHRF1 intron , as well as the oriLyt sequence . We conclude that the 2 . 2kb transcript is generated from large transcripts that encompass oriLyt , contains the BHRF1 intron and open reading frame and the BHRF1 3’UTR , requires the presence of miR-BHRF1-1 for its generation and accumulates in the absence of miR-BHRF1-2 . Taking into account that the distance between miR-BHRF1-1 and the end of the BHRF1 polyA tail is 2 . 2 kb , we conclude that the 2 . 2 kb RNA represents a polyadenylated transcript that begins at miR-BHRF1-1 and runs through to the BHRF1 polyA . It is important to note that this transcript runs slightly higher in the LCLs infected by Δ23 and even higher in those infected with Δ2 . This fits with the fact that the BHRF1 transcript is 46 and 126 ribonucleotides longer in the Δ23 and the Δ2 mutants , respectively , than in the wild type BHRF1 gene . Altogether , we conclude that the 2 . 2 kb polyadenylated transcript is very likely to represent a precursor form of the 1 . 3 kb transcript identified in the total RNA blot that is itself generated upon miR-BHRF1-2’s processing . The abundance of the larger BHRF1 transcripts was inversely related to the existence of the 2 . 2 kb transcripts and their existence was strongly dependent on the absence of miR-BHRF1-2 . In the absence of miR-BHRF1-1 , there were also some larger but less strongly expressed high molecular weight BHRF1 transcripts . This points to the central role of miR-BHRF1-2 in the processing of these large transcripts . We noticed that there is an inverse correlation between the expression of the BHRF1 protein and the presence of shorter BHRF1 RNA forms . This suggests that the large polyadenylated mRNAs that encompass BHRF1 are the source of the BHRF1 protein and that they are destroyed during BHRF1 miRNA processing , in particular though processing of miR-BHRF1-2 . LCLs generated with Δ3 express high levels of miR-BHRF1-2 [6] and this fits with the observation that this cell line does not express any of the BHRF1 intermediates but only their final product , the 1 . 3 kb RNA fragment . We then went on to assess the functional consequences of BHRF1 overexpression . To this end , we provoked apoptosis in LCLs established with B-cells from multiple donors and generated by infection with Δ123 or with wild type controls . This was achieved by incubating the cells with etoposide , staurosporine or simvastatin for a variable length of time and staining these cells with Annexin-V and 7AAD ( Fig 7a , 7b and 7c ) . Treatment with these drugs gave rise to massive cell death in all samples , although the killing efficiency was significantly higher in the wild type LCLs than in LCLs generated with Δ123 . We confirmed these data with a PARP cleavage assay ( Fig 7d ) . Whilst some cells infected with Δ123 retained some intact PARP after treatment with etoposide or staurosporine , this was not the case for cells infected with wild type viruses . We then performed similar experiments with B-cells infected with ΔBHRF1 and Δ123ΔBHRF1 . We found that LCLs generated with ΔBHRF1 or with wild type viruses do not significantly differ in their response to apoptotic stimuli ( Fig 7e , 7f and 7g ) . This fits with the observation that BHRF1 is expressed at very low levels in established LCLs [11] . Fig 7h , 7i and 7j show that B-cells infected with Δ123ΔBHRF1were much more sensitive to provoked apoptosis than Δ123 LCLs , and even showed more cell death upon induction than the wild type controls . This suggests that in the absence of the BHRF1 protein , infected B-cells rely on the BHRF1 miRNAs to withstand apoptotic stimuli . These results concur with our observation that at an early time point after infection , B-cells infected with a virus such as Δ123 that expresses low BHRF1 protein levels are crucially dependent on this residual activity as seen by infection with the Δ123ΔBHRF1 mutant ( Fig 3 ) . We assessed the cell cycle characteristics of LCLs that had been established for more than 35 days using a BrdU incorporation assay ( Fig 8a ) . Although the percentage of cells in S phase remained lower in LCLs infected by Δ123 relative to those generated with wild type viruses , the ratio between these numbers increased from 0 . 42 ( Fig 1c 5 . 7/13 . 5 ) to 0 . 65 ( Fig 8a 14 . 7/22 . 6 ) . We also generated growth curves with LCLs transformed by Δ123 and wild type virus and found that both types of LCLs grow at very similar rates ( Fig 8b ) . However , as seen in Fig 8c , PTEN levels remained higher in cells infected with the Δ123 mutant relative to wild type . Therefore , LCLs infected with Δ123 must have acquired additional changes in their cell cycle regulation after several weeks in culture . We screened the expression of key regulators of the cell cycle , including p53 , Rb and the CDKN family in the different types of LCLs by western blot . This analysis revealed that CDKN1/p27 is frequently expressed at lower levels in established LCLs infected by the Δ123 mutant relative to wild type controls after more than one month in culture ( Fig 8d ) . We monitored expression of p27 during the day 5 to day 18 crucial period and found that the protein levels of p27 decreased rapidly after infection to nearly disappear at day 13 p . i . However , this expression re-increased after day 18 , albeit less strongly in B-cells infected with the Δ123 mutant ( Fig 8e ) . These data argue against a direct impact of the BHRF1 miRNAs on p27 and establish a parallel between the decrease in p27 and the recovery of the cell growth rate in B-cells infected with Δ123 .
The clarification of the molecular mechanisms fine tuned by miRNAs is frequently a difficult undertaking and the BHRF1 miRNAs are no exception to that rule . CLIP-based strategies have identified potential targets of these miRNAs but the identity of the crucial proteins that they modulate remains enshrouded in mystery [17] . We have evaluated in detail the role served by the BHRF1 miRNAs during the first weeks of EBV-mediated B-cell transformation and found that cells infected with a virus that lacks the three BHRF1 miRNAs undergo on average twice as much apoptosis than cells infected with wild type controls between day 5 and day 20 . Our results are only partially concordant with an earlier report that described a massive apoptosis in cells infected with a virus devoid of the BHRF1 miRNAs during the first five days of infection , followed by a quick recovery of cell numbers 10 days after infection [4] . These data are difficult to reconcile with the fact that the BHRF1 miRNAs are fully produced only 5 to 8 days after the onset of infection [21 , 22] and we indeed did not see any differences between cells infected with the Δ123 mutant and cells infected with wild type controls during the first 5 days of infection . One difference between the studies lies in that we investigated the infected B-cells at the single cell level and used a larger panel of markers to characterize apoptotic cells . Indeed , positive staining for Annexin-V used in the study by Seto et al . is not specific for apoptosis and is found in many other modes of cell death [23] . We found that the expression of the viral bcl-2 homolog BHRF1 is modulated by the BHRF1 miRNAs in an unexpectedly complex manner . During the first days of infection , the BHRF1 protein has previously been reported to be expressed at relatively high levels and we could confirm this observation [11] . However , B-cells infected with the Δ123 virus hardly express BHRF1 , suggesting that the observed increased apoptosis after infection with Δ123 is due to a reduction in BHRF1 expression . Indeed , a virus that lacks the BHRF1 protein and a virus that lacks the BHRF1 miRNAs induce similar phenotypes in infected B-cells . Interestingly , cells infected with ΔBHRF1 not only exhibited an increased apoptotic rate but also a reduced entry into cell cycle . At the present stage of the work , it is unclear whether the cell cycle abnormalities are a consequence of increased apoptosis , or whether this reflects a specific so far unrecognized , PTEN-independent , property of the BHRF1 protein . Our results are discordant with those obtained by Altmann et al . who found that only a virus lacking both BHRF1 and BALF1 has a reduced transforming ability as a consequence of enhanced apoptosis upon infection of B-cells [12] . However , in this study , the authors showed growth curves of B-cells infected with a virus that lacks the BHRF1 protein that are reminiscent of our own results and clearly differ from those obtained with wild type virus . Moreover , these authors also discussed the possibility that a virus that lacks BHRF1 only might also induce an abnormal phenotype without reaching definite conclusions . Importantly , B-cells infected with a virus that lacks both the BHRF1 miRNAs and the BHRF1 protein undergo a strong degree of apoptosis starting at day 5 post-infection and increasing until day 20 at which time infected cells die . Thus , the phenotype induced by Δ123ΔBHRF1 is much more severe than the one induced by Δ123 . Importantly , BHRF1 is expressed , although at low levels , shortly after infection with Δ123 . This suggests that the very low amounts of BHRF1 are sufficient to limit the apoptosis in cells infected by Δ123 . However , it is important to note that the construction of the Δ123ΔBHRF1 mutant led to a complete deletion of this gene . Therefore , although it is unlikely , we cannot exclude that so far unidentified genetic elements might be altered in this mutant . Moreover , we found that the low BHRF1 expression level in established LCLs infected with wild type EBV has no significant impact on drug-induced apoptosis , as a LCL infected with a virus that lacks the BHRF1 protein but expresses the BHRF1 miRNAs is not more sensitive to provoked apoptosis than wild type controls . This underlines again the role of the BHRF1 miRNAs in the regulation of the apoptotic status in the absence of the BHRF1 protein . The observation that the phenotype induced by Δ123ΔBHRF1 is more severe than the one observed in B-cells infected with ΔBHRF1 also indicates that the function of the BHRF1 miRNAs is not subsumed by the modulation on BHRF1 protein levels . Therefore , we looked for additional targets of the BHRF1 miRNAs and the recognition that B-cells infected by Δ123 display a reduced entry in cell cycle oriented the search . We also knew that a recombinant virus lacking miR-BHRF1-3 displays some cell cycle abnormalities and therefore looked for potential miR-BHRF1-3 targets identified in the PAR-CLIP assays [6 , 17] . This led to the identification of PTEN as a protein whose expression is downregulated by miR-BHRF1-3 . This effect was visible as early as day 5 post-infection with Δ123 and persisted in established LCLs . Treatment of wild type LCLs with the PI3K inhibitor wortmannin gave rise to a decrease in cell cycle entry , reproducing the cell cycle alterations observed in B-cells infected with Δ123 and suggesting that the PI3K pathway is active in EBV-transformed B-cells , as previously suggested [24] . Brennan et al . found that inhibition of PI3K leads to a downregulation of cyclin D2 and cyclin D3 , combined with an increase in p27 that causes growth arrest [24] . PTEN is an inhibitor of PI3K , and its increased expression in LCLs is likely to reduce the PI3K activity in these cells . Therefore , downregulation of PTEN through miR-BHRF1-3 is expected to facilitate cell division in LCLs generated with wild type EBV . PTEN is also targeted by cellular miRNAs such as the miR-17~92 cluster , that is expressed in LCLs [17 , 25 , 26] . Thus , viral and cellular miRNAs seem to collaborate to downregulate its expression . Moreover , PTEN , through its repression of the PI3K/Akt pathway , also negatively modulates the apoptotic status of the cell [27] . We found that miR-BHRF1-3 is able to weakly downregulate a luciferase-PTEN 3’UTR reporter gene . However , the difficulties in expressing miR-BHRF1-3 render interpretation of these results difficult . Altogether , these results , in combination with those of the PAR-CLIP assay suggest that miR-BHRF1-3 directly targets PTEN , but other mechanisms cannot be excluded . We found that the BHRF1 miRNAs influence the dynamic expression profile of Wp-driven and BHRF1-specific transcripts . In their absence , the peak of transcription shortly after infection was delayed by about one week , after which transcription remained higher than in the wild type counterparts . We could show that BHRF1 protein expression depends on miR-BHRF1-3 and miR-BHRF1-2 seed regions and identification of their targets should shed light on their exact mechanism of action . After 30 days , the infected cells recovered a normal growth rate , as assessed by growth curves and PH3 staining . The cell cycle profile of B-cells infected with Δ123 substantially improved at this point , with an increased proportion of cells that enter the S phase , although it did not reach wild type levels . These LCLs were also more resistant to drug-induced apoptosis . Accordingly , the BHRF1 protein expression profile differed between established LCLs and freshly infected B-cells . LCLs established with Δ123 expressed more BHRF1 protein than their wild type counterparts and this explains the increased resistance to apoptosis , as LCLs infected with Δ123ΔBHRF1 or ΔBHRF1 have lost this property . We could identify miR-BHRF1-2 as the miRNA that is mainly responsible for the downregulation of the BHRF1 protein after establishment of the LCLs . The sequential processing of miR-BHRF1-2 and miR-BHRF1-3 that results in the cleavage of the 3’ end of the BHRF1 mRNA suggests that miR-BHRF1-2 does not solely act through the classical RISC-mediated target mRNA regulation , but highlights another mechanism through which a miRNA regulates protein expression . Therefore , we have shown that the latent BHRF1 mRNAs can act as a template for BHRF1 translation but can also be used by the Microprocessor machinery to generate the BHRF1 miRNAs and that both processes are in competition for access to the mRNA . The alternative use between protein translation and miRNA processing was previously reported for the cellular gene follistatin and miR-198 that are coded on the same mRNA [28] . Northern blot analyses have shown that infected B- cells produce abundant probably Wp-initiated high molecular weight mRNAs that contain the BHRF1 ORF . The exact identity of the BHRF1 transcripts that are translated remains unknown but we know that they contain the BHRF1 intron . Furthermore , analysis of established LCLs has also shown that high molecular weight BHRF1 RNAs are polyadenylated and probably also contain oriLyt-specific sequences . BHRF1 transcripts initiated at oriLyt have been previously identified and could potentially be the source of the protein [29–31] . Moreover , large polyadenylated RNAs containing intronic BamHI W sequences linked to the EBNA-LP , EBNA2 , and BHRF1 polyA sequences have been also detected in EBV-infected cells [29] . The observation that the same miRNAs can successively activate and repress BHRF1 protein expression is puzzling but we could identify clear differences between these 2 stages . The BHRF1 transcription rate is much higher at an early time point than after establishment of the LCL . At this time , Wp-driven transcription dominates and it has previously been found that this promoter drives the expression of BHRF1 [11] . Furthermore , we also found that whilst the stimulatory effects of miR-BHRF1-2 and -3 depended on their seed regions , the repressive effects of the BHRF1 cluster was mainly mediated by miR-BHRF1-2 processing . We hypothesize that , at the beginning of infection the strong Wp-driven BHRF1 transcription , boosted by miR-BHRF1-2 and -3 favors BHRF1 translation . This would imply that high transcription rates disadvantages miRNA processing or that Wp-driven transcripts are less accessible to the Microprocessor machinery . At this stage , the ratio between transcripts processed or not by the Microprocessor machinery allows BHRF1 protein synthesis . When Wp-mediated transcription ceases , BHRF1 transcription is reduced , the BHRF1 miRNAs are efficiently processed and the BHRF1 translation is reduced to a minimum . Why would infected cells first express BHRF1 and then downregulate its expression ? Our data show that BHRF1 expression protects infected B-cells against apoptosis at the beginning of the infection . The kinetic of expression of BHRF1 and of the viral latent membrane protein 1 ( LMP1 ) , two proteins endowed with anti-apoptotic properties [9 , 32] , is strikingly opposite . LMP1 is expressed at low levels shortly after infection and reaches its plateau of expression only 21 days after infection , which is exactly the period at which the increased apoptotic rate in cells infected with Δ123 normalizes [33] . Thus , it is possible that BHRF1 assumes LMP1 anti-apoptotic functions until this protein reaches its optimal expression level . However , virus-infected cells need to protect themselves from the immune response . As BHRF1 is efficiently targeted by the T cell response , its expression needs to be downregulated to a minimum when it is not anymore needed [34] . In contrast to BHRF1 , PTEN levels remained high in cells infected by Δ123 and this is likely to explain the persisting reduced entry in S phase in these cells . This fits with the concept that this protein is targeted by miR-BHRF1-3 , whose levels reach a plateau after day 5 to 8 in infected cells [21 , 22] . Nevertheless , the improvement in S phase entry in established LCLs prompted us to screen the expression of cell cycle regulators in these cells and we identified p27 as a protein whose expression is reduced after 30 days in culture . The expression of p27 varies markedly over time in LCLs and follows a complex pattern . As expected for non-dividing cells , the expression of this protein was high one day after infection . It decreased then to become hardly visible 2 weeks after infection . However , it then started to increase again but remained lower from this stage in cells infected with Δ123 viruses . The reduced expression of p27 suggests that infected cells counteract the effects of PTEN overexpression . Indeed , PTEN blocks the repressive effects of Akt on p27 and thus increases p27 expression [27] . At this stage , we cannot distinguish between a selection process that facilitates the growth of cells with low p27 in the context of increased PTEN expression and an indirect effect of the BHRF1 miRNAs . We favor the first possibility because the expression of p27 re-increases in wild type LCLs , even in the presence of the BHRF1 miRNAs and thus seems to be independent of them . In conclusion , we have identified BHRF1 , PTEN and p27 as direct or indirect targets of the BHRF1 miRNA cluster . These 3 proteins regulate the apoptotic status and entry into S phase , two essential cell functions . Although it is likely that other important targets of the BHRF1 miRNAs remain to be discovered , many of the phenotypic traits evinced by the Δ123 can be explained by the modulation of these 3 proteins . The expression of the BHRF1 miRNAs increases BHRF1 protein production and reduces PTEN production after 5 days post-infection to facilitate cell division . At a later time point , the BHRF1 miRNAs reduce expression of the BHRF1 anti-apoptotic protein and indirectly increase expression of p27 , two events associated with a reduced propensity to neoplastic cell transformation [35 , 36] that is beneficial for long-term persistence in the host . BHRF1 has gained increasing attention in recent years for its important function in LCLs and in Burkitt’s lymphomas . BHRF1 also represents a potentially very attractive therapeutic target [37] . It is therefore not surprising that its expression is tightly modulated within the infected cell . It is interesting to note that the appearance of a miRNA cluster within the BHRF1 locus is relatively new during evolution and is restricted to gammaherpesviruses with a tropism for B-cells and may be related to the need to restrict efficient MHC class I and class II presentation from B-cells [34 , 38 , 39] .
Cells were kept in RPMI-1640 ( Life Technologies ) supplemented with 10% FBS ( Sigma ) at 37°C in a 5% CO2-buffered humid atmosphere . Primary B-cells infected with EBV were kept in RPMI/20%FBS until establishment of the cell line and 100μg/ml Hygromycin B ( Calbiochem ) was added to HEK 293 producer cells to induce selection pressure in cells stably carrying the EBV-BAC . HEK 293 cells are neuro-endocrine cells obtained by transformation of embryonic epithelial kidney cells with adenovirus ( ATCC: CRL-1573 ) . Primary B-cells were isolated from adult human blood buffy coats by Ficoll ( GE healthcare ) density gradient centrifugation and the CD19+ B-cell population was purified using CD19 PanB Dynabeads ( Life technologies ) and DETACHaBEAD CD19 ( Life technologies ) . Elijah-5E5 is an EBV-negative subclone of the EBV positive Burkitt’s lymphoma cell line Elijah ( kindly provided by A . B . Rickinson ) . BJAB is an EBV-negative Burkitt’s lymphoma cell line ( kindly provided by A . B . Rickinson ) [40] . Oku-BL is a Wp-restricted Burkitt’s lymphoma cell line ( kindly provided by A . B . Rickinson ) [10] . WI38 are primary human lung fibroblasts ( ATCC: CCL-75 [41] ) . All human primary B-cells used in this study were isolated from anonymous buffy coats purchased from the Blood Bank of the University of Heidelberg . No ethical approval is required . The recombinant EBV wild type BAC ( B95-8 wt/2089 ) was constructed by introducing the bacterial F-factor , GFP , a chloramphenicol resistance gene and a hygromycin selection marker in the EBV strain B95-8 [42] . The construction of the miR-BHRF1 deletion mutant and revertant [5] , of the miR-BHRF1 single mutants [6] and of the BZLF1 deficient virus ΔZ [43] were reported previously . The Δ123ΔBHRF1 mutant was obtained by exchanging the complete BHRF1 locus of the B95-8 wild type BAC ( EBV coordinates 53758:55278 ( V01555 . 2 ) ) with a kanamycin resistance cassette by homologous recombination [44] . This cassette was amplified from pCP15 using the primers 1422 ( ATGTGGGGGT GGAAATATGA GCAAGAATAA GGACGGCTCC AACAGCTATG ACCATGATTA CGCC ) and 683 ( ATTTTAACGA AGAGCGTGAA GCACCGCTTG CAAATTACGT CCAGTCACGA CGTTGTAAAA CGAC ) . We used En Passant Mutagenesis [45] to construct a BHRF1 deficient recombinant virus ( ΔBHRF1 ) in which the BHRF1 ATG start codon ( 54376:54378 ( V01555 . 2 ) ) was replaced by ATTAG in GS1783 , a bacterial strain that contains the B95-8 wild type BAC . To amplify the kanamycin resistance gene from the pepKanS plasmid , the primers 1855 ( CCTCTTAATT ACATTTGTGC CAGATCTTGT AGAGCAAGAT TAAGTAGGGAT AACAGGGTAA TCG ) and 1856 ( TATACACAGG GCTAACAGTA TCTCCCTTGT TGAATAGGCC TAATCTTGCT CTACAAGATC TGGCACAAAT GTAATGCCAG TGTTACAACC AATTAACC ) were used . To generate the 3SM seed mutated recombinant virus , three point mutations were introduced in the seed region of miR-BHRF1-3 ( AACGGGA converted to AACGTTG ) . To this end , the nucleotides in the mature seed as well as the complementary miR-BHRF1-3* strand ( 55261:55263 and 55307:55309 ( V01555 . 2 ) ) were mutated using En Passant mutagenesis . The kanamycin resistance gene from the pepKanS plasmid was amplified using the primers 2125 ( CAATTGGGTG TCCTAGGTGG GATATACGCC TGTGGTGTTC TAACGTTGAG TGTGTAAGCA CACACGTAAT TTGCAAGCGG ATAAGTAGGG ATAACAGGGT AATCG ) and 2126 ( CTCAGTTATT TCTTTAGTAT CTTGTCCTTG TGTTATTTTA ACGCCAAGCG TGAAGCACCG CTTGCAAATT ACGTGTGTGC TTACACACTC AACGTTAGCC AGTGTTACAA CCAATTAACC ) and introduced into B95-8 Δ3 in GS1783 . Successful construction of all clones was verified by sequencing of the mutated region and integrity of the complete genome was confirmed by restriction enzyme digestion of the BAC clones and of the stably transfected HEK 293 producer cells . Induction of HEK 293 producer cells to generate virus supernatants for B-cell infection and the quantification of virus titers were performed as previously described [5] . 106 primary B-cells were infected using a MOI of 10 EBV genome equivalents per cell for 2 hrs at RT and cultured in RPMI/20%FBS . Four to five days after infection with EBV , B-cells initiate permanent growth that gives rise to the establishment of cell lines termed lymphoblastoid cell lines ( LCL ) . For transformation assays , 5×102 cells were infected with a MOI of 0 . 01 , that is enough viruses to induce GFP expression in 1% of infected cells [5] , for 2 hrs at RT and plated on 96-well plates coated with 50Gy irradiated WI38 fibroblasts . 48 wells per donor and virus supernatant were seeded and the number of transformed wells was determined 30 days post-infection . Cells were washed once in 1xPBS , spread on glass slides ( Medco ) and air-dried . For phospho-histone 3 staining ( PH3; 1:100; Cell Signaling ) , cells were fixed in 4% PFA for 20 min at RT , washed 5 min in 1xPBS , permeabilized by immersion in 1xPBS/0 . 5% Triton-X for 2 min and washed again for 5 min in 1xPBS . Cells were incubated with the primary antibody diluted in PBS/10%HINGS ( heat inactivated; Gibco ) for 30 min at 37°C in a humid chamber and washed 3 times 5 min in 1xPBS . Secondary goat anti-rabbit Cy3 conjugated antibody ( 1:1200; Dianova ) was applied for 30 min at 37°C in a humidity chamber , cells were washed 3 times for 5 min in 1xPBS and the DNA was counterstained using Hoechst 33258 for 2 min at RT . The cells were embedded in 90% glycerol and analyzed by fluorescence microscopy ( Leica DM5000 B ) . A triple staining of the mitotic spindle ( α-tubulin ) , the centromeres ( centrin-2 ) and the DNA ( DAPI ) was used on cells flattened by cytospin to analyze mitosis . To this end , cells were harvested , washed twice in 1xPBS/3% FBS and 5×104 cells were spun on slides ( Tharmac , Cytoträger ) in 100μl PBS/3%FBS using the cytospin 4 ( Thermo; EZ single cytofunnel , Thermo ) for 10 min at 2000rpm , maximum acceleration . Cells were air-dried , fixed in PFA and permeabilized using PBS/Triton-X as described above . Unspecific protein binding was blocked for 45 min in PBS/3%BSA at room temperature in a humid chamber . Cells were stained with rabbit α-centrin-2 ( 1:100; Santa Cruz ) and mouse α-α–tubulin ( 1:4000; Sigma ) in PBS/3%BSA for 2 hrs at 37°C in a humidity chamber , washed 5 times in 1xPBS , followed by incubation for 2 hrs in the secondary antibodies diluted in PBS/3%BSA ( goat α-mouse IgG-Alexa488 , 1:300 , Invitrogen; goat α-rabbit Cy3 , 1:1200 , Dianova ) . After washing 5 times in 1xPBS , the cells were embedded in ProLong Gold antifade reagent with DAPI ( Life technologies ) and analyzed at a magnification of 630x . For BrdU incorporation assays , cells were adjusted to 5×105 cells per ml one day prior to cell cycle analysis . 5×105 cells were pulsed with 10μM BrdU for 35 min at 37°C and stained using the APC BrdU Flow Kit ( BD Pharmingen ) according to the manufacturers instructions . The cell cycle profile was determined with a FACSCalibur flow cytometer . This assay allowed recognition of cells in G1/G0 ( BrdU negative , 7AAD single DNA content ) , S phase ( BrdU positive , shifting from single to double DNA content ( 7AAD ) ) and the G2/M phase ( BrdU negative , double DNA content ( 7AAD ) ) . The growth rate of established LCLs was determined using the trypan blue ( Sigma ) dye exclusion method . The number of viable cells was determined 24 hrs and 48 hrs after adjusting the cultures to a density of 3x105 cells per ml . Cells were harvested at indicated time points , washed once in 1xPBS , spread on glass slides , fixed and stained for cleaved caspase 3 ( Casp3; 1:400; Cell Signaling ) as described above for PH3 and single cells were analyzed by fluorescence microscopy . For detection of apoptosis using the TUNEL technology ( terminal deoxynucleotidyl transferase ( TdT ) -mediated dUTP nick end labeling ) , cells were washed once in 1xPBS , spread on glass slides ( Medco ) , the DNA double strand nicks were enzymatically labeled with the In Situ Cell Death Detection Kit , TMR red ( Roche ) according to the manufacturers instructions and analyzed by fluorescence microscopy . Following induction of apoptosis , cell death was determined using Annexin-V-Alexa647 ( Roche ) . To this end , 106 cells were pelleted , washed once in 1xPBS , incubated for 15 min at room temperature in 100μl Annexin-V-binding buffer ( Annexin-V ( 3μl/106 cells ) , 7AAD ( eBiosciences; 4μl/106 cells ) in 10mM Hepes pH 7 . 4 , 140mM NaCl , 5mM CaCl2 ) . Annexin-V positive cells were quantified using a FACSCalibur flow cytometer . Apoptosis was induced in LCLs of different B-cell donors 2–3 months after infection . Cells were treated with etoposide ( 4μg/ml , Sigma ) , staurosporine ( 4μg/ml , Sigma ) or a DMSO solvent control for 20 hrs . We also used simvastatin ( 2mM , Calbiochem ) or an ethanol solvent control for 5 days [46] . After treatment , cells were subjected to western blotting ( PARP ) or Annexin-V viability staining . We used a 10nM concentration of 17β-hydroxy wortmannin ( Cayman Chemical ) to inhibit the PI3K in EBV wild type-infected primary B lymphocytes 14 days post-infection . Cells were treated for 17 hrs and the cell cycle profile was determined using a BrdU incorporation assay as described above . Cells were harvested , washed once in 1xPBS , lysed in RIPA buffer supplemented with a protease inhibitor cocktail ( 1:1000 , Sigma ) and sonicated . 50μg of protein were separated on a 7 . 5% ( PARP1 ) or 15% ( p27 , BHRF1 , PTEN , Actin ) SDS-polyacrylamide gel and electroblotted on a protran membrane ( Amersham , 0 . 45 NC ) by wet blotting . Incubation with primary and secondary antibodies was performed as described previously [5] using primary antibodies specific for PTEN ( 1:8000 , Abcam ) , BHRF1 ( 1:100 , kindly provided by J-Y . Chen [47] ) , p27 ( 1:1000 , Santa-Cruz ) , PARP1 ( 1:1000 , Cell Signaling ) and Actin ( 1:10000 , Dianova ) . Total RNA was isolated using TRIzol ( Life technologies , 1ml per 107 B-cells ) according to the manufacturers protocol . polyA+ RNA was isolated by hybridization of the total TRIzol purified RNA to oligo-dT-coupled latex beads using the nucleotrap mRNA mini isolation kit ( Machery-Nagel ) according to the manufacturers instructions . Northern blots of polyA+ RNA and total RNA were performed by separating 1 . 5μg or 7 . 5μg RNA , respectively , alongside with 5μl RNA marker ( 0 . 5-10kb RNA marker , Life Technologies ) on a denaturing 1% agarose/1xMOPS gel containing 2 . 2M formaldehyde for 5 hrs at 100V . RNA was transferred to a Hybond-XL membrane ( Amersham ) by capillary blotting over night in 10xSSC and cross-linked to the membrane by baking for 2 hrs at 80°C . The blot was hybridized with 50ng of a [32P]-α-dCTP ( Perkin Elmer ) radiolabeled DNA probe specific for the 3’UTR of BHRF1 ( coordinates 55389:55567 in the EBV reference genome V01555 . 2 ) , the BHRF1 open reading frame ( coordinates 54360:54853 ( V01555 . 2 ) ) , the BHRF1 intron ( coordinates 53953:54359 ( V01555 . 2 ) ) , or the left origin of lytic replication ( coordinates 53351:53756 ( V01555 . 2 ) ) . Labeling was obtained by using the random primed DNA labeling kit ( Roche ) . The blot was hybridized over night at 65°C in Church buffer , washed 4 times in 0 . 1%SDS/1xSSC and exposed to Hyperfilm-MP ( Amersham ) at -80°C as indicated in the figure legends . For microRNA northern blots , 20μg of total RNA per sample was separated on 15% Mini-Protean TBE-urea acrylamide gels ( BioRad ) at 80V for 2 hrs in 0 . 5xTBE ( Ambion ) . RNA was transferred on a Hybond N+ membrane ( Amersham ) by semi-dry blotting for 2 , 5 hrs at 250mA and UV-crosslinked to the membrane ( 1200μJ ) . The blot was hybridized with 20pmol of an [32P]-γ-ATP ( Perkin Elmer ) labeled oligonucleotide ( MWG Eurofins ) complementary to the seed-mutated mature miR-BHRF1-3 ( TGTGCTTACACACTCAACGTTA ) or the seed-mutated mature miR-BHRF1-2* of the 2/2*DSM recombinant ( GCAAACGGCTGCAACAACGTTT ) at 37°C for 1h in ExpressHyb solution ( ClonTech ) . Blots were washed twice at 37°C in 2xSSC/0 . 05%SDS , and twice in 0 . 1xSSC/0 . 1% SDS at room temperature and exposed to Hyperfilm-MP ( Amersham ) at -80°C for 7 days . MicroRNAs were quantified using stem-loop RT qPCR [48] and miR-BHRF1 specific primers as described previously [49] . 110ng total RNA was reverse transcribed with the TaqMan miRNA reverse transcription kit ( Applied Biosystems ) using 12 . 5μM of each RT primer . Per sample , 10ng of template were mixed with 1 . 5μM forward primer , 0 . 7μM reverse primer , 0 . 2μM probe and TaqMan universal PCR mastermix ( Applied Biosystems ) . The samples were incubated at 50°C for 2 min , 95°C for 10 min , followed by 40 cycles of 95°C for 15 sec and 56°C for 1 min using a StepOnePlus real-time PCR system ( Applied Biosystems ) . All samples were measured in duplicate and the RNU48 TaqMan microRNA control assay ( Applied Biosystems ) was used as internal control for normalization of all samples . To quantify different BHRF1 containing transcritps and the Wp promoter activity , 400ng total RNA was reverse transcribed using the AMV reverse transcriptase ( Roche ) with a mix of RT primers specific for GAPDH , combined with primers specific for the W2W1 exon junction or BHRF1 [11 , 50] . qPCR was performed on 20ng of template using the TaqMan universal PCR mastermix ( Applied Biosystems ) and incubated at 50°C for 2 min , 95°C for 10 min , followed by 40 cycles of 95°C for 15 sec and 60°C for 1 min on a StepOnePlus real-time PCR system ( Applied Biosystems ) . The activity of the Wp promoter and the expression of W2-BHRF1 spliced transcripts were quantified as described earlier [11 , 50] . For the quantification of all BHRF1 transcripts , RNA was reverse transcribed using the universal BHRF1 RT primer [11] , and amplified using 0 . 3μM forward primer ( CCCTCTTAAT TACATTTGTG CCAGAT ( 54337:54362 ( V01555 . 2 ) ) , 0 . 3μM reverse primer [11] , and 0 . 2μM of the Fam-labeled probe ( TAGAGCAAGA TGGCCTATTC AACAAGGGAG A ( 54367:54397 ( V01555 . 2 ) ) . The VIC-labeled human GAPD ( GAPDH ) endogenous control ( Applied Biosystems ) was used as internal reference for normalization , All samples were measured in duplicate . Oligonucleotide primers ( MWG Eurofins ) encoding part of the wild type 3’UTR of PTEN ( coordinates 102288:102331 ( NG_007466 . 2 ) ) or a seed-matched mutant PTEN 3’UTR ( coordinates 102288:102331 ( NG_007466 . 2 ) in which the position 102311:102318 ( NG_007466 . 2 ) was replaced for a stretch of eight adenines were annealed and introduced in the 3’UTR of the firefly luciferase reporter plasmid pGL4 . 5 ( Promega ) , which had been modified to contain an EcoR1 and Xho1 cutting site 3’ of the luc2 open reading frame . Constructs were confirmed by sequencing . HEK 293 cells were seeded at a density of 7*104 cells per well in a 24-well cluster plate . The following day , 210ng of the wild type PTEN 3’UTR firefly luciferase fusion or of the seed-matched mutant control plasmids , and 840ng of miR-BHRF1-3 ( 55198:55395 ( V01555 . 2 ) in pcDNA3 . 1 ( + ) ) or of the empty vector control pcDNA3 . 1 ( + ) ( Invitrogen ) , and 210ng of a pRL-SV40 plasmid ( Promega ) encoding the renilla luciferase to control for differences in cell numbers and transfection efficiency were cotransfected using 3μl metafectene per μg of plasmid DNA . The activity of the firefly and renilla luciferase were determined 2 days after transfection using the dual-luciferase reporter assay system ( Promega ) according to the manufactures instructions and measured using a Fluoroskan Ascent FL luminometer ( Thermo Scientific ) in triplicate measurements for each sample . GraphPad Prism 6 was used to conduct all statistical analysis . The error bars represent the standard deviation of the data sets . Statistical significance was determined using the student's t-test and all data that were derived from LCLs generated from the same blood donor were analyzed as paired samples . | This paper explains some of the molecular mechanisms used by the Epstein-Barr virus ( EBV ) BHRF1 microRNA cluster to enhance transformation of B-cells after infection . We find that B-cells exposed to a virus that lacks the BHRF1 microRNAs ( Δ123 ) undergo more apoptosis and grow more slowly between the second and the fourth weeks after infection than cells infected by an intact virus . These effects are partly mediated by the viral protein BHRF1 , a homolog of the anti-apoptotic bcl-2 protein . The viral microRNAs allow abundant expression of BHRF1 early after infection and its down-regulation when transformation has been established . The first effect is mediated by the seed regions of miR-BHRF1-2 and -3 , whereas the second is dependent on RNA cleavage mediated by processing of miR-BHRF1-2 . Furthermore , we found that the ability of the BHRF1 microRNAs to increase cell cycle entry is related to their ability to downregulate PTEN , a crucial negative regulator of the cell cycle . We also study the consequences of the absence of the microRNAs for the infected cells . B-cells infected with Δ123 become more resistant to apoptosis and express lower levels of p27 , two events that facilitate the development of genome instability . Thus , the viral microRNAs allow rapid and innocuous expansion of infected B-cells , their long-term reservoir , thereby facilitating the life-long coexistence between the virus and its host . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [] | 2016 | A Viral microRNA Cluster Regulates the Expression of PTEN, p27 and of a bcl-2 Homolog |
In a wide range of studies , the emergence of orientation selectivity in primary visual cortex has been attributed to a complex interaction between feed-forward thalamic input and inhibitory mechanisms at the level of cortex . Although it is well known that layer 4 cortical neurons are highly sensitive to the timing of thalamic inputs , the role of the stimulus-driven timing of thalamic inputs in cortical orientation selectivity is not well understood . Here we show that the synchronization of thalamic firing contributes directly to the orientation tuned responses of primary visual cortex in a way that optimizes the stimulus information per cortical spike . From the recorded responses of geniculate X-cells in the anesthetized cat , we synthesized thalamic sub-populations that would likely serve as the synaptic input to a common layer 4 cortical neuron based on anatomical constraints . We used this synchronized input as the driving input to an integrate-and-fire model of cortical responses and demonstrated that the tuning properties match closely to those measured in primary visual cortex . By modulating the overall level of synchronization at the preferred orientation , we show that efficiency of information transmission in the cortex is maximized for levels of synchronization which match those reported in thalamic recordings in response to naturalistic stimuli , a property which is relatively invariant to the orientation tuning width . These findings indicate evidence for a more prominent role of the feed-forward thalamic input in cortical feature selectivity based on thalamic synchronization .
Sensory systems serve the purpose of allowing us to extract perceptually relevant features from the environment . Although there are certainly examples of sensory features whose coding originates in the sensory periphery ( e . g . auditory frequency , visual color , etc . ) , the more intriguing and less well understood phenomena involve the emergence of feature selectivity in more central brain structures that do not just inherit the selectivity from the periphery . Perhaps the most well studied of these phenomena is that of orientation selectivity in primary visual cortex ( V1 ) , where many if not most neurons in the mammalian primary visual cortex exhibit differential firing activity for visual stimuli at different orientations , despite the fact that the neurons projecting from the lateral geniculate nucleus ( LGN ) serving as input to V1 exhibit little to no orientation preference on their own [1] ( see [2] for a review ) . This implies that the thalamocortical link is a transformative location for representation of stimuli as collections of particular features rather than samples ( i . e . it does far more than simply relay luminance values to the cortex ) . This transformation can serve as a general model for how sensory systems convey increasing feature selectivity as the information moves to higher-order brain areas . How do these convergent thalamic structures drive cortical feature selectivity , and in what way do populations drive this selectivity ? The mechanistic origin of orientation tuning in V1 has been vigorously explored in the literature [1]–[5] . In their seminal work , Hubel and Wiesel outlined a conceptual model that involved the projection of LGN neurons along a particular axis of orientation to a common cortical target [1] , the core connectivity of which was subsequently confirmed in recordings from connected pairs of neurons in LGN and V1 [6]–[8] . Although the relative roles of this feedforward architecture versus cortico-cortico connectivity in sharpening and refining orientation selectivity in such phenomena as contrast-invariance and cross-orientation suppression has been intensely debated [2] , [9] , the thalamic basis for the origin of the basic selectivity is not in dispute , and by its nature implies a role for the timing of thalamic inputs to the cortical target . That is , the several decade old proposal by Hubel and Wiesel conceptually suggests that an edge activating the subset of thalamic neurons projecting to a common cortical target at the same time would naturally drive the cortical neuron more so than when the thalamic inputs are activated at different times , establishing the orientation tuning for the cortical neuron . However , the precise role of timing of thalamic inputs in the downstream cortical orientation selectivity is not known . In the context of the natural visual environment , it has been shown that LGN neurons ( individually and across pairs ) are temporally precise to a time scale of 10–20 ms , a level that is matched to what is necessary to capture the timescale of changes exhibited in natural scenes [10]–[12] . Further it has been demonstrated that neurons in the primary visual cortex are extremely sensitive to short intervals between incoming thalamic spikes also on the time scale of approximately 10 ms [13]–[22] and that common cortical convergence is most probable when receptive fields overlap [7] , [13] . All of these findings collectively suggest that feature selectivity is likely to arise from the modulation of precise timing among overlapping populations of neurons in LGN and that this modulation drives the coactivation of neurons within the populations . Finally , we have recently shown that considering just the coactivation between pairs of electrophysiologically recorded thalamic neurons reveals in many cases extremely sharp orientation tuning even when the receptive fields are highly overlapped [23] . Here , to explore the role of the precise timing of thalamic spiking in the orientation tuning of the downstream cortical neurons to which the thalamus projects , we utilized experimental population recordings of single units from the LGN region of the visual thalamus in concert with a large-scale thalamocortical model . Specifically , based on anatomical and physiological evidence concerning the convergence of thalamic input to cortical layer 4 , we constructed thalamic sub-populations from experimentally recorded thalamic spiking in response to oriented visual stimuli , and systematically controlled the precise timing across the sub-population and its direct impact on the downstream orientation tuning . We found that the conventionally measured tuning sharpness was remarkably invariant over a wide range of peak LGN timing precisions , but the trial-to-trial variability in cortical response was strongly influenced by the timing precision of the LGN inputs . From a decoding perspective of an ideal observer of the cortical response , this complex relationship led to a decreasing error in estimation of orientation with increasing thalamic precision , and a corresponding increase in the information rate , both saturating for peak thalamic precisions of 10–20 ms , a finding which was invariant to the overall width of cortical orientation tuning . Taken together , the results here provide a compelling picture for the role of stimulus-driven thalamic synchrony in the emergence of cortical feature selectivity .
Neurons in layer 4 of primary visual cortex are driven by sub-populations of projecting LGN neurons with receptive fields that are highly overlapped , thus representing a relatively limited area of visual space [24] . Although individual LGN neurons are relatively insensitive to the orientation of drifting sinusoidal gratings , the synchrony across neuron sub-populations is often highly sensitive to the orientation , a product of the relative spatial geometry of the receptive fields and the underlying temporal dynamics of component neurons [23] . LGN populations which share a convergent cortical neuron are both large ( approximately 30 neurons [8] ) and highly overlapped . Since it is not currently possible to record from such dense and numerous clusters in the LGN , we implemented a population-filling method to quantify the synchronization properties of the sub-population . Specifically , in the population-filling method we utilized simultaneous recordings of spiking activity of small sub-populations of LGN neurons whose receptive fields span a small area of visual space ( see Methods ) . Single unit activity was collected in response to spatiotemporal white noise , and receptive fields ( RFs ) were mapped using standard spike-triggered averaging ( see Methods ) . The RFs of a pool of simultaneously recorded LGN neurons are shown in Figure 1A , where the RF for each neuron is represented as the 20% contour . Note that in this recording , 5 neurons were recorded simultaneously , where each of these neurons is represented as a different color in the figure . We have previously provided experimental measures of the distribution of receptive field spacing of pairs of LGN neurons monosynaptically connected to a single cortical cell [8] and populations of LGN neurons to a single cortical orientation column [24] , as shown with the dashed gray curve in Figure 1B . Specifically , this measure provides a probability distribution of the distances between receptive fields , as measured by the distance between the RF centers normalized by the diameter of the larger of the two RFs , referred to here in units of receptive field center diameter ( RFCD ) - see [24] . From experimental data in [24] , the distribution of separations was modeled as , where x is the separation in units of RFCD , which is described only for the range of 0 . 4 to 2 . 0 . Using the neurons in Figure 1A as templates and the relationship in Figure 1B ( dashed line ) as a rule , we filled out the assumed remainder of the population by translating the receptive fields in visual space , creating a dense and accurate convergent LGN population , as shown in Figure 1C . The receptive field centers were randomly shifted such that the amount of visual space covered did not change relative to the visual space covered by the original simultaneously recorded population . This method resulted in a distribution of RF separations consistent with previous experimental findings ( simulated distribution shown with solid black circles , Figure 1B ) . Note that because the original population was itself elongated in the horizontal axis , the resultant shifts for this population were also mostly horizontal although some receptive field locations also moved vertically . The resultant cluster of receptive fields would be typical for a population that has a major and minor axis as opposed to being more circularly arranged . The resulting aspect ratio of the cluster of RFs in Figure 1A is approximately 2 . 4∶1 , when measured as the ratio of the longer dimension to the shorter dimension of the area covered by the RF contours . It is important to note that this aspect ratio is lower than the majority of existing models [1] , [3]–[5] , where aspect ratios range from 3 to 4 ( but see [5] for a much smaller aspect ratio ) . Spiking activity was also collected in response to drifting sinusoidal gratings ( 0 . 5 cycles/degree , 5 Hz , 100% contrast - see Methods ) . The individual LGN neurons had mean firing rates that ranged from 16 to 28 Hz which were relatively insensitive to the stimulus orientation . To generate the population activity in response to the drifting gratings , we utilized the spatially translated RFs as described above , and imposed temporal shifts in the spiking activity based solely on the geometry related to the RF centers , as illustrated in Figure 1D . Specifically , a spatial translation of the RF by x degrees horizontally and y degrees vertically imposes a latency shift of the neural response by an amount proportional to the component of the vector connecting the centers of the two RFs orthogonal to the edge of the drifting grating , scaled by the speed of the drift ( see Methods ) . For the collected datasets , spiking activity was collected at each of eight drifting directions with sinusoidal gratings . For each stimulus condition , each randomly placed neuron was assigned a random trial from the original neuron from which it was derived and the shift latency value was added to all spike times in the chosen trial . In this spirit , we view the trial to trial variability in spiking activity for a single neuron as representative of the across neuron variability on a single trial . The resulting population response at each orientation is shown in Figure 1E . For most orientations , spike times within the population uniformly distributed across the entire trial timespan . However , at 90 and 270 degrees , the spike times line up rather precisely between all neurons in the population , reflecting a high degree of synchrony at these orientations . The degree of synchrony across this population of neurons is a function of the orientation of the drifting gratings , as well as the variability in spiking timing across neurons within the population . To quantify the synchrony , we used a timing jitter metric , which utilizes the width of the spike-time auto-correlation computed from all spikes in the population ( roughly equivalent to the PSTH width ) . A brief overview of how the auto-correlation is calculated is demonstrated in 2A . The collection of spike times across the input population is collapsed into a single spike train , which represents all the projecting thalamic input on the cortical target neuron . This spike train is then used to calculate all of the pair-wise timing differences between every input spike in the population , the histogram of which forms the auto-correlation estimate . There are two values of interest: the population PSTH ( with a width of ) and the “response timescale” of the auto-correlation function ( given by ) . These related values provide us with an approximation for the synchronization within the neural population . When synchrony is high , the spike time auto-correlation has a narrow width and thus there is little jitter . Alternatively , when synchrony is low , the auto-correlation has an increased width and jitter is very high , a property that is demonstrated in Figure 2B . From top to bottom in the figure , the level of synchrony in the population increases , spike times become more clustered , and the auto-correlation has a correspondingly decreasing width . Note that each auto-correlation covers the lag range from −400 ms to +400 ms . Each auto-correlation function was fit with a Gaussian between −100 and +100 ms to eliminate any effects of periodicity in response to the drifting sinusoidal grating . The corresponding width of this Gaussian fit was then utilized as the measure of timing jitter . As in [10] , the timing jitter was defined as the half the latency at which the Gaussian fit is equal to ( see Methods and Figure 2A ) . The timing jitter of the population is shown as a function of the stimulus orientation in Figure 2C , where the random sampling of single trials of the template neuron was repeated 50 times . At the most asynchronous stimulus orientations ( in this case perpendicular to the elongated axis of the RFs of the population ) , the timing jitter was approximately 100 ms . At the preferred orientations , when synchrony was maximized , the timing jitter was approximately 24 ms . The timing jitter as a function of stimulus orientation was fit with a Gaussian function ( gray dashed line in Figure 2C ) and exhibited a characteristic tuning width of approximately 31 degrees ( standard deviation ) , a finding which was consistent for two of the three animals . In the third animal there was an insufficient number of strongly-driven neurons with identical polarities ( ON- versus OFF-center ) to allow for a reasonable reconstruction of a population with more than 2 or 3 neurons . With so few neurons , the population displayed more and more properties of the response of a single neuron as opposed to a rough average of multiple neurons and the overall orientation tuning decreased as the population approached the orientation-agnostic response properties of a single input neuron . To determine the generality of our findings here , we utilized other metrics from previously published studies , with a focus on the reliability method used in [25] which is easily adaptable to population data . We found that qualitatively the results were similar to our own findings; just as jitter decreases in our sample population at 90 and 270 degrees ( Figure 2C ) the reliability across all the neurons in the population is significantly higher at 90 and 270 degrees . We thus expect that the synchronization observed across all neurons in the population is not affected by the metric chosen to measure it . By construction , the degree of synchrony across the population of neurons in Figure 1D is a function of the orientation of the drifting gratings and across neuron variability in spiking , independent from geometry . The across neuron variability in timing thus set the lower bound of timing jitter in Figure 2C . To more fully explore the role of synchrony in shaping the feature selectivity in the downstream cortical response , we effectively replaced the across-neuron variability in spike timing with variability under our control . Specifically , we utilized a single trial spike-train for a template neuron and introduced the latency associated with the translation of the receptive field as in Figure 1D , but subsequently added variability to each spike time in the form of a Gaussian random variable with zero mean and variance . So long as the population firing rate reaches a particular minimum mean level it does not matter which template neuron is chosen to provide the spike train; we found that nearly all neurons from all three animals provided consistent simulations of cortical activity . Using a single trial has the effect of removing the effects of variable spike count across trials for a particular neuron in addition to providing the exact control over the timing jitter . Of key importance is the value , which is the stimulus-dependent component of timing jitter ( see Methods for expanded description ) . This value is related to but not equal to the timing value measured from the full populations; represents the underlying stimulus-based modulations to synchrony that give rise to the full timing jitter relationship shown in Figure 2C . This timing variability quantity was parameterized as a Gaussian function of and was manually tuned to reproduce the population timing variability curve in Figure 2C . From here on out , when we refer to “minimum timing jitter” we are referring to the minimum value of that occurs at the preferred orientation . To determine how different levels of input synchrony affect the downstream cortical response and the corresponding feature selectivity , we simulated the cortical layer 4 neuron response to the drifting gratings at different orientations . The previously described populations were used as input to this model , modulating the minimum value of to cover a range of 6 to 40 ms of population timing jitter . To model the cortical neuron , we used a biophysically inspired integrate and fire model — illustrated in 3A — that generates a continuous membrane potential and corresponding firing activity , similar to that in [22] and [23] - see Methods . In brief , the model lumps all input spike times together in a common spike train , laying down a superimposed EPSC for each input spike ( all of which thus have equal weighting ) . This model is represented by the differential equationwith a fixed parameter set to determine the point by point membrane potential and with a fixed time step of 0 . 05 ms . Membrane potential traces show a clear stimulus-driven modulation [26]–[28] that increases in amplitude towards the population's preferred orientation when averaged over 1000 trials , as shown in Figure 3B . Single trial responses , with the exception of the nonphysiological mechanics of the hard reset , match typical recordings from cortical neurons using examples from Carandini & Ferster [29] as a primary source for comparison . Further , the tuning properties ( firing rate and tuning half-width at half-height ) match reported values , as will be shown later . The reset mechanics did not adversely affect the accuracy of the results as the spiking statistics and tuning curves were consistent with experimental observations . Cortical spike counts , as shown in Figure 3C rastergrams , increased dramatically as the stimulus approached the preferred orientation , and the underlying stimulus driven events became very clear . Again , these spike count rastergrams are representative of what would be expected from cortical neurons , although this is easier to see in the cortical tuning curves . By construction of the thalamic input , the model generated cortical responses that exhibited orientation selectivity . Although the original experimental data was collected only for 8 grating orientations , the parameterized construction described in Figure 2 allowed simulation at an arbitrarily fine grain ( chosen to be at 1 degree increments here ) . The resulting mean cortical firing rate across all orientations for a minimum jitter of 6 ms is shown in Figure 4A , which is stereotypical of recorded responses of neurons in the primary visual cortex [29] , with higher firing rates possible when using different neurons for thalamic spike times . The cortical firing rate as a function of stimulus orientation was fit with a local Gaussian over a 180 degree span , as shown with the dashed curve . The parametric fits for each of a range of minimum jitter cases are shown in Figure 4B . The colors indicate decreasing levels of synchrony with dark red representing high synchrony ( 6 ms of jitter ) and dark blue representing low synchrony ( 40 ms of jitter ) . The overall magnitude of the cortical response decreased with increasing amounts of jitter , as reflected in the overall amplitude of the tuning curves . The sharpness of orientation tuning is quantified though the half-width at half-height ( HWHH ) of the tuning curve [29] , [30] . Consistent with reported values for firing rate , the HWHH tuning width for firing rate was approximately 15 to 16 degrees and was relatively insensitive to the LGN input synchrony ( Figure 4C ) up until 35 ms of input jitter at which point the tuning width increases by approximately 1 . 5 degrees . These values are on the lower end of expected tuning widths [9] , [29] , [30] . Carandini & Ferster [29] noted that due to experimental limitations they cannot discriminate half-widths less than 17 degrees , a value that they find for almost all recorded neurons . On the other hand different studies [31] , [32] have reported tuning widths with significant numbers of neurons with small ( 10–15 degree ) tuning widths . Note that the primary results of the analysis were relatively invariant to the actual tuning width , as we will demonstrate later . The tuning curve is illustrative to see how well a particular stimulus orientation drives a cortical neuron but by itself it does not convey any context as to how well the cortical neuron transmits information about the stimulus . Synchrony clearly modulates the overall amplitude of this tuning but it is unclear how it modulates the transmission of the underlying stimulus information . The ability of an ideal observer of neural activity to extract meaningful information regarding the features of a visual stimulus depends not only on the shape of the tuning curve , but also on the variability of the cortical response and how this variability changes with the stimulus feature . The statistics of the cortical response are summarized in Figure 5 . In Figure 5A , the underlying relationship between the mean and variance of the cortical spike count for all stimulus orientations ( each individual dot ) is illustrated . The relationship clearly demonstrates an increase of spike count variance relative to spike count mean with a slope of approximately 3 , which begins to drop when the input is relatively synchronous ( 6–10 ms of jitter ) . The variance begins to drop at extreme levels of synchrony as the decreased amount of added timing variance approaches the size of the integration window of the model , and higher synchrony values effectively make the spike count more deterministic . With respect to the relationship between the mean and variance of the cortical response , experimental results have been variable , exhibiting both sub- and supra-linear variability [33]–[42] . So while the orientation tuning width was relatively invariant to the level of synchrony , as shown in Figure 4C , the increased level of synchrony was accompanied by an increased mean firing rate , and thus an increased variance , the effects of which are not immediately obvious from the perspective of an ideal observer . Figure 5B shows the corresponding spike count distributions for the tuning curves in Figure 4B , for the preferred stimulus orientation ( 90 degrees ) . The spike count distribution changed dramatically as input synchrony decreased , with asynchronous inputs pinning spike count distributions at the origin and restricting the discriminability at adjacent distributions , a problem not encountered for highly synchronous inputs . From these results we might qualitatively expect that increasing synchrony would lead to increases in information because synchronization appears to give response distributions a greater range over which to vary with stimulus orientation . Results from both the mean-variance relationship and the per-synchrony peak spike count response distributions thus lead to conflicting expectations on what level of input population synchrony would drive the maximum amount of information about stimulus orientation . In order to solve this inconsistency we must implement a metric that describes concisely how discriminable different stimulus orientations are and determine the effect input synchrony has cortical information transfer . Fisher information quantifies the degree to which response distributions are discriminable , and thus , provide unambiguous information about stimulus features captured in the response distributions . The simplest understanding of Fisher information in the context of the problem here is that it represents the derivative of the tuning curve with respect to the stimulus orientation; regardless of the underlying firing statistics , the peak Fisher information will occur near orientations where the derivative of the tuning curve is highest . We use the peak amount of information across all stimulus orientations for each level of input synchrony as the metric for the capacity for any particular neuron to inform estimations about the stimulus orientation . By itself the absolute amount of information is an unintuitive quantity . With the goal of determining how synchrony changes the capabilities of cortical neurons to decode specific stimulus features , it is more natural to look at properties of the feature estimator . The inverse of Fisher information is the Cramér-Rao lower bound , a theoretical lower bound on the variance of a maximum-likelihood estimator; decreases in this quantity yield estimates that are more precise and have more confidence . Under the assumption that the stimulus orientation estimator is unbiased , lower estimator variance guarantees lower estimator error . Since we could directly calculate Fisher information in our model , we could also determine what this lower bound was , as shown in Figure 6A . The estimator standard deviation decreased nonlinearly with increasing synchrony , covering a range of relatively precise estimates to very imprecise estimates with a notable saturation at around 20 ms of jitter; synchrony higher than this does not yield rapid gains while decreases in synchrony rapidly decrease the estimator precision . As the Fisher information is directly related to the local slope of the tuning curve this qualitative observation was unaffected , in a relative sense , by the discretization of the tuning curve . The raw information decreased approximately linearly with increasing minimum jitter as shown in Figure 6B ( error bars are ±1 S . D . ) . However , as we will show the degree to which this is not linear has important implications for the efficiency of information transmission by the cortical neuron . From these results , we naively assumed that a strategy which absolutely increased synchrony would always be best as it would always result in increasing stimulus information . As has been noted in other models which bear some similarities to our own [43] , there is a metabolic cost to increasing firing rate which can affect the efficiency of some information representations relative to others . In this case , as shown in Figure 6C , when we normalize the absolute amount of information by the number of cortical spikes , it becomes clear that the peak in transmission efficiency occurred at around 15 ms of thalamic jitter , and a quadratic fit had a peak at 16 ms with a clear decrease in information efficiency away from this peak . In previous studies [10]–[12] we identified that pairwise LGN synchrony in response to natural scenes tends to be from 10 to 20 ms as measured by our scale . As noted , this result was consistent across all simultaneously recorded neurons when these neurons were used as sources for single-trial spike times . A few neurons maintained this quadratic relationship between information transmission efficiency and input synchronization at a peak efficiency closer to 25 ms of timing jitter , slightly lower than expected . These results indicate that populations in the LGN are uniquely arranged to be effectively synchronized by a preferred orientation . This synchronization allows information transmission to be more efficient without sacrificing precision in estimating orientation . The results presented so far have demonstrated that information efficiency saturates at levels of minimum timing jitter between 10 and 20 ms , without addressing the effect of tuning width . It is clear from existing literature that there is a wide range of tuning widths that are typically measured in neurons in visual cortex [9] , [29]–[32] and these changes are reflected in the width of and thus the width of the tuning curve . To investigate the effect of changes in just tuning width we modulated both the minimum timing jitter as well as the tuning width , with the results shown in Figure 7 . From 4 . 1 to 30 . 8 degrees ( HWHH; maroon to light blue dots in Figure 7 ) , which covers the rough range one could expect tuning width to vary , it is clear that the normalized information per spike ( IPS ) has approximately the same pattern regardless of tuning width . We show normalized information per spike because Fisher information is directly related to the slope of the curve , higher slopes monotonically lead to higher absolute levels of information and as such 4 . 2 degree and 30 . 8 degree tuning widths have an order of magnitude difference in their absolute amount of information . The relationship between tuning width and information efficiency is made clearer in the breakouts in Figure 7B for each individual tuning width; with the exception of extremely narrow tuning widths , as the tuning width increases the optimal level of minimum jitter increases but still stays in the range of 10–20 ms . Narrow tuning curves fail to saturate information per spike because very narrow tuning curves effectively contain information about a very small range of orientations and the amount of information is directly related to the diference between baseline and peak firing rates . As an example consider a tuning curve that goes from baseline firing rate to peak firing rate in the span of 2 or 3 degrees ( a very narrow tuning curve ) . In this case higher peak firing rates have a very pronounced affect on the overall amount of information . Since lower jitter always provides higher peak firing rates , narrower tuning curves are always most efficacious at extremely low amounts of jitter . We thus see that the results are valid for a range of primary visual cortex neurons so long as they have tuning widths that are within physiologically measured ranges .
In this work we investigated the role of stimulus-driven synchrony in thalamic populations in the emergence of feature selectivity in primary visual cortex . The complete understanding of this role requires observation of entire thalamic sub-populations which are convergent onto single cortical neurons . Since these populations are too large to record electrophysiologically using current experimental methodologies , we synthesized representative populations from experimental data by randomly choosing recorded trials of neurons from which we could record , when obeying anatomical rules of thalamocortical connectivity [24] ( also see below ) . These populations had an amount of stimulus-driven synchronization that was a direct function of the orientation of a drifting grating stimulus . These synthesized populations allowed us to systematically modulate the underlying spike timing synchrony to investigate the way in which different levels of synchronization affect information transmission . Through a biophysically inspired integrate and fire model that simulates cortical responses , we estimated the resultant cortical orientation selectivity and the corresponding information conveyed about visual stimulus orientation by the cortical response . Ultimately we found that the level of synchronization of the input population had a nonlinear effect on the resulting information contained in the cortical response; higher levels of synchrony led to higher levels of information , but at the expense of a nonlinear increase in firing rate . When taking into account the potential cost of increased firing rate , we found that the most efficient transmission of information was at a level of thalamic synchrony in the range of 10 to 20 ms . It is important to note that the synchronization of neurons has been widely studied in a number of different contexts . Notably , synchronization of neurons across cortical columns has been previously reported in the visual cortex , proposed as a means to form relationships across regions of the visual field [44] . Additionally , in the context of convergence and divergence of retinal afferents projecting to the LGN , precise correlations have been observed across geniculate neurons that were present in the absence of stimulus driven correlations , and were attributed to the projections of common retinal ganglion cell inputs [13] . In contrast , the current study ( and previous studies from our group [11] , [23] ) specifically examines the role of stimulus driven synchronization/correlation of neuronal firing in the visual thalamus . Our previous investigations have shown that many neurons in the LGN do not exhibit appreciable noise correlations [11] . The focus here is thus on the relationship between the visual input and the resultant synchronization of firing activity across geniculate ensembles , a requisite for robust activation of the downstream cortical neurons to which they project . In the most general case , however , as described in Gray et al . [44] , the propagation of neuronal signals would involve a combination or interaction between the synchronization due to ongoing spontaneous activity and the stimulus-driven synchronization due to coordinated activation of neurons sharing the same topology and feature selectivity . Such a “from-any-source” view of synchronization carries with it the possibility that neurons with receptive fields from disparate regions of the visual field could be synchronized by spatially correlated stimuli . For example two very spatially distant LGN neurons could be simultaneously activated by either two unrelated objects or one very long bar of light; synchronization due to these origins are not considered in this model . It is important to note that we explicitly consider only recordings from spatially localized populations , as widely-spaced LGN units do not converge at the same cortical target . The emergence of orientation selectivity in primary visual cortex is perhaps the most well-studied example of cortical computation to date . As a result , there have been a large number of modeling studies seeking to capture the mechanistic explanation for the primary observation of orientation selectivity , and also to capture a number of related , and more complex functional properties ( e . g . contrast invariant orientation tuning , cross-orientation suppression , etc . ) . Given that there is little if any dispute as to the role of direct feed-forward geniculate input to cortical layer 4 in establishing the basic orientation preference for cortical neurons , models of orientation selectivity have invariably been constructed around a backbone of thalamic input . Although the model from Ringach introduced structured synaptic weightings and connectivity probabilities of thalamic inputs to cortex as a key model element [5] , the majority of other models assume relatively simple feedforward excitation structure and differ primarily in the relative strengths of the feedforward or intracortical inhibition [3] , [4] , [26]–[28] , [45] , [46] . A specific limitation of most of these previous models is that they explicitly do not directly involve electrophysiological data as thalamic input . For example , one class of models use simulations of thalamic or retinal responses based on the stereotypical difference-of-Gaussians representation of center-surround receptive fields [3]–[5] , [45] , [46] , while others rely on assumed or derived cortical conductances or membrane potential but not on actual thalamic input [26]–[28] . The large majority of previously published models also assume that sinusoidal inputs ( i . e . drifting gratings ) elicit sinusoidal thalamic responses and that the cortical membrane potential itself is perfectly sinusoidally modulated ( as in [9] or [26] , [45] ) . Dating back to the early 1980s there was the observation that drifting sinusoidal gratings produced asymmetric LGN response PSTHs ( i . e . a sharp peak at the onset of the stimulus followed by a long tail of decaying response ) [47]–[49] and more recently we have directly analyzed the effects of this synchrony in the context of cortical orientation and direction selectivity [23] . We assert that the precise timing and stimulus-driven synchronization of thalamic inputs serves a prominent role in the thalamocortical circuit and in the emergence of cortical feature selectivity . It is important to note that most , if not all , existing models designed to capture the mechanism behind cortical orientation selectivity rely on spatial arrangements of projecting thalamic inputs that in some cases exceed those observed experimentally [24] . More specifically , the relevant measure for thalamic input is the aspect ratio of the scatter of thalamic receptive fields that form the input to a single cortical layer 4 neuron . Recently , Jin et al . experimentally observed thalamic clusters and showed that the thalamic input to cortical orientation columns has receptive fields that are highly overlapped [24] . Because the scatter of the thalamic receptive fields covers 2 . 5 receptive field centers in visual space , the average layer 4 cortical neuron should have a maximum aspect ratio of 2 . 5∶1 . The thalamocortical model from Somers et al . was built on an aspect ratio of 3∶1 [3] , whereas the model from McLaughlin et al . was built on an aspect ratio of 4∶1 [4] . Similarly large aspect ratios are apparent from the Kayser et . al . model and Finn et . al . models , with ratios approximately 6∶1 and 2 . 5∶1 respectively [26] , [46] . It is clearly the case that inhibitory mechanisms play a significant role in the shaping of the cortical feature selectivity [2] , and would only serve to further refine the selectivity established by the direct feedforward thalamic input shown here . Many of the above-mentioned models differ from our presentation here in that they include OFF-center sub-populations in the thalamic population , most commonly offset from the ON-center population as would be implied by the common Gabor-type simple cell receptive field . To keep the model relatively straightforward and simple , we have chosen to focus on just ON-center populations . The majority of existing models were optimized to explain extra-classical effects of cortical receptive fields with a particular focus on the contrast invariance of cortical tuning width and as such constructed mechanisms specific to this issue . Specifically , it has been widely observed that although peak cortical firing rates are strongly dependent upon stimulus contrast , cortical orientation tuning is largely invariant to stimulus contrast ( for review , see [2] ) . This observation called into question the purely feedforward model of cortical orientation selectivity [2] . Subsequent models augmented the feedforward thalamic input with inhibitory feedforward connections [26] or cortico-cortico inhibition [46] or some combination [2] , [3] . We have previously shown that thalamic synchrony is largely unaffected by stimulus contrast [11] , and the cortical tuning based on thalamic synchrony is also contrast invariant . The model we have proposed here thus potentially demonstrates a completely feed-forward explanation for contrast invariance . For a fixed minimum jitter amount , as the underlying LGN firing rates across the entire population are modulated by changes in the stimulus contrast , the peak induced firing in the cortical neuron rises and falls . Since the changes in LGN firing are correlated across the LGN population , the synchrony across such a population ( with particularly focus on the relationship between stimulus orientation and the synchrony ) remains unchanged as a function of stimulus contrast . As demonstrated in Figure 4B for the span of biophysical levels of preferred orientation population synchrony ( ∼5 to 20 ms ) , the tuning width of the cortical neuron does not change , indicating that changes in the degree of underlying synchrony do not change the tuning properties . Although the results are not presented here directly , the combination of past and present results suggest that changes in the LGN population response ( i . e . the population becomes less active in general ) lead to a decreased or increased peak cortical response but the tuning curve widths will be invariant to stimulus contrast . We used Fisher information as a measure of the efficacy of cortical neurons in representing stimulus features ( orientation ) in response to changes in the synchrony of an input population . Specifically , we used the peak Fisher information irrespective of the orientation at which the peak occurs . Contrary to previous investigations [50]–[52] in which the absolute value of the Fisher information was used as an important measure of the performance of neural populations , here we sought to capture the relative effects of varying degrees of thalamic synchrony on the information conveyed by a single recipient cortical neuron target . In this case , we assumed that the Cramér-Rao lower bound need not be met and that whatever bias causing deviations from the lower bound is consistent across all simulation conditions . We ensure this by using the same input data and model structure for all conditions so that we can compare relative levels of information across different synchrony conditions for a single neuron . Although this is a simplification of the true amount of information ( and indeed no single neuron will saturate this lower bound ) , in either case the absolute information was consistent with previous studies utilizing experimental cortical data . Yarrow et . al . [52] computed Fisher information for both real and simulated neural populations and found an information level which was approximately consistent with the findings presented here ( see their Figure 4 as well as [51] Figure 3 , with axes in [52] helping in the conversion from SSI bits to Fisher Information in units of ) . This assumption ultimately only affects the reporting of estimator standard deviation ( as in Figure 6A ) which was not the primary result of the work . It is also important to note that the application of Fisher Information to cortical tuning curves has deeper roots in estimating cortical population response information transmission . Past work [53]–[57] has in general used constructions where a collection of identical cortical neurons have preferred orientations that uniformly span the orientation spectrum ( 0 to 360 degrees ) . In this study we considered only a single neuron in the population . We claim , though , that results which demonstrate information in a single neuron at all stimulus orientations are fundamentally identical to results which demonstrate information in a population at a single orientation . As long as we assume every neuron in the cortical population is conditionally independent , for the questions we ask these two formulations are fundamentally interchangeable . As identified in [54] under the assumption that each cortical neuron in this population is independent , then at every stimulus orientation the overall Fisher information isFurther , in the case that every neuron in the population is also assumed to be identical in response properties , then we can modify the above to read ( for any choice of ) It is clear though that not all cortical tuning curves are identical and the absolute amount of information is strongly negatively correlated with tuning width . Using this fact as inspiration , we show in Figure 7 that the optimally efficient level of input timing jitter is widely insensitive to the tuning width of the cortical neuron . In this case , even if a cortical population is composed of non-identical independent neurons , each neuron , as well as the population as a whole , will be optimally efficient as long as the thalamic input is synchronous to the 10–20 ms level ( thus implying we need no longer assume neurons within the population have identical , but shifted , tuning curves ) . If we further consider the effects of correlated variability , as in [55] , then we can no longer assume the units are independent . Regardless of whether the correlated variability increases or decreases the absolute amount of information ( and neither is guaranteed ) , correlated variability would raise or lower the response rate of the individual neurons in a coordinated manner . Since again our metric is one of relative comparisons , the results presented here are expected to be invariant to correlated variability in the sense that the efficiency of any single neuron may decrease , the peak efficiency will still occur between 10–20 ms ( which would still be true for all neurons in the cortical population ) . Thus our findings directly translate to cortical populations regardless of the independence and homogeneity of tuning properties of the component neurons . In previous studies of timing precision of individual thalamic neurons [10] and across thalamic pairs [11] in response to natural scenes , we have reported characteristic timescales on the order of 10–20 ms . In these previous studies , measures were taken across long segments of natural scene movies , representing the aggregate of instantaneous firing events whose timing precision clearly varies on an event-by-event basis [12] , [58] . The instantaneous synchronization of firing activity across a sub-population of neurons in the context of natural scenes is undoubtedly a complex function of the local properties of the scene , including spatial frequency , temporal frequency , and orientation of the local spatial structure . It is thus the case that the 10–20 ms average timescale reflects a distribution of synchronous events , spanning from synchrony on just a few milliseconds to more asynchronous firing over a timescale of 10€s of milliseconds , unlikely to drive the cortical target . Here , we report that in the context of the modulation of thalamic synchrony through visual stimulus orientation with drifting sinusoidal gratings , the most efficient level of thalamic synchrony in conveying relevant information to cortex is in the 10–20 ms range . This means that , on average , amongst natural scenes and all their various features , the thalamic neural response is tuned to maximize the efficiency of information transfer to the cortex ( similar to [22] ) . As we have investigated only the effects of orientation changes on synchronization and feature selectivity , we expect that this result implies that information efficiency will be similarly optimized for other visual features such as spatial and temporal frequency . Furthermore , it is possible that synchronization optimizes information transmission in entirely different sensory systems , given previous findings in the somatosensory system [59] .
Surgical and experimental procedures were performed in accordance with United States Department of Agriculture guidelines and were approved by the Institutional Animal Care and Use Committee at the State University of New York , State College of Optometry . The experimental data collection has been previously described [23] . Briefly , single-cell activity was recorded extracellularly in the lateral geniculate nucleus ( LGN ) of anesthetized and paralyzed male cats , with a total of three animals . As described in [60] , cats were initially anesthetized with ketamine ( 10 mg kg−1 intramuscular ) and acepromazine ( 0 . 2 mg/kg ) , followed by propofol ( 3 mg kg1 before recording and 6 mg kg−1 h−1 during recording; supplemented as needed ) . A craniotomy and duratomy were performed to introduce recording electrodes into the LGN ( anterior , 5 . 5; lateral , 10 . 5 ) . Animals were paralyzed with vecuronium bromide ( 0 . 3 mg kg−1 h−1 intravenous ) to minimize eye movements , and were artificially ventilated . Using a seven-electrode matrix , layer A geniculate cells were recorded extracellularly . The multielectrode array was inserted into the brain to record from iso-retinotopic lines across the depth of the LGN , using an angle of 25–30 degrees antero-posterior , 2–5 degrees lateral-central . To a multielectrode array ( with inter-electrode separation of 254 µm ) we attached a glass guide tube with an inner diameter of 300 µm . As the elevation axis is better represented in LGN than the azimuth axis , some of the populations of LGN receptive fields showed greater lateral than vertical scatter in the visual field [61] . Layer A of LGN was physiologically identified by performing several electrode penetrations to map the retinotopic organization of the LGN and center the multielectrode array at the retinotopic location selected for this study ( 5–10 degrees eccentricity ) . While recording , the RASPUTIN software ( Plexon , Dallas , TX ) was used to capture voltage signals after being amplified and filtered . We isolated single units by independently moving each electrode and the resulting units were spike-sorted online and verified offline using a commercially available algorithm ( Plexon , Dallas , TX ) . Cells were eliminated from this study if they did not have at least 1 Hz mean firing rates in response to all stimulus conditions . Cells were classified as ON or OFF according to the polarity of the receptive field estimate . For each cell , visual stimulation consisted of multiple repetitions of a drifting sinusoidal grating at 0 . 5 cycles/degree , at either 100% or 64% contrast . The direction of the drifting grating was varied . The orientation of a particular drifting grating was one of eight possible values: 0 , 45 , 90 , 135 , 180 , 225 , 270 , 315 degrees . The convention was that a vertically oriented grating drifting rightward was referred to as 0 degrees , a horizontally oriented grating drifting downward was referred to as 90 , and so on . The temporal frequency for all datasets was 5 Hz or 4 Hz . The spatial resolution for the drifting gratings was 0 . 0281 degrees per pixel . All stimuli were presented at a 120 Hz monitor refresh rate . Biophysiological levels of LGN population synchrony were measured from multiple sets of simultaneous electrophysiological recordings ( between 5 and 7 neurons were recorded simultaneously ) . A cortical neuron is thought to receive approximately 30 LGN inputs [8] but these neurons are substantially more densely arrayed than we can reasonably hope to record with penetrating electrodes . Population response estimates were achieved by expanding the simultaneous recorded neurons into a population of 30 neurons by replicating the recorded responses and then shifting to a new visual location , restricted within the visual space bounded by the original receptive field locations . This restriction resulted in a population that has a receptive field center diameter distribution that is consistent with [24] ( see Figure 1D ) . To create the population random shifts were allowed in both the vertical and horizontal directions ( i . e . the major and minor axes of the population ) but the restrictions placed by the original population layout often required greater shifts along one or the other axis . For the example in Figure 1C , the shift restrictions resulted in a visual space coverage of approximately 5 degrees ( horizontal ) by 2 degrees ( vertical ) . Shifting responses required knowledge of the timing difference in excitation between the old and the new location , defined as the shift latency . The replicated input spike trains occurred in response to sinusoidal gratings and , due to the regularity in the stimulus , the shift latency was relatively easy to calculate . This shift latency was estimated simply by measuring the timing latency between the maximum excitation at the centroids of the receptive fields at both the original location and the shifted locationwhere is the center to center separation of the original and shifted locations , represent the spatial ( cycle/deg ) and temporal ( Hz ) frequencies ( fixed ) of the stimulus itself , and is the angle between the axis connecting the two receptive fields and a line from the shifted location perpendicular to the oriented stimulus bar . A graphical representation of this is in Figure 1D . Each newly created neuron is assigned a random trial from all recorded trials of the original neuron and the shift latency value is added to all spike times within that chosen trial . For the representation of this process in Figure 1E each neuron received a trial from the appropriate stimulus orientation . As the overarching cortical model , though , expands to a much larger set of orientations than originally recorded from , for consistency each newly created neuron was assigned a trial from the recordings performed with a stimulus at a 0 degree orientation . This allows us to preserve the baseline across-neuron timing changes , while capturing the stimulus-driven timing modulations with our parameter , discussed below . The model was constructed such that all input synapses to the cortical neuron have equal strength and no particular synaptic location ( i . e . along the dendrite or at the soma ) , and accordingly the source of the spikes from within the LGN population has no effect on the actual model output . Since this is the case , we can estimate the input population auto-correlation by collapsing all LGN spike times into a single vector . The auto-correlation is then calculated by subtracting each spike time from all other spike times and calculating the histogram of these pair-wise interspike intervals . Synchronous populations will have a much higher proportion of small intervals ( neglecting stimulus periodicity ) than asynchronous populations . The auto-correlations are also appropriately normalized to be between 0 and 1 . To smooth the auto-correlation and eliminate correlations caused by the periodicity of the input , a Gaussian was fit to the central 200 ms lags in the correlation . We use timing jitter as a metric of synchrony , which is determined by normalizing the Gaussian fit and locating the lag at which this curve is equal to . To relate this number to the PSTH timing jitter ( i . e . combined population timing jitter ) we must divide by two ( see Supplemental in [10] for a complete description ) . In brief , we define a value which is the “response timescale” . This value is equal to the latency at which the auto-correlation equals . By construction this has the relationship that , where is the timing jitter in the PSTH , our value of interest . This process was performed for all stimulus orientations ( in order to maintain phase and timing differences that arise from differences in neuron properties and not just spatial relationships ) to describe timing jitter as a function of stimulus orientation . This function was calculated multiple times for different randomly generated populations to estimate the variance that is created by choosing either different visual locations for the component neurons or choosing different recorded trials to represent the neurons in the population . The observed timing variability in spike times across the population is composed of two aspects; intrinsic neural variability and variability caused by the interaction between the stimulus and the population organization . Our model captured the intrinsic variability by using spike times that were recorded in vivo . On the other hand , while the grating stimulus always evokes firing in the thalamic neurons the timing differences in spike times from neuron to neuron will vary according to orientation of these gratings and the arrangement of the population itself . We capture this stimulus-evoked timing variability in a parameter . This parameter , as a function of stimulus orientation , was manually calibrated such that when used with recorded data we could reconstruct the exact plot shown in Figure 2C . This procedure allows us to capture both the intrinsic and stimulus-evoked sources of spike timing variability even at orientations for which we were not able to collect data . All simulations and computations were performed in the Matlab programming language ( Mathworks , Inc . , Natick , MA ) using a 64-node grid computer . The integrate and fire model [62] , illustrated in Figure 3A , takes spiking activity from the simulated LGN population as input and outputs cortical membrane potential and the associated cortical spike times . It was assumed that each synapse has equal strength . To create the synaptic input current , an exponentially decaying EPSC of defined amplitude ( ) and time constant ( ) was generated for all spike times in the input LGN population . The EPSCs were summed linearly across all LGN inputs to create a single current input at every simulation time point . The cortical membrane potential was modeled with the following first-order differential equation:where is the membrane potential , is the total synaptic current , is the membrane resistance ( ) , is the resting potential ( −70 mV ) , and is the membrane time constant ( 2 ms ) . The integration was performed using the forward euler method with a step size of 0 . 05 ms; since the step size is significantly smaller than any other temporal dynamics or spike timing precision use of a simple euler method is sufficient . When exceeds the threshold membrane potential ( ) , a cortical spike is generated by setting the instantaneous potential to 0 mV followed by a 3 ms refractory period at the reset potential of −65 mV . These values are similar to those we have used previously for similar models [22] , [23] . An analysis was performed to determine the approximate sensitivity of the model to each of the above indicated parameters . In general the model is sensitive to parameters which modulate the strength ( or efficacy ) of input spikes relative to the generated EPSC . Thus the model is sensitive to the EPSC amplitude ( ; effective values 0 . 05 to 0 . 1 nA within acceptable ranges ) and the EPSC decay ( ; effective values 2 to 5 ms ) while being robust to changes in threshold and reset potentials ( ) . Sensitivity manifests itself as a change between one of three states; impoverished cortical firing , sufficient cortical firing , and strong cortical firing . Impoverished firing results in a peak information per spike ( see Figure 6C ) at very low jitter values ( as this maximizes the chance to get any spikes ) and strong firing demonstrates no discernible peak information per spike for any particular jitter value ( as it results in very wide tuning curves ) . Taking the perspective of an ideal observer , we approximated the capability of the observer to discriminate between visual stimulus orientations based on cortical activity alone . More specifically , the Fisher information [54]–[57] at each orientation captures the discriminability between where the expectation is taken with respect to . In the case that the probability is zero , we set . We calculated the derivative numerically using increments of 1 degree which was the resolution at which the simulations were performed . To reduce the results of this calculation to a single descriptive value , we report the estimator minimum standard deviation , which is related to the Fisher information through the Cramér-Rao lower bound ( assuming the estimator is unbiased ) :As a metric of efficiency with which the cortical output conveys information about the stimulus , we divide the peak output information by the peak spike count with the goal of identifying how much each individual spike contributes to the overall information; higher values indicate each spike is more efficient at conveying information about stimulus features . This established a penalty for higher firing rates , realizing that there is a metabolic cost to generating action potentials . Response distributions of the cortical firing rate were estimated based on the simulated data , in order to calculate the Fisher Information . The firing rate varied as a function of and the distributions are given by . The data were explicitly fit to a Poisson distribution , consistent with previous findings [33]–[42] as well as explicitly verified for appropriate fitting against our own data:To generate an accurate estimation of the response distributions at a minimum 250 simulation trials were run , with more trials providing no significant change in the estimated distributions . Note that the distributions change as a function of stimulus orientation , as indicated by . Further , in order to create a smooth description of Fisher information it was necessary that the response distributions be smooth functions of , as even minor fluctuations in the parameter get magnified by differentiation and squaring . To alleviate this , was smoothed with a Gaussian fit which was empirically verified to describe well . | While the visual system is selective for a wide range of different inputs , orientation selectivity has been considered the preeminent property of the mammalian visual cortex . Existing models of this selectivity rely on varying relative importance of feedforward thalamic input and intracortical influence . Recently , we have shown that pairwise timing relationships between single thalamic neurons can be predictive of a high degree of orientation selectivity . Here we have constructed a computational model that predicts cortical orientation tuning from thalamic populations . We show that this arrangement , relying on precise timing differences between thalamic responses , accurately predicts tuning properties as well as demonstrates that certain timing relationships are optimal for transmitting information about the stimulus to cortex . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"circuit",
"models",
"visual",
"system",
"computational",
"neuroscience",
"sensory",
"systems",
"biology",
"neuroscience",
"coding",
"mechanisms"
] | 2014 | The Role of Thalamic Population Synchrony in the Emergence of Cortical Feature Selectivity |
Bacterial superantigens ( SAgs ) cause Vβ-dependent T-cell proliferation leading to immune dysregulation associated with the pathogenesis of life-threatening infections such as toxic shock syndrome , and necrotizing pneumonia . Previously , we demonstrated that staphylococcal enterotoxin-like toxin X ( SElX ) from Staphylococcus aureus is a classical superantigen that exhibits T-cell activation in a Vβ-specific manner , and contributes to the pathogenesis of necrotizing pneumonia . Here , we discovered that SElX can also bind to neutrophils from human and other mammalian species and disrupt IgG-mediated phagocytosis . Site-directed mutagenesis of the conserved sialic acid-binding motif of SElX abolished neutrophil binding and phagocytic killing , and revealed multiple glycosylated neutrophil receptors for SElX binding . Furthermore , the neutrophil binding-deficient mutant of SElX retained its capacity for T-cell activation demonstrating that SElX exhibits mechanistically independent activities on distinct cell populations associated with acquired and innate immunity , respectively . Finally , we demonstrated that the neutrophil-binding activity rather than superantigenicity is responsible for the SElX-dependent virulence observed in a necrotizing pneumonia rabbit model of infection . Taken together , we report the first example of a SAg , that can manipulate both the innate and adaptive arms of the human immune system during S . aureus pathogenesis .
Staphylococcus aureus is an opportunistic pathogen responsible for a wide array of human diseases in both the hospital and community settings [1] . The diversity of disease types and the strain-dependent variation in pathogenic potential is due in part to the large array of virulence factors that are produced by S . aureus [1] . The staphylococcal superantigens ( SAgs ) are a family of at least 26 secreted proteins that modulate the immune system by stimulating dysregulated T-cell proliferation [2–4] , contributing to a variety of different diseases including toxic shock syndrome , necrotizing pneumonia and Kawasaki disease [2] . The diversity of SAgs produced by S . aureus strains facilitates interaction with the large repertoire of variable-β chains ( Vβ ) found in the T-cell receptor leading to dysregulation of a critical component of the adaptive immune response [2 , 5] . The SAg SElX is encoded in the core genome of over 95% of S . aureus isolates and contributes to lethality in a rabbit model of necrotising pneumonia [6] . Although a member of the SAg family , SElX exhibits greater sequence homology with the staphylococcal superantigen-like protein ( SSl ) family comprising of proteins that are structurally similar to SAgs but lack the capacity to induce Vβ-specific T-cell proliferation [5] . The SSls are associated with a diversity of immune evasion functions including the blockade of complement activation , interference with bacterial cell wall opsonisation ( e . g . SSl7 and SSl10 ) and disruption of neutrophil function ( e . g . SSl3 , SSl4 and SSl5 ) [7–11] . Of note , SSl5 can bind to neutrophils via a direct interaction with CD162 ( P-selectin glycoprotein ligand-1; PSGL-1 ) , reducing neutrophil migration [7 , 12] , and SSl3 and SSl4 are toll-like receptor 2 antagonists which prevent neutrophil activation by bacterial lipopeptides [11 , 13 , 14] . Fevre et al . ( 2014 ) previously demonstrated that SElX can interact with neutrophils and monocytes , binding via the CD162 molecule on the surface of neutrophils [15] . However , very high concentrations of SElX were required for a relatively low-affinity interaction suggesting that CD162 may not be the main neutrophil receptor involved [15] . In addition to SSls , S . aureus produces other molecules which subvert the innate immune response including chemotaxis inhibitory protein of staphylococcus ( CHIPS ) , which binds to the formylated peptide and C5a receptors on neutrophils , blocking chemotaxis to the site of infection [16] . Furthermore , the formyl peptide receptor-like 1 inhibitor ( FLIPr ) and its homologue FLIPr-like can antagonise the formylated peptide receptor and bind to Fcγ receptors disrupting IgG-mediated phagocytosis of neutrophils [17 , 18] . Other examples of multifunctional determinants include; extracellular adherence protein ( EAP ) which can act as a host cell invasin and inhibit the activity of neutrophil elastase , Panton–Valentine leucocidin ( PVL ) which has been demonstrated to induce inflammation independent of cell lysis , and collagen binding protein ( CNA ) which , in addition to its role binding to extracellular collagen , can bind C1q and block the complement cascade [19–21] . These examples of multi-functionality highlight the apparent functional redundancy exhibited by S . aureus with regard to pathogenesis , providing a robust , multi-faceted response to innate immunity during the early stages of infection . In the current study , we further investigated the role of SElX in S . aureus disease pathogenesis . We discovered that SElX binds to neutrophils via multiple glycosylated neutrophil surface receptors , inhibiting phagocytosis and contributing to the pathogenesis of severe lung infection . Importantly , the neutrophil binding and superantigenic functions of SElX are mechanistically independent providing a bi-functional disruption of both the innate and adaptive arms of the human immune system .
In order to test the hypothesis that SElX can bind to human leukocytes , recombinant SElX was incubated with human leukocytes isolated from healthy donors . SElX bound to both human neutrophils and monocytes at a high level , comparable to or greater than the known neutrophil-binding protein SSl5 ( Fig 1ii and 1iii ) . In addition , SElX exhibited low level binding to CD4+ and CD8+ T-lymphocyte subsets but not to B-lymphocytes ( CD19 ) ( S1 Fig ) . In order to examine the host-dependent binding of SElX , neutrophils from cattle , rabbits and mice were employed . SElX exhibited a higher level of binding to murine neutrophils compared to human neutrophils , but this was reduced compared to bovine and laprine neutrophils . However , binding to neutrophils from all species was observed at a concentration below 100 nM ( Fig 1v , 1vi and 1vii ) . Overall SElX exhibited binding with neutrophils from multiple mammalian species at a level that exceeded that of the known neutrophil-binding protein SSl5 . Amino acid sequence alignment of SElX with previously characterised SSl-proteins revealed a conserved sialic acid-binding motif ( YTxExxKxLQx[H/N/D]Rxx[D/E] ) matching 54% amino acid sequence identity with that of SSl5 ( Fig 2 ) [7 , 11 , 22 , 23] . Of note , 4 amino acids of the motif have been demonstrated to interact with sialic-Lewis X ( sLEX ) [23] , and 3 of these residues ( E154 , K156 , and Q169 ) are conserved in SElX ( Fig 2 ) and the 5 SSls that contain the motif ( Fig 2 ) . The fourth residue ( [H/N/D]161 ) identified to interact with sLEX in crystallography studies [23] , varies among the SSl-proteins and allelic variants of SElX encoded by different strains suggesting that this residue is not essential for sialic acid binding ( Fig 2 ) . SElX exhibits considerable allelic variation , with at least 17 allelic variants , and a total of 40 ( 13% ) variable amino acid positions identified [6] . However the sialic acid binding motif contains only one variable residue among all known allelic variants , consistent with functionality ( Fig 2 ) . To examine the hypothesis that SElX binding is sialic acid-dependent , neutrophils and monocytes were pre-treated with neuraminidase prior to binding . Neuraminidase treatment abolished the binding of SElX to both neutrophils and macrophages , suggesting that the interaction is sialic acid-dependent ( Fig 3i ) . This is consistent with the observation of Fevre et al ( 2014 ) who also noted a sialic acid requirement for SElX binding [15] . In order to investigate the role of the predicted sialic acid-binding motif of SElX in neutrophil binding , single site-directed mutants of each of the four predicted binding residues were constructed in addition to a combination mutant of all four residues ( Fig 3ii ) . Mutant proteins were expressed in E . coli , purified and structurally validated by circular dichroism and thermal shift analysis to ensure the mutations did not destabilise or influence protein structure ( S2 Fig ) . Binding analysis indicated that each of the SElX single site-directed mutants of E154A , K156A , Q169A and the combination mutant SElX-EKQD-A exhibited almost complete loss of binding to both neutrophils and monocytes compared to the wild-type SElX protein , indicating that each of these 3 residues are required for SElX-mediated neutrophil and monocyte binding ( Fig 3iii ) . The mutant SElX-D161A did not demonstrate a reduction in neutrophil or monocyte binding compared to wild-type SElX , indicating that this residue is not required for neutrophil or monocyte binding ( Fig 3iii ) . Taken together these data demonstrate that the predicted sialic acid-binding motif of SElX is essential for binding to both human neutrophils and monocytes . In order to identify the receptors for SElX-binding on the surface of neutrophils , affinity precipitation analysis of recombinant SElX and the neutrophil binding-deficient SElX-EKQD-A mutant was carried out with human neutrophil lysates , followed by quantitative mass spectrometry ( MS ) analysis . At least 12 proteins were enriched 5-fold or higher on SElX-coated Ni-NTA beads compared to Ni-NTA beads coated with SElX EKQD-A mutant protein suggesting that SElX binds to many neutrophil receptors in a sialic-acid dependent manner ( S1 Table; Fig 4i ) . To test the hypothesis that individual neutrophil receptors could support binding of SElX we performed an ELISA-type assay with recombinant SElX and recombinant CD50 ( ICAM-3 ) , a protein that was enriched in the affinity precipitation analysis ( Fig 4ii ) . The data demonstrated that SElX could bind directly to immobilized CD50 in a manner dependent on the sialic-acid binding motif ( Fig 4ii ) . All of the cell surface proteins identified to bind SElX are glycoproteins , consistent with the requirement for sialic-acid binding and represent an array of functional pathways in the neutrophil . Of note , the most enriched proteins are glycoproteins that display integrin or cell activation functions ( CD45 , CD13 , CD31 , CD50 and CD148 ) [24–27] . In addition , several cytosolic or granule-associate proteins were identified to bind to SElX , including p22-PHOX and the enzyme maltase-glucoamylase ( MGAM ) ( found in the gelatinase and ficolin granules of neutrophils ) , both of which are involved in microbicidal functions of neutrophils [28 , 29] . To determine if SElX inhibits neutrophil-mediated phagocytosis of S . aureus , neutrophils were pre-incubated with SElX before the addition of opsonised bacteria . SElX reduced the ability of neutrophils to take up bacteria when heat-inactivated human pooled serum ( ΔHPS ) or IgG alone were applied as opsonins ( Fig 5i ) . These data demonstrate that IgG-mediated phagocytosis is inhibited by SElX independently of the complement-mediated phagocytic pathway . SElX exhibits a highly potent activity at a concentration as low as 20 nM with a reduction in phagocytosis of up to 30% , compared to up to 70% for the Fcγ antagonist FLIPr ( Fig 5ii ) . Of note , no reduction in phagocytosis was observed when the SElX EKQD-A mutant was pre-incubated with neutrophils ( Fig 5ii ) , consistent with a requirement for sialic acid-binding dependent neutrophil interactions for phagocytosis inhibition . To test the hypothesis that SElX enhances S . aureus survival in the presence of neutrophils , S . aureus strain USA300 LAC , a derivative selx deletion mutant , or a derivative that produces SElX-EKQD-A ( Table 1 , see S3 Fig for mutant validation ) were each incubated with human neutrophils . The USA300 LAC parent strain was highly resistant to neutrophil killing consistent with previous reports [30 , 31] , but the selx deletion mutant and EKQD-A-expressing derivatives both exhibited increased susceptibility to killing ( Fig 5iii ) . Importantly , re-introduction of selx to the LACΔselx mutant restored killing resistance to wild-type levels ( Fig 5iii ) . These data indicate that SElX enhances the capacity of S . aureus to survive neutrophil-mediated killing . To determine if the superantigenic activity of SElX is distinct from its capacity to bind to neutrophils and inhibit phagocytosis , sialic acid-binding mutants of SElX were examined for their ability to induce T-cell proliferation by [3H] thymidine incorporation ( Fig 6 ) . All 5 mutants were capable of inducing T-cell proliferation at comparable levels inferring that the residues essential for neutrophil binding are not required for superantigenic activity . These data indicate that SElX exhibits two independent mechanisms of immune modulation affecting distinct cell types . To examine the role of SElX-neutrophil binding in the pathogenesis of skin infections , we carried out experimental murine skin abscess infections with USA300 LAC and derivative mutant strains ( Table 1 ) . Lesions developed within 24 h post injection and generally reached peak size within three days . No differences were seen in lesion size between the different mutant groups and the number of bacteria recovered at each time point declined over the course of the study up to 144 h ( Fig 7 ) . At each time point there were no differences seen in bacterial load between the different mutant and wild-type groups ( Fig 7 ) . For histopathological analysis , 6 mice were analysed from each experimental cohort and infected animals were sacrificed at 72 and 144 h post infection . Processed tissue slides were assessed for abscess morphology , severity of tissue damage and leukocyte infiltration . Diversity was observed in terms of both lesion morphology and severity although these features were not associated with a particular bacterial genotype ( Fig 7 ) . High levels of leukocyte infiltrations were observed in all slides but no differences detected among the different mutants groups in comparison to LAC wild-type infected tissue ( Fig 7 ) . Taken together , these data indicate that selx does not contribute to virulence in this murine skin abscess infection model . Previously , SElX has been demonstrated to contribute to lethality in a rabbit model of necrotising pneumonia [6] . Here , the experiment was repeated including the selx-EKQD-A-expressing USA300 LAC to examine the relative contribution of superantigenicity and neutrophil-binding functions of SElX to the outcome of infection . At an inoculum of 6 x 109 CFU , rabbits infected with LACΔselx or LAC selx EKQD-A had longer survival times ( Fig 8i ) and had less elevated body temperatures than animals infected with wild-type LAC or repaired strain ( LACΔselx rep ) ( Fig 8iii ) . Similarly , at a lower inoculum of 2 x 109 CFU animals challenged with the LAC selx EKQD-A mutant strain had increased survival ( p = 0 . 0027; Fig 8ii ) and less elevated body temperature compared to animals infected with the wild-type ( p = 0 . 0014; Fig 7iv ) . On gross examination , lungs from rabbits infected with wild-type LAC were dark red to purple indicating severe haemorrhage ( Fig 8v left ) . In contrast , lungs from rabbits infected with the LAC selx EKQD-A mutant strain demonstrated signs of haemorrhage in one lobe but tissue was relatively healthy in the second lobe ( Fig 8v right ) . Taken together , these data indicate that residues essential for sialic acid-binding of SElX are required for the pathology caused by SElX in a rabbit model of necrotising pneumonia . As a role for SElX in pathogenesis in a murine skin abscess model was not observed in the current study , these data highlight important host- or tissue-specific differences relevant to the choice of infection model employed for investigating S . aureus pathogenesis .
As previously reported in Wilson et al ( 2011 ) , SElX shares sequence homology with SAgs and SSls , suggesting that SElX may have properties of both families of proteins . In this study we have demonstrated that SElX exhibits characteristics of both groups and can therefore be described as a functional hybrid of the SAgs and SSls . SElX is a bi-functional protein that binds to neutrophils and inhibits IgG-mediated phagocytosis via a mechanism that is distinct from its superantigenic activity . This is the first time that bi-functionality has been observed in superantigens independent of the emetic activity of staphylococcal enterotoxins . The common multifunctionality of S . aureus virulence factors is an emerging theme in S . aureus pathogenesis research . For example , SSl5 binds to neutrophils via PSGL-1 and also inhibits the function of metalloprotease 9 resulting in the inhibition of cell activation by chemokines , rolling and migration of neutrophils to the site of infection , and reduction in the formation of thrombi [34–38] . In addition , Staphylococcal protein A ( SpA ) subverts opsonisation by binding the Fc region of IgG molecules and can also act as a SAg for B-lymphocytes leading to disruption of the humoral immune response [39] . Cytolytic toxins from S . aureus ( including Hla , HlgACB , PVL and LukAB ) in addition to their cytolytic activity on haematopoietic cells have been shown at sub-lytic concentrations to activate the intracellular NOD-like Receptor protein 3 ( NLRP3 ) inflammasome in neutrophils , monocytes and macrophages which lead to pro-inflammatory cytokine release [21 , 40–44] . Of note , we have identified a protein that employs distinct mechanisms to target two distinct immune cell types linked to innate and acquired immune responses , respectively . The SaeRS two-component system controls an array of S . aureus virulence factors such as CHIPS , SCIN , the SSls , and SAgs including TSST-1 , which are involved in immune evasion in response to host stimuli [45–47] . Transcriptomic analysis of a USA300 deletion mutant of SaeRS , performed by Nygaard et al , resulted in almost a five-fold decrease in the transcription levels of selx ( SAUSA300_0370 ) [47] . These data suggest that selx is regulated by SaeRS in co-ordination with an array of other immune modulators . Considering its dual modes of function , SElX is likely produced by S . aureus early in infection in response to neutrophil signals but may persist at the site of infection as the adaptive immune response is recruited , leading to induction of T-cell proliferation and further immune dysregulation . Alternatively , SElX activation of lymphocytes resulting in the release of cytokines and chemokines ( such as interferon-γ and interleukin 6 ) may stimulate the recruitment of neutrophils , as reported previously for SEA [48] . As neutrophils are recruited to form an abscess , SElX may enhance bacterial survival by inhibiting phagocytosis and stimulating misdirecting cytokines from lymphocytes which could also inhibit neutrophil function . We observed that SElX exhibited low levels of binding to human lymphocytes and this was limited to CD4+ and CD8+ T-cells with no binding to B-lymphocytes ( S1 Fig ) . This is a characteristic of the SSl-proteins that bind to human leukocytes and may reflect the low activity of glycosyl-transferases in peripheral lymphocytes prior to activation [49] . In lymphocytes , glycosyl-transferases are activated by the presence of IL-2 , an interleukin induced during SAg-mediated activation of the lymphocyte [49] . It is possible that SElX induces glycosylation on the surface of lymphocytes through SAg-mediated activation and then acts as a glycoprotein antagonist modulating lymphocyte function . However , further experiments would be required to investigate this hypothesis . SElX demonstrates the ability to interact with neutrophils from multiple mammalian species including humans , rabbits , mice , and cattle . Dairy cows and farmed rabbits represent important veterinary hosts of S . aureus [50] and the ability of SElX to bind to neutrophils from these species , in addition to humans , suggests a broad role for SElX in pathogenesis of S . aureus mammalian hosts ( Fig 1 ) . Previous studies of S . aureus virulence have demonstrated that rabbits may represent more appropriate infection models than mice for the analysis of some virulence factors such as Panton-Valentine leucocidin ( PVL ) [51] . Although we identified that SElX bound strongly to murine neutrophils , we did not detect any SElX-dependent effect on virulence in a murine skin abscess model ( Fig 7 ) . This may reflect low levels of SElX expression during murine skin infection or alternatively it may be due to the previously observed differential glycan decoration on murine cell receptors compared to humans [52–55] , which could impact on the capacity of SElX to bind to specific target receptors with downstream pathogenic consequences . Our data suggest that the effect of SElX on neutrophils is independent of cytolysis and apoptosis ( S4 Fig ) and that it can mediate an inhibition of IgG-mediated phagocytosis , consistent with its capacity to bind to the Fcγ receptor CD16b . However , the observed promiscuous binding of SElX to an array of glycosylated neutrophil proteins suggests that SElX may disrupt other neutrophil functions . For example , another SElX binding partner ICAM-3 ( CD50 ) is an important signalling molecule associated with the lymphocyte function-associated antigen 1 ( LFA-1 ) which was also identified as a ligand for SElX ( α-L integrin ) [56 , 57] . In addition , several proteins linked to the phagosome including MGAM ( a glucamase present in neutrophil granules ) [29] and p22-PHOX ( a key component of the cytochrome B complex ) [28] were able to bind to recombinant SElX in a sialic acid dependent manner ( Fig 4 ) , suggesting that SElX may interfere with neutrophilic enzymes involved in phagosomal killing . The multiple functions exhibited by many S . aureus virulence factors contribute to the remarkable apparent redundancy of function promoting the evasion of the immune response . This is an important consideration for the development of vaccines and anti-virulence therapy as therapeutic or preventive measures targeting a small number of virulence factors may prove ineffective if similar functions can be mediated by alternative proteins . Therefore , developing an understanding of the full functional repertoire of S . aureus virulence factors will inform the rational design of targeted therapies that by-pass the intrinsic redundancy in S . aureus pathogenesis . In conclusion , we have demonstrated that SElX is a bi-functional SAg with distinct modulatory effects on critical functions of the innate and acquired immune response . Given the large family of SAgs identified to date , our findings imply that further investigations into the alternative functions of SAgs are warranted .
Human venous blood was taken from healthy donors in accordance with a human subject protocol approved by the national research ethics service ( NRES ) committee London City and East under the research ethics committee reference 13/LO/1537 . Volunteers were recruited by a passive advertising campaign within the Roslin Institute ( University of Edinburgh ) and following an outline of the risks , written informed consent was given by each volunteer before each sample was taken . De-identified human blood packs from consenting-healthy individuals were also purchased from the New York Blood Center . The New York City Blood Center obtained written informed consent from all participants involved in the study and risks were outlined to each donor prior to sampling . The use of the de-identified samples was reviewed and approved by the New York University School of Medicine University Committee on Activities Involving Human Subjects . All experiments using animals were handled in strict accordance with good animal practice as defined by the relevant national and local animal welfare bodies . Specifically , experimentation using rabbits were performed under a University of Iowa approved Institutional Animal Care and Use Committee ( IACUC ) protocol ( 4071100 ) . Animals were maintained in accordance with the Animal Welfare Act ( 1966 ) ( administered by the US Department of Agriculture ) and the U . S . Department for Health and Human Services ( DHHS ) ‘‘Guide for the Care and Use of Laboratory Animals” . University of Iowa is accredited by the Association for Assessment and Accreditation of Laboratory Animal Care International ( AALAC ) . For intra-pulmonary administration of bacterial strains , animals were anesthetized with use of ketamine ( 10 mg/kg ) combined with xylazine ( 10 mg/kg ) . In agreement with the University of Iowa IACUC , animals that failed to exhibit escape behaviour and at the same time could not right themselves were prematurely euthanized according to predetermined experimental endpoints . Animals were euthanized with intravenous injection of 2 ml of Euthasol , whether prematurely or at the end of experimentation . Euthanized animals were then subjected to bilateral thoracotomy to ensure euthanasia . All animals received pain-relieving medication ( buprenorphine; 0 . 05 mg/kg intramuscular twice daily ) for the duration of experimentation . All murine experiments were carried out under the authority of a UK Home Office Project License ( PPL 70/08663 ) within the terms and conditions of the regulations of the UK Home Office Animals ( Scientific Procedures ) Act 1986 and the code of practice for the housing and care of animals bred , supplied , or used for scientific purposes . The study protocol was reviewed and approved by the Roslin Institute Small Animal Review Group ( University of Edinburgh ) which includes the Named Veterinary Surgeon ( NVS ) , Named Animal Care and Welfare Officers ( NACWOs ) and a statistician services prior to each experiment . Animals were monitored twice daily to ensure that no animal exceeded predetermined severity agreed euthanasia criteria according to the study protocol . Animals were euthanized , whether prematurely or at the end of experimentation by asphyxiation using carbon dioxide . Euthanized animals were then subjected to cervical dislocation to ensure euthanasia . Chicken immunisation was provided by the Scottish national blood transfusion service ( Pentland Science Park , Midlothian , UK ) using unembryonated hen’s eggs . Procedures performed were carried out under the authority of a UK Home Office Project License ( PPL 60/4165 ) within the terms and conditions of the regulations of the UK Home Office Animals ( Scientific Procedures ) Act 1986 and the code of practice for the housing and care of animals bred , supplied , or used for scientific purposes . S . aureus strains were cultured in tryptone soya broth ( TSB ) or brain heat infusion broth ( BHI ) ( Oxoid , UK ) with shaking at 200rpm , or on tryptone soya agar ( TSA ) ( Oxiod , UK ) at 37°C for 16 h unless otherwise stated . E . coli strains were cultured in Luria broth ( LB ) ( Melford laboratories , UK ) with shaking at 200rpm , or on LB-agar ( Melford laboratories , UK ) at 37°C for 16 h unless otherwise stated . Media was supplemented , where appropriate , with antibiotics . Strains were stored in the appropriate liquid culture media containing 40% ( v/v ) glycerol ( Sigma-Aldrich , UK ) in cryovials ( Nunc , Thermo Scientific , UK ) at -80°C . 50 ml of venous blood was drawn from healthy human volunteers and mixed with 6 ml of acid-citrate-dextran ( ACD ) ( 25 g D-glucose ( Sigma-Aldrich , UK ) and 20 . 5 g trisodium citrate ( Sigma-Aldrich , UK ) added to 1 L of ddH2O ) . Human neutrophils and peripheral blood mononuclear cells ( PBMC ) were isolated as outlined previously [58] and re-suspended in RPMI1640 ( Gibco , UK ) containing 0 . 05% human serum albumin ( HSA ) ( Sigma Aldrich , UK ) for further analysis . Murine bone marrow derived neutrophils were isolated from femurs of wild-type mice from BALB/C backgrounds , as described previously [59] . Neutrophil cell identity was confirmed using Lys6C and Lys6G staining by incubating the isolated cells with 1:500 dilutions of Rat anti-Lys6C PerCP-Cy5 . 5 ( AL-21; BD Bioscience , Oxford , UK ) and Rat anti-Lys6G PE-Cy7 ( 1A8; BD Bioscience , Oxford , UK ) for 30 min on ice and then analysed by flow cytometry ( FACSCalibur; Becton Dickinson , Franklin Lanes , NJ ) . Expression of neutrophil cell surface markers was confirmed by a high fluorescent signal for Lys6G and Lys6C . Bovine neutrophils were isolated from Holstein-Friesian cattle aged 18 to 36 m via jugular vein puncture ( 6 ml 10 x PBS/EDTA ( 100 mM KH2PO4 , 9% ( w/v ) NaCl , and 2 mg/l EDTA in sterile H2O adjusted to pH 7 . 4 ) anticoagulant used for 50 ml of blood ) . Neutrophils were isolated from bovine blood using a previously described protocol [60] . Rabbit neutrophils were isolated from peripheral rabbit blood ( using ACD as an anticoagulant ) purchased from the National Transfusion Centre , Moredun Institute , Midlothian , UK . Leukocytes were isolated using an equal volume of red blood cell ( RBC ) lysis buffer ( 10 x buffer contained; 155 mM NH4Cl , 10 mM KHCO3 and 100 μM EDTA in ddH2O ) . RBCs were lysed in 1x lysis buffer for 10 min at 37°C and the remaining cells were pelleted at 400 x g for 10 min and washed with HBSS ( Mg2+ and Ca2+ free ) ( Gibco ) and this was repeated until the pellet was clear of RBCs . All cells were counted with a TC20 automated cell counter ( Biorad , Hemel Hempstead , UK ) and re-suspended to a concentration of 1 x 107 cells/ml in the desired assay media/buffer . Liquid cultures of S . aureus strains were grown in BHI broth to stationary growth phase ( OD600 = 4 . 0–7 . 0 ) . Cultures supernantants were concentrated using Amicon Ultra-15 Centrifugal Filter units ( 10000 MWCO ) ( Millipore , Watford , UK ) . Concentrated secreted proteins were separated by sodium dodecyl sulfate polyacrylamide gel electrophoresis ( SDS-PAGE , 12% acrylamide gel ) and transferred to nitrocellulose membranes using the Trans-Blot Turbo Blotting system ( Bio-Rad , Hemel Hempstead , UK ) , according to the manufacturer’s specifications . The membrane was incubated in 1 x PBS ( pH 7 . 3 ) containing 8% ( w/v ) powdered milk ( Sigma-Aldrich , UK ) , at 4°C overnight . The membrane was incubated with primary antibody ( anti-SElX chicken IgY ) , at a dilution of 4μg/ml for 2 h . Anti-SElX IgY was obtained from immunised unembryonated hen’s eggs using the Eggspress IgY purification kit ( Gallus Immunotech , NC , USA ) according to the manufacturer’s instructions . Following primary antibody incubation the membrane was washed 3 times with washing buffer , 1 x PBS ( pH 7 . 3 ) containing 1% ( w/v ) powdered milk and 0 . 05% ( v/v ) Tween 20 ( Sigma-Aldrich , UK ) . The membrane was incubated with a horse radish peroxidase-conjugated ( HRP ) secondary antibody ( goat anti-chicken-IgY IgG ( Source Bioscience , Nottingham , UK ) ) , at 0 . 5 μg/ml in washing buffer , for 1 h , followed by a further 3 washes and then incubated with ECL reagent for 5 min . The blot was then exposed on Hybond ECL film ( Amersham , Systems , Buckinghamshire , UK ) for 20 s and developed using an X-ray developer ( SRX-101A , Konic Minolta , Japan ) . E . coli expression strain BL21 DE3 containing plasmid constructs ( pET15b::selx2 , pET15b::ssl7 [6] , and pRSETB::ssl5 [12] ) for protein expression were cultured in LB containing selective antibiotics and induced in mid-exponential phase of growth ( OD600 = 0 . 6 ) , with 1 mM isopropyl b-D-1-thiogalactopyranoside ( IPTG ) ( Formedium Ltd . , Norfolk , UK ) for 4 h at 37°C . Cells were recovered by centrifugation at 4000 x g , disrupted using a One-Shot cell disruptor ( Constant Systems , Northants , UK ) , and His-tagged recombinant proteins were purified by affinity chromatography on a His-Trap FF crude nickel affinity column ( GE healthcare , Buckinghamshire , UK ) using an AKTA fast protein liquid chromatography ( FPLC ) OPC900 P920 system ( GE Healthcare , Buckinghamshire , UK ) . SSl5 was purified using the buffer system outlined previously [9] . SSl7 purification was performed using a buffer containing 50 mM NaH2PO4 , 300 mM NaCl and 10 mM imidazole ( pH 8 . 0 ) . For washing , imidazole concentration was increased to 20mM and a concentration of 250 mM imidazole was used for elution . SElX was purified under denaturing conditions in a lysis buffer of 100 mM NaH2PO4 , 10 mM Tris•Cl and 8 M urea ( pH 8 ) , the protein was washed with the same buffer at pH 6 . 3 then eluted at a pH of 4 . 5 . To further purify SElX , ion exchange chromatography was used by diluting the affinity chromatography elution ( 1/10 ) in 50 mM HEPES ( pH 7 . 5 ) and binding it to a Hitrap SP FF column . Protein was eluted over a NaCl gradient . Purified proteins were dialysed in 1 x PBS ( pH 7 . 3 ) using Spectra/Por Float-A-Lyzer tubing with an 8000 to 10000 Da molecular weight cut off ( MWCO ) ( Spectrum Laboratories , CA , USA ) . Proteins were quantified using a Nano-Drop ND1000 spectrophotomer ( Thermo scientific , USA ) set on the A280 program . After the protein solution spectra were obtained , the concentration of the protein was calculated at an absorbance of 280 nm and the extinction coefficient calculated from the protein sequence as described previously [61] . For binding of recombinant proteins to leukocytes , neutrophils ( 5 × 106 cells/ml ) and PBMCs ( 5 × 106 cells/ml ) were incubated with increasing concentrations of 6 x HIS-tagged recombinant proteins in RPMI 1640 ( Gibco , UK ) and 0 . 05% HSA ( human serum albumin ( Sigma-Aldrich , UK ) , RPMI-HSA ) for 30 min on ice . For some leukocyte binding experiments the cells were also co-incubated with cell surface subset–specific antibodies including anti-CD4 ( Leu-3a ) , -CD8 ( Leu-2a ) , and , -CD19 ( Leu-12 ) ( PE labelled ) ( BD bioscience , Oxford , UK ) . Cells were washed with RPMI-HSA and pelleted at 400 x g for 10 min at 4°C . Binding of the proteins was detected with a FITC-labelled monoclonal mouse anti-HIS tag monoclonal antibody at a 1 . 25 μg/ml final concentration ( AD1 . 1 . 10; LS Bioscience , WA , USA ) , antibodies were incubated with cells for 30 min on ice , washed with RPMI-HSA and pelleted at 400 x g for 10 min at 4°C . Following washing , cells were re-suspended in 200 μl of RPMI-HSA and leukocyte populations were identified based on forward and sideway scatter on a BD FACSCaliburflow cytometer ( Becton Dickinson , Franklin Lanes , NJ ) , and fluorescence was measured . To analyse leukocyte apoptosis , cells were washed one additional time following the binding experiment and re-suspended in assay media containing the nuclear dye DRAQ 5 ( Biostatus , UK ) ( diluted 1/250 ) prior to flow cytometry analysis . For neuraminidase experiments , isolated leukocytes ( 2 x 106 cells/ml ) were pre-treated with 0 . 2 U/ml neuraminidase ( from Vibrio cholera; Sigma-Aldrich , UK ) at 37°C for 45 min at pH 6 . 0 , prior to washing and subsequent incubation with proteins . Site-directed mutagenesis was performed to exchange amino acids in the sequence of SElX with alanine by introducing mutations into the pET15b::selx2 construct using PCR with oligonucleotide primers listed in Table 2 . The reactions were performed using the PfuUltra II Fusion HS DNA polymerase ( Agilent Technologies , UK ) . Primers were used at a final concentration of 250 nM along with 2 mM dNTPs ( Promega , Hampshire , UK ) . PCR cycle conditions were as follows; 1 cycle at 95°C for 2 min , 30 cycles of 95°C for 20 s , 50°C for 20 s and 72°C for 90 s , followed by a final extension of 3 min at 72°C . Following the PCR reaction , DpnI endonuclease ( NEB , Herts , UK ) was added to a final concentration of 0 . 8 U/μl , the reaction was then incubated for 1 h at 37°C followed by an enzyme deactivation step of 10 min at 65°C . Following digestion , 1 μl of the amplified vector product was transformed into E . coli Solopack cells from the Strataclone blunt cloning kit ( Agilent technologies , UK ) following the manufacturer’s instructions . Transformation plates were incubated overnight at 37°C and screened for transformed colonies , which were randomly selected for Sanger sequencing ( Edinburgh Genomics , University of Edinburgh ) . When the mutations were confirmed the plasmid constructs were transformed into BL21 DE3 E . coli cells for protein expression and purification . Following the purification of the mutated proteins , analysis was performed to show the proteins had folded correctly and not become unstable . Far UV circular dichroism ( CD ) spectra of samples were acquired on a Jasco J-710 spectrometer ( Japan Spectroscopic Co . Ltd , Japan ) . Spectra of the proteins , at a concentration of 0 . 4 mg/ml to 1 mg/ml in 0 . 1 x PBS ( pH 7 . 3 ) , was recorded between 190 nm to 250 nm using a cuvette with a path length of 0 . 05 mm . Thermal shift analysis was performed using SYPRO Orange dye ( Life technologies , UK ) and conducted in a Lightcycler 480 ( Roche , West Sussex , UK ) using the melt curve function from 25°C to 95°C . Solutions of 5 μM recombinant staphylococcal protein were mixed with SYPRO Orange dye diluted 1/125 of the stock dye . The plate was heated from 25°C to 95°C in temperature increments of 0 . 1°C/sec . The fluorescence intensity was measured with excitation/emission at 465/510 nm . To prepare an S . aureus construct containing the site-directed mutations in selx the same PCR protocol was applied to pMAD::selx rep [6] as described previously for pET15b::selx2 . Following confirmation of the mutations by sequencing , pMAD::selx EKQD-A was transformed into S . aureus RN4220 , then transduced into LACΔselx using phage 80α [62] . To aid transduction and allelic replacement , all USA300 mutants were made erythromycin sensitive by serial passage on TSA . Following transduction , allelic replacement was performed as described previously [6] . Phenotypic analysis and whole genome sequencing was performed on USA300 LAC constructs ( Table 1 ) to confirm no spurious mutations had occurred ( S3 Fig ) . 250bp Paired End Illumina HiSeq 2500 reads were mapped to the USA300 FPR3757 ( NC_007793 ) reference genome using various tools implemented in the Nesoni pipeline ( https://github . com/Victorian-Bioinformatics-Consortium/nesoni ) . Genome sequence data has been deposited at the European Nucleotide Archive under study number PRJEB20077 with the accession numbers ERS1625171 , ERS1625172 , ERS1625173 and ERS1625174 . Growth curve analysis was performed on a Floustar Omega microplate reader ( BMG Labtech , Germany ) . Overnight cultures of bacteria were diluted 1/100 in either TSB or BHI and growth was recorded over 18 h . SDS-PAGE analysis was performed on concentrated supernatants and cell wall extracts ( CWA ) . To extract CWA proteins , pelleted cells were washed with 1 ml PBS ( Oxoid , Cambridge , UK ) , re-suspended in 1 ml lysis buffer ( 50 mM TrisHCl , 20 mM MgCl2 , 30% Raffinose ( Fluka , UK ) , adjusted to pH 7 . 5 ) containing 200 μg/ml Lysostaphin ( AMBI products LLC , NY , USA ) and protease inhibitors ( Roche , Switzerland ) and incubated at 37°C for 20 min . Samples were centrifuged at 6000 x g for 20 min and CWA proteins were recovered from the supernatant fraction . S . aureus were cultured to an OD600 of 4 . 0 in BHI broth , washed by diluting in PBS , followed by centrifugation at 4000 x g for 10 min . The cells were then diluted to 1x104 CFU/ml in RPMI containing 10% ( v/v ) complement-inactivated serum ( inactivated by incubation at 56°C for 30 min ) . Isolated human neutrophils were infected at a ratio of 1 bacterium to 1000 neutrophils , and incubated for 60 min at 37°C with vigorous shaking . 250 μl of the neutrophil bacterial suspension was diluted into 750 μl of PBS containing 0 . 05% ( v/v ) Triton-X 100 and incubated for 5 min at room temperature to lyse the neutrophils . A control reaction was also prepared; no neutrophils were added to the RPMI with heat inactivated serum and treated in the same way as the test samples . Viable bacteria in each reaction mixture were enumerated by serial dilution and plating on to TSA , followed by overnight incubation at 37°C . Surviving bacteria were counted and compared to the no neutrophil control to determine bacterial survival . Phagocytosis efficacy was measured using FITC-labelled S . aureus USA300 spa::Tn ( Table 1 ) , bacteria were labelled as outlined previously [63] . To determine the level of neutrophil phagocytosis , FITC-labelled S . aureus were mixed with complement-inactivated human serum or purified IgG for 15 min at 37°C in 2 ml 96-well v-bottomed plates ( Corning , USA ) to facilitate opsonisation . Subsequently , isolated human neutrophils , with or without recombinant staphylococcal protein ( pre-incubated for 30 min at room temperature ) , were added at a 10:1 ( bacterium/cell ) ratio and incubated for 15 min at 37°C with shaking at 750 rpm . The reaction was stopped with 1% ( v/v ) paraformaldehyde ( Fisher Scientific , UK ) , and cell-associated fluorescent bacteria were analysed by flow cytometry . Phagocytosis was defined as the percentage of cells with a positive fluorescent signal . Reduction in phagocytosis was calculated by normalising the percentage from test samples to that of uninhibited cells for each opsonin . Protein mediated cytotoxicity was determined on human neutrophils , this was analysed with 1x105 cells/ml suspended in RPMI 1640 ( Gibco , USA ) supplemented with 10% heat-inactivated fetal bovine serum ( Gemini Bio Products , CA , USA ) at a concentration of 1x105 cells/ml . Recombinant S . aureus proteins were added and incubated for 1 h at 37°C with 5% CO2 . Following protein treatment , cells were pelleted by centrifugation at 450 x g at 4°C for 5 min . Lactate dehydrogenase ( LDH ) release was assayed as a measure of neutrophil viability using the CytoTox-ONE homogeneous membrane integrity assay ( Promega , WI , USA ) according to the manufacturer’s specifications . Fluorescence was measured using a PerkinElmer Envision 2103 multilabel reader ( excitation , 555 nm; emission , 590 nm ) ( PerkinElmer , MA , USA ) , and data were normalized to 100% lysis as determined by the addition of 0 . 2% Trition X ( v/v ) to the neutrophils . Detergent-solubilized proteins from primary human neutrophils were incubated with 6 x HIS-tagged SElX and interacting proteins were purified using Ni-NTA resin ( Qiagen , Manchester , UK ) . Proteins eluted from Ni-NTA resin were separated by SDS-PAGE ( 4% to 15% gradient gel ) and stained with SYPRO-Ruby ( Life Technologies , UK ) . Gels were imaged and the lanes were excised for trypsin digestion , protein extraction and liquid chromatography-tandem mass spectrometry ( LC-MS/MS ) analysis as described previously [64] . 96-well ELISA plates ( Maxisorb; Nunc , Thermo , UK ) were coated with 10 μg/ml recombinant SElX , SElX EKQD-A , SSL5 or SSl7 in carbonate/bicarbonate buffer ( 50 mM Na2CO3 and NaHCO3 pH 9 . 6 ) diluted 1/10 in PBS . The plates were incubated overnight at 4°C . Plates were washed with PBS-0 . 05% ( v/v ) Tween-20 ( Sigma-Aldrich , UK ) and blocked with PBS-0 . 05% ( v/v ) Tween-20 and 8% ( w/v ) skimmed milk powder ( microbiology grade; Sigma-Aldrich , UK ) for 1 h at 37°C . Plates were washed with PBS-0 . 05% ( v/v ) Tween20 and incubated with different concentrations of human CD50-HIS recombinant protein ( purchased from Life Technologies , UK ) for 1 h at 37°C . Bound protein was detected using a specific anti-CD50 mouse monoclonal antibody ( MEM-171; Biorad , UK ) and a secondary peroxidase-conjugated rabbit-anti-mouse monoclonal antibody ( Abcam , Cambridge , UK ) . Peroxidase activity was detected with 3 , 3′ , 5 , 5′-Tetramethylbenzidine ( TMB ) liquid substrate ( Sigma-Aldrich , UK ) for 40 s and the reaction was stopped using 1M H2SO4 . Absorbance was measured at 450 nm using a Synergy HT plate reader ( BioTek , Vermont , USA ) . Human PBMC were adjusted to a concentration of 1x106 cells/ml in RPMI 1640 ( Sigma Aldrich , UK ) supplemented with 10% ( v/v ) heat-inactivated fetal calf serum ( Gibco , UK ) , 100 U/ml penicillin , 100 μg/ml streptomycin and 292 μg/ml L-glutamine . ( PSG ) ( Gibco , UK ) . Cells were cultured in 96-well round bottomed tissue culture plates ( Nunc , Fisherbrand , UK ) at 37°C in humidified air with 5% CO2 . Protein samples were tested in triplicate and added at varying concentrations before incubation . Cells were cultured in medium only or with 1 μg/ml of Concanavalin A as negative and positive controls respectively . Proliferation of cells was determined using the incorporation of [3H]-thymidine , by pulsing with 1 μCi/well of [3H]-thymidine , after a 72 h incubation , and harvested after 18 h using a Tomtec Mach III M Harvester 96 ( Hamden , USA ) onto Wallac A filters ( Perkin Elmer , MA , USA ) . A Meltilex A wax scintillant strip ( Perkin Elmer , MA , USA ) was melted onto the filter pad and the [3H]-thymidine incorporation into cellular DNA was determined by scintillation counting using a β-radiation counter ( Wallac 1450 Microbeta PLUS , Perkin Elmer ) and recorded as counts per minute ( CPM ) . The rabbit pneumonia model was performed as described previously with some modification [6] . Briefly , wild-type LAC and the selx mutants ( Table 1 ) were cultured in Todd Hewitt broth for 16 h and washed once in Todd Hewitt broth . The bacteria were re-suspended in Todd Hewitt broth at 1×1010 CFU/ml for use in injections . New Zealand White rabbits ( 3–4 per group ) were anesthetized with ketamine and xylazine . Their tracheas were exposed and a high ( 6x109 CFU ) or low ( 2×109 CFU ) dose of USA300 CA-MRSA strain LAC or mutant derivatives ( Table 1 ) were administered intra-tracheally through catheters in 0 . 4 ml volumes . The animals were closed and monitored for 4 d for development of fatal necrotizing pneumonia . Female , inbred 8–12 week old female BALB/cOlaHsd mice aged between 8–12 weeks were obtained from ENVIGO ( UK ) and acclimatised for 1 to 2 weeks in the facility before being used in infection challenge studies . All mice were housed under specific pathogen-free conditions in individually ventilated cages ( IVCs ) at the Bioresearch Facility ( BRF ) of the Roslin Institute ( University of Edinburgh , UK ) . 1 day prior to infection challenge , mice were anesthetized with isoflurane and the fur on the back of each animal was removed by using clippers . Overnight cultures of S . aureus were inoculated 1:100 into fresh BHI broth and cultured to mid-log phase ( OD600 = 0 . 6; approximately 2 h ) with shaking at 37°C . Staphylococci were harvested by centrifugation , washed , and suspended in sterile PBS to obtain an inoculum of 1x108 CFU/ml . Inocula were determined by CFU enumeration following serial dilution , plating on TSA plates , and overnight growth at 37°C . Prior to inoculation the animals were weighed and then every 24 h post infection . Mice were inoculated with 1x107 CFU by subcutaneous injection to a depth of 5 mm , in the shaved area of skin on the back of each mouse . Infected animals were monitored for health status and lesion development over a period of 6 d using a standardized and Home Office project licence approved monitoring protocol . The size of each skin lesion was measured each day ( one measurement across the longest dimension of the lesion ) . Bacterial tissue load was determined post-mortem , the skin lesion was excised , weighed and homogenised in 1 ml of sterile PBS using fast prep tubes containing lysis matrix D ( MP Biomedicals , UK ) . Bacterial loads were determined by CFU enumeration following serial dilution , plating on TSA plates , and overnight growth at 37°C . CFU were normalised to tissue weight . For histopathological analysis lesions were excised and fixed in 10% neutral buffered formalin ( NBF ) ( Leica microsystems , UK ) for 24 h . Lesions were processed to parrifin blocks , sectioned and then stained with haematoxylin and eosin by the Veterinary Pathology Service Unit of the Royal ( Dick ) School of Veterinary Studies ( University of Edinburgh ) . All statistical analysis was performed in Graphpad Prism 7 . Grouped data was analysed to determine if a Gaussian distribution was true with the Shapiro-Wilk normality test . Parametric data was analysed using student t-test with Welches correction if required . Tests were unpaired and two-tailed and significant differences were considered when the p-value was <0 . 05 . Protein dose curves were tested using two-way ANOVA analysis , multiple comparisons were performed using Sidak’s method . Animal survival curves were plotted using Kaplan-Meier method and significance was determined using log-rank ( Mantel-Cox ) analysis ( p-value <0 . 05 ) . | Staphylococcus aureus is a bacterial pathogen responsible for an array of disease types in healthcare and community settings . One of the keys to the success of this pathogen is its ability to subvert the immune system of the host . Here we demonstrate that the superantigen ( SAg ) staphylococcal enterotoxin-like toxin X ( SElX ) contributes to immune evasion by inducing unregulated T-cell proliferation , and by inhibition of phagocytosis by neutrophils . We observed that the capacity to bind neutrophils appears to be central to the SElX-dependent toxicity observed in a necrotising pneumonia infection model in rabbits . We report the first example of a staphylococcal SAg with two independent immunomodulatory functions acting on distinct immune cell types . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Material",
"and",
"methods"
] | [
"blood",
"cells",
"medicine",
"and",
"health",
"sciences",
"chemical",
"characterization",
"immune",
"cells",
"pathology",
"and",
"laboratory",
"medicine",
"pathogens",
"immunology",
"cell",
"processes",
"microbiology",
"rabbits",
"vertebrates",
"staphylococcus",
"aureus"... | 2017 | The Staphylococcus aureus superantigen SElX is a bifunctional toxin that inhibits neutrophil function |
Ethiopia is one of the few African countries where Plasmodium vivax is co-endemic with P . falciparum . Malaria transmission is seasonal and transmission intensity varies mainly by landscape and climate . Although the recent emergence of drug resistant parasites presents a major issue to malaria control in Ethiopia , little is known about the transmission pathways of parasite species and prevalence of resistant markers . This study used microsatellites to determine population diversity and gene flow patterns of P . falciparum ( N = 226 ) and P . vivax ( N = 205 ) , as well as prevalence of drug resistant markers to infer the impact of gene flow and existing malaria treatment regimes . Plasmodium falciparum indicated a higher rate of polyclonal infections than P . vivax . Both species revealed moderate genetic diversity and similar population structure . Populations in the northern highlands were closely related to the eastern Rift Valley , but slightly distinct from the southern basin area . Gene flow via human migrations between the northern and eastern populations were frequent and mostly bidirectional . Landscape genetic analyses indicated that environmental heterogeneity and geographical distance did not constrain parasite gene flow . This may partly explain similar patterns of resistant marker prevalence . In P . falciparum , a high prevalence of mutant alleles was detected in codons related to chloroquine ( pfcrt and pfmdr1 ) and sulfadoxine-pyrimethamine ( pfdhps and pfdhfr ) resistance . Over 60% of the samples showed pfmdr1 duplications . Nevertheless , no mutation was detected in pfK13 that relates to artemisinin resistance . In P . vivax , while sequences of pvcrt-o were highly conserved and less than 5% of the samples showed pvmdr duplications , over 50% of the samples had pvmdr1 976F mutation . It remains to be tested if this mutation relates to chloroquine resistance . Monitoring the extent of malaria spread and markers of drug resistance is imperative to inform policy for evidence-based antimalarial choice and interventions . To effectively reduce malaria burden in Ethiopia , control efforts should focus on seasonal migrant populations .
Despite considerable progress towards malaria control , two-thirds of the population in Ethiopia , i . e . , approximately 66 million people , reside in areas of low or high malaria transmission [1] . Apart from human factors such as population mobility , urbanization , and agricultural development , emergence of drug resistant parasites and insecticide resistance present a major hurdle to malaria control programs in Ethiopia and worldwide [2] . Reports of emerging Plasmodium vivax resistance to chloroquine ( CQ ) in Ethiopia threaten the efficacy of P . vivax treatment [3–6] . Also , the well-documented emergence of P . falciparum resistance to artemisinin in Southeast Asia may endanger current malaria treatment programs in Ethiopia , given that both CQ and sulfadoxine-pyrimethamine ( SP ) resistance originated in Southeast Asia and spread quickly to East Africa [7] . Thus , knowing how malaria parasites spread as well as monitoring prevalence of drug resistant markers in high-risk areas are important to informing antimalarial interventions . While P . vivax is the most widespread human malaria parasite , it is rare in Africa where P . falciparum predominates . Due to its low prevalence in the continent , little is known about the transmission patterns of P . vivax in Africa . Ethiopia is unique in that P . vivax is co-endemic with P . falciparum at approximately equal case incidence rates . Other African countries with significant P . vivax infections are Eritrea , Sudan , and Madagascar [1] . Although Ethiopia carries a substantial malaria burden , information on the transmission dynamics and spread of drug resistance across the country is scarce . P . vivax and P . falciparum exhibit different biological and epidemiological features . Compared to P . falciparum , P . vivax has a broader temperature tolerance , an early onset of gametocyte development , and a dormant life cycle stage , the hypnozoite , in the host liver that can cause relapse . Relapse infections may present opportunities for P . vivax to exchange and disseminate alleles at any time of the year rather than only the transmission season [8] . Population genetic diversity and structure are thus expected to be different between P . vivax and P . falciparum even when the two species coexist . For instance , in Cambodia [9] , the Indo-West Pacific [10–12] , and the Brazilian Amazonia [13] , P . vivax revealed a higher microsatellite diversity than its sympatric P . falciparum . A similar contrast was observed in Papua New Guinea where P . vivax showed a higher AMA1 gene diversity than P . falciparum [14] . Globally , both P . falciparum and P . vivax in Africa were markedly differentiated from those in Southeast Asia and Oceania [15 , 16] , reflecting a clear continental disjunction . While P . vivax is genetically most diverse in Southeast Asia [17 , 18] , P . falciparum diversity is the highest in East and West Africa compared to Southeast Asia and Oceania [19 , 20] . Such differences could be tightly associated with the historical levels of transmission intensity . Comparing genetic diversity and structure between the two species at the same endemic setting would shed light on the biological relevance on malaria epidemiology . Apart from transmission dynamics , the biological differences between P . falciparum and P . vivax have added a layer of complexity to antimalarial treatment programs in Ethiopia . First-line treatment for P . falciparum is arthemether-lumefantrine ( AL ) , which replaced SP in 2005 due to increasing and widespread SP resistance . While CQ was withdrawn in 1998 due to a high prevalence of CQ resistance in P . falciparum , it remains the first-line treatment for P . vivax in Ethiopia [2] . Genetic markers for CQ ( pfcrtT76 ) , SP ( pfdhfrI51-R59-N108 + pfdhpsG437-E540 ) , and artemisinin ( Kelch13-propeller region ) resistance in P . falciparum have been well documented [21–24] . For P . vivax , although there is no clear evidence that variants in pvcrt-o and pvmdr1 are associated with CQ resistance , mutations including T958M , Y976F and F1076L in pvmdr1 , as well as a K-10 insertion ( lysine ( K ) insertion on chromosome 10 ) in pvcrt-o have been suggested as possible genetic markers [25 , 26] . Mutation frequency of these genes largely depends on the level of drug usage and the extent of the spread of resistant genotypes . For instance , the pfcrtT76 mutation frequency was almost 100% among clinical and asymptomatic P . falciparum samples from 2004–2012 in south-central Ethiopia [27–29] . In the case of mixed infections with the two species where only P . vivax diagnosed , P . falciparum is still exposed to CQ despite the change of P . falciparum first-line treatment more than a decade ago . On the contrary , the frequency of pfdhfr and pfdhps quintuple mutations had decreased significantly from 2005 to 2008 since the withdrawal of SP in 2005 [30 , 31] , indicative of relaxed selection in the P . falciparum populations . Mutations in the kelch ( K13 ) -propeller region are markers of artemisinin resistance [24] . Resistance-associated pfK13 mutations are widely prevalent in Southeast Asia but those mutations have not yet been common in Africa [32] . Recently , a new nonsynonymous mutation at pfK13 position 579 ( M579I ) was detected in a P . falciparum strain that was indigenous to Equatorial Guinea and shown to be artemisinin resistant based on in vitro testing [33] . Careful surveillance of pfK13-propeller region mutations in Ethiopia will be especially useful to detecting the spread of artemisinin resistance from Southeast Asia to Africa . This study examined and compared population diversity and gene flow patterns between P . falciparum and P . vivax in the northern , eastern , and southern parts of Ethiopia with low to moderate level of malaria transmission . Specifically , we investigated if landscape heterogeneity impacts parasite gene flow by testing the association between landscape factors and population genetic structure . We tested three competing hypotheses of factors that may influence gene flow: 1 ) factors related to vector ecology ( land cover and precipitation ) ; 2 ) factors related to human movement ( distance to roads ) ; and 3 ) factors related to environment ( elevation , which is tightly correlated with temperature ) . Further , we inferred how gene flow pattern relates to the prevalence of drug resistance markers . This knowledge will help inform how malaria parasites and drug resistance spread , how P . vivax and P . falciparum epidemiology influences genetic structures , as well as antimalarial drug efficacy in Ethiopia . Monitoring for markers of antimalarial drug resistance is imperative to informing public health interventions .
Scientific and ethical clearance was obtained from the institutional scientific and ethical review boards of Jimma and Addis Ababa Universities in Ethiopia and University of California , Irvine , USA . Written informed consent/assent for study participation was obtained from all consenting heads of households , parents/guardians ( for minors under age of 18 ) , and each individual who was willing to participate in the study . Clinical samples from six study sites representing the northern highland ( MA: Mankush and BU: Bure ) , eastern Rift Valley ( SR: Shewa Robit and ME: Metehara ) , and southern basin area ( JM: Jimma and HA: Halaba ) of Ethiopia were collected during the peak transmission season ( September-November ) of 2014 ( Fig 1; S1 Table ) . This area encompasses an elevation gradient from ca . 50m in the basin to over 2 , 500m in the highlands west of the Great Rift Valley . Finger-prick blood samples were collected from malaria symptomatic ( who has fever with axillary body temperature > 37 . 5°C and with confirmed asexual stages of malaria parasite based on microscopy ) or febrile patients visiting the health centers or hospitals at each of the study sites . Thick and thin blood smears were prepared for microscopic examination and three to four spots of blood , equivalent to ~50 μl , from each individual were blotted on Whatman 3MM filter paper . Parasite DNA was extracted from dried blood spots by the Saponin/Chelex method [34] . Nested and quantitative PCR were performed to identify and confirm parasite species of the infected samples [35] . A total of 226 and 205P . falciparum and P . vivax samples ( ranged from 18–58 samples per site ) were included in microsatellite analyses . Thirteen single-copy microsatellites with tri- or tetranucleotide repeats , which mapped to 14 chromosomes , were typed for P . falciparum and P . vivax , respectively ( S2 Table ) . Alleles were PCR-amplified with the published oligonucleotide primers [36–38] . For each PCR reaction , 2 μl of genomic DNA were used with 2 mM MgCl2 , 2 μM of each primer ( all forward primers were labeled with fluorescent dyes; Applied Biosystems , Foster City , CA ) , and 10μl of 2×DreamTaq Green PCR Master Mix ( Thermo Scientific , Waltham , MA ) in a final volume of 20 μl . PCR cycling conditions were as follows: 2 min , 94°C; ( 30 sec , 94°C; 40 sec , 58°C; 50 sec , 72°C ) for 40 cycles; 5 min , 72°C . After PCR amplification , products were pooled into four groups based on size differences: TAA87+PFPK2+POLY2+9735 , TAA42+TAA81+TAA109 , PE87a+PFG377+POLYα , TAA60+TA80+TA116 for P . falciparum; MS1+MS3+MS4+MS5 , MS8+MS9+MS16 , MS10+MS12+MS15 , MS20+Pv1 . 501+Pv3 . 27 for P . vivax ( S2 Table ) . The pooled products were separated on an ABI 3730 sequencer and all allele sizes were determined and visualized in Peak Scanner . To avoid background signal and potential artifacts , a threshold of 500 relative fluorescent units was set for peak detection . For each sample , the dominant allele and any alleles with a minimum of 33% height of the dominant allele were scored [36] . A model-based Bayesian method implemented in STRUCTURE v2 . 3 . 4 was performed to examine partitioning of individuals to genetic clusters [43] . The number of clusters ( K ) was determined by simulating a range of K values from 1 ( no genetic differentiation among all sites ) to 6 ( all sites were genetically differentiated from one another ) . The posterior probability of each value was then used to detect the modal value of ΔK , a quantity related to the second order rate of change with respect to K of the likelihood function [44] . Posterior probability values were estimated using a Markov Chain Monte Carlo ( MCMC ) method . A burn-in period of 500 , 000 iterations followed by 106 iterations of each chain was performed to ensure convergence of the MCMC . Each MCMC chain for each value of K was run ten times with the ‘independent allele frequency’ option that allows individuals with ancestries in more than one group to be assigned into one cluster . Individuals were assigned into K clusters according to membership coefficient values ( Q ) ranged from 0 ( lowest affinity to a cluster ) to 1 ( highest affinity to a cluster ) . The partitioning of clusters was visualized with DISTRUCT [45] . Neighboring-joining trees were constructed using T-REX [46 , 47] to show the genetic relatedness among P . falciparum and P . vivax samples . The squared Euclidean distance , which is based on the number of times a certain allele found in two individuals [48] , was used for tree constructions . The resulted trees were visualized in FigTree v1 . 4 . 2 . An FST analysis was conducted using θ , an FST-estimator in SPAGeDi v1 . 2e [49] . FST values were tested for significance using 10 , 000 permutations . Genetic differentiation among sites was displayed by multidimensional scaling plot based on the estimated DS values ( an analog of FST ) in R v3 . 3 . 0 . Furthermore , an analysis of molecular variance ( AMOVA ) was used to determine the hierarchical distribution of genetic variance within and among populations , as well as among regions ( north , east , and south of Ethiopia ) using GENALEX [50] . The relationships between genetic distances ( DS values ) and Euclidean geographical distance ( estimated from spatial coordinates using R for multivariate and spatial analysis; [51] ) were examined by Mantel tests ( 10 , 000 randomizations ) and reduced major axis ( RMA ) regression in the Isolation By Distance v3 . 23 [52] . Signature of genetic bottleneck was detected with BOTTLENECK v1 . 2 . 02 [53] . Only sites with a sample size of 20 or above were included for statistical significance . Two tests were performed using three different mutation models: the infinite alleles model ( IAM ) , the stepwise mutation model ( SMM ) , and a combination of those two extreme hypotheses , the two-phase model ( TPM ) . First was the overall distribution of allele frequency classes . Second was the Wilcoxon-signed rank test to compare the number of loci that present a heterozygosity excess to the number of such loci expected by chance only . Frequency of gene flow among populations was estimated for each parasite species by a maximum-likelihood analysis implemented in Migrate-N v2 . 4 . 4 [54] . Parameters including Θ ( defined as 4Neμ , where Ne is the effective population size and μ is the mutation rate per generation and site ) and M ( m/μ , where m is the immigration rate scaled by mutation rate ) were estimated . Four independent runs were conducted with the Brownian motion model using 10 short chains with 5 , 000 sampled genealogies and three long chains with 50 , 000 sampled genealogies to obtain the mean and range of Θ and M values . In addition , we inferred migration rate using a Bayesian approach implemented in BayesAss v3 [55] , which is not dependant on the assumption of equilibrium and can be used with populations that are not in migration-drift or Hardy-Weinberg equilibrium . A MCMC algorithm was used to estimate the posterior probability distribution of the proportion of migrants from one population to another . We performed the analyses with 9×106 iterations , with a burn-in of 106 iterations , and a sampling frequency of 2 , 000 to ensure the parameters of the model were converged . The correlation between migration rate and geographical distance was tested for all pairs of populations . To test for the effects of landscape factors on gene flow between populations , we performed a landscape genetic analysis as follows . First , we created landscape resistance surfaces based on our predictions of the factors influencing gene flow of Plasmodium species , specifically factors influencing vector ecology ( land cover and precipitation ) , human movement ( distance to roads ) , and Plasmodium biology ( elevation , which is tightly correlated with temperature ) . A resistance surface is a spatial layer in which each cell in a grid is assigned a value that represents the degree to which that cell constrains gene flow or movement [56] . These values were often based on numerous assumptions about relationships between a landscape or environmental feature and the ability of a given organism to move through that feature . Landscape resistance surfaces were derived from publicly available data: NASA MODIS MCD12Q2 for land cover ( forest , shrubland , woody savanna , savanna , grassland , cropland , and sparsely vegetated ) [57 , 58]; NASA SRTM v4 . 1 for elevation [59]; WorldClim v1 . 4 for precipitation [60]; and Roads Africa shapefile in ArcGIS for distance to roads . All raster files were resampled to a resolution of 1km in ArcGIS 10 . Next , we used ResistanceGA to optimize landscape resistance surfaces based on our genetic data [61] . ResistanceGA uses a genetic algorithm to unbiasedly assign landscape resistance values to continuous or categorical data . Circuitscape v . 4 . 05 was used to measure resistance distance between populations [62] . Circuitscape relies on electrical circuit theory to predict landscape connectivity and incorporates all possible pathways between populations into the resistance distance measure . To test the fit of resistance surfaces in relation to the genetic data , linear mixed effects models with the maximum likelihood population effects ( MLPE ) were fit in lme4 [63] . Finally , Akaike information criterion with a penalty for extra parameters ( AICc ) was calculated from the linear mixed effect model and used as the means for model selection . Five gene regions that are putatively associated with CQ ( pfcrt and pfmdr1 ) , SP ( pfdhps and pfdhfr ) , and artemisinin ( pfK13 ) resistance were sequenced with P . falciparum samples . Polymorphisms were examined for the following codons: pfcrt–codon76; pfmdr1–codons 86 , 184 , 1042 , and 1246; pfdhps–codons 396 , 436 , 437; pfdhfr–codons 51 , 59 , 108; pfK13–codons 476 , 493 , 519 , 532 , 539 , 543 , 578 , 579 , 580 , 582 , and 590 . In addition , two gene regions that are putatively associated with CQ resistance ( pvcrt-o and pvmdr1 ) were sequenced with P . vivax samples . Polymorphisms were examined for the following: pvcrt-o–a ( AAG ) insertion at codon 10 ( K10 insert ) , codon 117; pvmdr1–codons 958 , 976 , and 1076 . Amplification was conducted in a 20μl reaction mixture containing 3μl of genomic DNA , 12 . 5μl of 2×DreamTaq Green PCR Master Mix , and 10 nmol of forward and reverse primers based on the published protocols [21–23 , 32 , 64] . PCR products were then purified the by the SAP-ExoI method ( Affymetrix , Santa Clara , CA ) and sequenced in both directions by Sanger sequencing ( GENEWIZ ) . The pfmdr1 gene copy number of P . falciparum was assessed by real-time PCR . Genomic DNA of P . falciparum clones 3D7 ( which has a single copy of pfmdr1 ) was used as a calibrator and pfβ-tubulin , a house-keeping gene , was used as an internal control . The primers for pfmdr1 and β-tubulin were described previously [65] . For P . vivax , the Salvador I strain was used as a calibrator and the pvaldolase gene , which is known to be a single copy gene in P . vivax , was used as an internal control using the published primers [66] . Amplification was performed in triplicate in a total volume of 20 μl containing 10μl of SYBR Green PCR Master Mix , 0 . 75 μl of each of the sense and anti-sense primers ( 10 μM ) , 20 ng of genomic DNA and 3 . 5 μl of water . Reaction was performed in CFX96 Touch™ Real-Time PCR Detection System ( Bio-Rad ) , with an initial denaturation at 95°C for 3 min , followed by 45 cycles at 94°C for 30 sec , 55°C for 30 sec , and 68°C for 1 min with a final 95°C for 10 sec . This was then followed by a melting curve step of temperature ranged from 65°C to 95°C with 0 . 5°C increment to determine the melting temperature of each amplified product . A negative control with no template was used in each run . Each sample was run in triplicates and the Ct values and melting temperature were recorded at the end of the reactions . The average and standard deviation of the three Ct values were calculated , and the average value was accepted if the standard deviation was lower than 0 . 32 . The 2ΔΔCt±SD method for relative quantification was used to estimate the gene copy number [66] and the results were expressed as the N-fold copy number of the targeted gene in relative to the reference . Fisher’s exact test ( given small sample size ) was used to test for significant differences in mutation prevalence and gene copy number among the study populations . All statistical analyses were performed in R ( R Core Team 2013 ) .
No significant LD was detected for all pairwise combinations of microsatellite loci among the P . falciparum and P . vivax samples ( Bonferroni corrected P>0 . 05 ) . However , when all locus were pooled together in the analyses , P . falciparum in general showed a higher level of linkage and/or rate of recombination ( IAS values ranged from 0 . 005 in Bure ( BU ) to IAS = 0 . 13 in Halaba ( HA ) ; all sites IAS = 0 . 03 , P<0 . 05 except BU ) than P . vivax ( IAS values ranged from 0 . 003 in Mankush ( MA ) and Jimma ( JM ) to 0 . 02 in Shewa Robit ( SR ) ; all sites IAS = 0 . 001 , P>0 . 05; Table 1 ) . Compared to P . falciparum ( 8 . 8%; 20/226; Table 1 ) , P . vivax indicated a lower rate of polyclonal infections ( 4 . 3%; 9/205 ) . Polyclonal samples were observed in all sites for P . falciparum , with the highest rate of polyclonal infections in the southern lowlands ( HA: 16 . 7% and JM: 11 . 8% ) . For P . vivax , polyclonal infections ranged from 5 . 3% in Bure ( BU ) to 0% in Shewa Robit ( SR ) , despite a slightly smaller sample size . Likewise , MOI for P . falciparum from all sites ( mean MOI = 1 . 10; Table 1 ) was significantly higher than that of P . vivax , ( mean MOI = 1 . 04 , P<0 . 01 ) , indicative of a higher complexity within P . falciparum infections . Among all the polyclonal infections , 18 were bi-clonal of which two equally dominant alleles were detected in a single locus . We separated the genotypes of the two strains and included them in the analyses . For the 11 samples that showed >1 alleles in two or more loci , we were unable to confidently differentiate the genotypes of the different strains and thus these samples were discarded in the analyses ( S3 Table ) . Both P . falciparum and P . vivax revealed similar levels of allelic and genotypic diversity ( Table 2 ) . Nevertheless , genotypic evenness in P . vivax from the highlands ( E = 0 . 75; BU and MA ) was significantly lower than the other samples ( P<0 . 01; two-tailed t-test; Table 2 ) . This suggested a less even distribution of genotypes in these populations and that some genotypes were more common than the others . AMOVA indicated that most of the genetic variation was within populations in both P . falciparum and P . vivax ( >84%; S4 Table ) . In P . falciparum , a greater proportion of variation was found among regions ( 13% among the north , east and south Ethiopia ) than among populations in a region ( 3% ) ; whereas in P . vivax , the proportion of variation among regions ( 8% ) and populations ( 7% ) were comparable and significant . Populations in north and east Ethiopia were slightly differentiated from those in the south ( Fig 1 ) . This pattern was shown in both P . falciparum and P . vivax , though populations were more scattered in P . falciparum . The two northern populations ( BU and MA ) were genetically close to samples in the eastern Rift Valley ( SR and ME; Fig 1 ) . In P . falciparum , three most probable genetic clusters were detected by STRUCTURE analyses ( Fig 2 ) , but these clusters did not clearly represent geographical regions . The red cluster was most apparent in the northern ( MA and BU ) and eastern ( SR and ME ) populations , but less significant in the southern populations ( JM and HA ) . All three genetic clusters were found in sites BU and ME , but the blue cluster was almost absent in MA and SR ( Fig 2 ) . In P . vivax , samples from north and east Ethiopia constituted predominantly the purple cluster , contrast with those from the south that constituted an admixture of the purple and yellow clusters . Unlike P . falciparum , the genetic composition between the northern and eastern populations was largely similar . Neighbor-joining trees did not indicate clear distinction among the P . falciparum and P . vivax population samples ( Fig 3 ) . Both trees had relatively short internodes but long terminal branches , which suggested that the parasite lineages were rapidly diverged from one another and that frequent gene exchange occurred among the populations . In P . falciparum , there were a number of subclades where parasites from the same population were genetically closely related ( e . g . , subclades I , II , and III in Fig 3A ) . However , in P . vivax , samples from different sites were clustered together in the same clade without clear differentiation ( Fig 3B ) . Mantel tests indicated no significant association between geographical and genetic distances among populations of P . falciparum ( R2 = 0 . 14 , P>0 . 05 ) and P . vivax ( R2 = 0 . 09 , P>0 . 05 ) , respectively . These results suggest that parasite gene flow was not limited by geographical distance . Further , for both P . falciparum and P . vivax , we found that none of the tested landscape factors explained pairwise genetic distance ( FST ) among populations more than the Euclidean distance alone based on AICc ( Table 3 ) . These results indicated that the differences in land cover , elevation , precipitation , and distance to roads ( a proxy for accessibility ) did not significantly influence parasite gene flow and that our study populations were clearly connected ( S1 Fig ) . All populations of P . falciparum showed a normal L-shape distribution in allele frequency ( Table 4 ) , suggesting that these populations did not experience a recent severe bottleneck . In P . vivax , allele frequency was shown with a shifted mode in site SR ( east Ethiopia ) , indicative of a significant genetic bottleneck . In addition , a significant excess of heterozygosity was observed in this population as well as other northern and eastern populations ( ME and BU ) under the IAM and SMM mutation models , suggestive of a deviation in the mutation-drift equilibrium . Based on Migrate-N analyses , both P . falciparum and P . vivax showed a relatively small effective population size ( Θ = 0 . 1–0 . 6 and 0 . 17–0 . 88 , respectively; S5 Table ) , which suggested that the effect of drift was unequivocally as significant as migration . Given that most values of M ( m/μ ) were >1 , the effect of migration ( m ) was larger than the effect of mutation ( μ ) . For P . falciparum , the effective number of migrants per generation Nem ranged from 0 . 12–8 . 58 . The greatest migration was observed between the north and east Ethiopia ( e . g . , from BU to SR: M 30 . 12 and Nem 8 . 58 ) and these north-east migrations were frequent and mostly bidirectional . While migrations between the north-east and south Ethiopia appeared to be less significant , these migrations in most cases were bidirectional ( Fig 4 ) . For P . vivax , the effective number of migrants per generation Nem ranged from 0 . 12–3 . 58 . The greatest migration was between the northern populations ( e . g . , from MA to BU: M 22 . 58 and Nem 3 . 58 ) followed by the migration from the north and east to south Ethiopia ( e . g . , from ME to JM: M 18 . 77 and Nem 2 . 02; from MA to JM: M 16 . 60 and Nem 1 . 96 ) . Interestingly , the migrations to the south were primarily unidirectional ( Fig 4 ) . BayesAss analyses supported the estimates of migration rates from Migrate-N . No significant correlations were found between migration rate and geographical distance in P . falciparum ( R2 = 0 . 13 , P = 0 . 09 ) and P . vivax ( R2 = 0 . 02 , P = 0 . 38; S2 Fig ) . Among the 226 P . falciparum and 204 P . vivax samples , we successfully amplified and obtained complete resistance gene data in 199 ( 88% ) and 185 ( 90% ) of the samples ( S7 Table ) . Samples with incomplete data were excluded in the analyses . Plasmodium falciparum samples from north , east and south Ethiopia all revealed a similar pattern of mutations in pfcrt and pfmdr1 , the genes that associated with chloroquine resistance . In pfcrt , about 54–62% of the samples were shown with a mutant 76T genotype ( north: 37/65 = 57%; east: 45/72 = 62 . 5%; south: 34/62 = 54 . 8%; Fig 5 ) . While the majority of P . falciparum samples showed the wild type N86 , N1042 , and D1246 of pfmdr1 , over 85% ( north: 56/65 = 86%; east: 72/72 = 100%; south: 53/62 = 85 . 5%; Fig 5 ) of the samples had the mutant 184F genotype . Also , qPCR data indicated that over 60% of the samples had two or more copies of the pfmdr1 gene ( north: 42/65 = 64 . 6%; east: 46/72 = 63 . 9%; south: 38/62 = 61 . 3%; Fig 5 ) . The rate of mutations observed in pfcrt and pfmdr1 was not significantly different among populations . By contrast , the pattern of mutation in genes pfdhps and pfdhfr that associated with SP resistance appeared to vary among geographical regions in Ethiopia ( Fig 5 ) . For instances , 62% ( 40/65 ) of P . falciparum in the northern populations had the mutant 396K of pfdhps , which was significantly higher than that in the eastern ( 8/72 = 11% ) and southern populations ( 24/62 = 38% ) . While both the eastern and southern populations showed a preponderance of pfdhfr mutations in codons 51 ( 51I genotype; east: 72/72 = 100%; south: 48/62 = 77 . 4% ) , 59 ( 59R; east: 43/72 = 59 . 7%; south: 41/62 = 66% ) and 108 ( 108N; east: 72/72 = 100%; south: 49/62 = 79% ) , the northern populations showed a significantly lower rate of mutations in these positions ( 51I: 34/65 = 52 . 3%; 59R: 15/65 = 23 . 2%; 108N: 34/65 = 52 . 3%; P<0 . 05 ) . Amplification and sequencing of the entire pfK13 indicated that this gene was highly conserved among all the P . falciparum samples . No polymorphisms were detected at the codon positions putatively related to artemisinin resistance ( Fig 5; S7 Table ) . For P . vivax samples from the north , east and south Ethiopia , all revealed a similar pattern of mutations in pvcrt-o and pvmdr1 , the genes that associated with chloroquine resistance . Sequences of these two genes were highly conserved among the samples . Almost all had the wild type genotype except pvmdr1 codon 976 where over 50% of the samples had the mutant 976F genotype ( north: 35/52 = 67 . 3%; east: 20/38 = 52 . 6%; south: 52/93 = 55 . 9%; Fig 5 ) . qPCR data indicated that less than 4% of the samples had two or more copies of the pvmdr1 gene .
Ethiopia is a unique malaria endemic country in sub-Saharan Africa where P . falciparum and P . vivax coexist . The present study showed that both species revealed similar population structure . The northern and eastern populations of P . falciparum and P . vivax were genetically closely related and slightly distinct from the southern populations . While parasite gene flow was most frequent between the northern highlands and the highland-fringe areas along the Rift Valley , the southern basin populations were not excluded . We did not find a significant association between any landscape factors and population genetic structure . This result may be caused by the chosen metric for human movement ( distance to roads ) , which does not fully capture the seasonal migration patterns that coincide with harvest season ( i . e . , September to November ) . The seasonal migration from highland to lowland areas for agricultural harvest appears to most closely reflect that of the observed gene flow patterns . Since the end of civil war in 1991 , seasonal migration of Ethiopians from highland and highland-fringe areas to lowlands has increased due to the growth of large-scale agricultural development projects [67] . Sugarcane production and coffee plantations in the lowlands are important sources of employment and income to people who live in the highlands where agricultural activities are scarce . It is possible that agricultural employments in the eastern Rift Valley and southern basin areas elicit seasonal human migrations and consequently enable spread of P . falciparum and P . vivax across broad areas without landscape or distance barriers [68 , 69] . For example , the lowland districts of which populations can increase by 20–30% as a result of the arrival of tens of thousands of farmers during the harvest season [70 , 71] . Hence , in addition to maintaining control efforts in the community , seasonal migrant populations should not be ignored in order to effectively reduce malaria burden in Ethiopia . This may include close monitoring of malaria symptoms when seasonal migrant farmers return to their home village after the harvest season , as well as offering additional instructions and aids on prevention and prompt treatment of malaria . At the gene level , P . falciparum and P . vivax revealed a comparable level of diversity within and among populations . The mean He in our P . falciparum populations ( 0 . 66 ) was similar to those reported in less endemic regions such as the Caribbean ( He = 0 . 61 ) [72] and Indonesia ( He = 0 . 53 ) [11] . Compared to endemic areas in South Pacific and Southeast Asia e . g . , Papua New Guinea ( He = 0 . 84 ) [12] , central Vietnam ( He = 0 . 88 ) [73] , Cambodia ( He = 0 . 84 ) [9] as well as a previous study in south-central Ethiopia ( He = 0 . 82 ) [74] , the mean He in our P . vivax populations ( 0 . 69 ) was slightly lower but similar to that in Asendabo ( He = 0 . 70 ) [15] , which is about 50km away from our study site Jimma in southern Ethiopia . The moderate and comparable genetic diversity observed in both species in Ethiopia contrasts with those reported in South Pacific and Southeast Asia where P . vivax showed a higher microsatellite diversity and less fragmented gene pool than the sympatric P . falciparum [9–12] . In countries or areas where malaria transmission is intense and stable , relapse could be common . This may , in turn , facilitate recombination and local spread of P . vivax leading to higher within-population diversity and lower among-population differentiation compared to P . falciparum . By contrast , our study sites in Ethiopia represent areas of low to moderate transmission settings where the relapse rate of P . vivax is largely unknown [35 , 75] . Malaria there displays a strong seasonal pattern with a lag time varying from a few weeks at the beginning of the rainy season to more than a month at the end of the rainy season [76] . The rainy season is relatively short in the highland and highland-fringe areas , and thus transmission season is usually short-lived [77] . These apparent seasonal and/or landscape differences not only influence the behavior and distribution of vector mosquitoes and the length of the parasite life cycle [78] , but also the patterns of human movement and settlement that can in turn determine transmission dynamics of malaria . For instances , although P . vivax infections can occur periodically throughout the year , human migration is seasonal in Ethiopia and this could limit the spread of relapse P . vivax . Also , the long arid or semi-arid season particularly in the highlands may constrain gametocyte development of P . vivax or local transmission even when relapse occurs . A recent study in southern Ethiopia indicated an approximately 9 . 4% ( ranged from 6 . 4–13 . 6% by sites ) of P . vivax patients showed recurrent infection by day 28 [5 , 6] . However , because relapses can occur as early as 21 days following initial treatment , it is unclear how many of the recurrent cases were due to relapse . Given that our P . falciparum and P . vivax samples were collected at the same time during the peak transmission season , the lack of contrast between P . falciparum and P . vivax population structure and diversity may suggest similar demographic factors and a less significant impact of relapse . Future study should investigate the incidence of relapse among transmission settings and how such influences parasite population diversity . The overall polyclonal infection rates observed in P . falciparum ( 8 . 8% ) and P . vivax ( 4 . 3% ) were low in our study area . The proportion of polyclonal P . vivax infections was similar to that reported in areas of low endemic setting such as Central China ( 2–19% ) [79] , but considerably lower than hypo-endemic areas such as Vietnam ( 71 . 4%; [80] ) , Sri Lanka ( 68%; [81] ) , Colombia ( 60–80%; [82] ) , and the Amazon Basin in Brazil ( 50%; [83] ) . Likewise , the polyclonal rate of P . falciparum was comparable to low transmission areas such as Haiti ( 12 . 9% ) [72] and southern China ( 10–23% ) [84] that are approaching elimination phase , but lower than endemic countries in West Africa such as Gambia and Senegal ( 36–50% ) [85 , 86] , East Africa such as Kenya ( 70–90% ) [87] , Papua New Guinea ( 39–45% ) [88] , and Southeast Asia such as Malaysia ( 65% ) [89] and Cambodia ( 47% ) [9] . The higher polyclonality in P . falciparum than in P . vivax may imply a potentially large P . falciparum reservoir present in asymptomatic hosts that emerges during the transmission/rainy season [90] . By contrast , because P . vivax infections can occur periodically throughout the year in Ethiopia , our samples may reflect only a fraction of the existing P . vivax gene pool . Although relapse and early production and circulation of gametocytes in P . vivax infection can lead to increased opportunities for recombination , drug-sensitive clones ( both blood-stage parasites or hypnozoites ) could be eliminated during the non-transmission season by antimalarial treatment and resulted in reduced within-host diversity [8 , 82] . The lack of demographic data and drug use history of our patients limits our exploration of immunity and other factors on the observed MOI . It is important to note that microsatellites have the potential to underestimate polyclonality due to their lower sensitivity and specificity in detecting minority alleles compared to amplicon deep sequencing [91] . Despite the concern of relapse by P . vivax , the asymptomatic P . falciparum and P . vivax reservoirs that remain undetected during non-transmission season or in low endemic areas could pose a long-term impact on local transmission . Ethiopia adopted AL as the first-line treatment for uncomplicated P . falciparum in 2004 in response to increased resistance to CQ and SP [2] . Unlike the Greater Mekong Subregion of SE Asia where delayed ACTs response has been reported , AL is shown to be highly effective in clearing parasite and fever within three days of drug treatment in Ethiopia [92] . The high efficacy of AL concords with our finding of predominantly wild type pfK13 genotypes in P . falciparum from the northern , eastern , and southern populations . Although CQ and SP have not been used for P . falciparum treatment in the last decade , the high prevalence of mutations in pfcrt 76 , pfmdr1 184 , pfdhps 396 , and pfdhfr 51 and 108 across Ethiopia suggested that strong selection may still exist possibly by the use of CQ for treating P . vivax malaria , as well as SP for intermittent preventive treatment ( IPT ) on pregnant women as part of the antimalarial schemes in sub-Saharan Africa [93] . It is concerning that the high prevalence of SP resistance mutations observed in the present study may indeed influence the outcome or effectiveness of IPT [94] . Another explanation for the high frequency of CQ and SP resistance mutations is the spread of resistance parasites from one population to another . Microsatellites indicated a substantial admixture of P . falciparum genotypes between north and east Ethiopia . It is possible that resistant parasites can spread via frequent human movements and become locally selected . The observations of an alarming level of CQ resistance prevalence in Papua New Guinea [95] , India [96] , SE Asia [97 , 98] , and South America [99] after a decade-application deepen the concern for the appearance of CQ resistance to P . vivax in Ethiopia . Recent reports of therapeutic failure of CQ ( 5 . 76–13% ) [3–6] as well as high rates ( 9–32% ) of recurrent infections subsequent to CQ usage in different parts of Ethiopia [6 , 100] have threatened the efficacy of P . vivax treatment . The highly conserved sequences of pvcrt-o as well as the predominantly wild type T958M , F1076L , and single copy of pvmdr1 suggest that these attributes may not be relevant to CQ resistance . By contrast , we detected a high prevalence of pvmdr1 976F mutation , which is considerably higher than that reported in India ( 22% ) where the observed resistance genotypes were confirmed by in vitro drug sensitivity testing [101] . Thus , one possible explanation for the high 976F prevalence in our study area is that this mutation may associate with emerging CQ resistance . Such association merits further clinical observations and/or in vitro testing to confirm its functional significance . Moreover , it is yet unclear whether our P . vivax samples were relapse or recrudescent infections . In Cambodia , 89% of the P . vivax samples that were isolated from patients with recurrent/relapse infections within a 42-day follow-up had pvmdr1 976F mutation [102] . This reiterates the importance of distinguishing recrudescent from relapse infections in order to clarify the implications of the observed mutations and accurately elucidate resistance prevalence of P . vivax . Given that chloroquine monotherapy has been the recommended regimen for P . vivax malaria in Ethiopia for the past decades [2] , selection as well as parasite gene flow may explain the emergence and spread of resistance genotypes across the country . Alternative P . vivax treatment regimes such as ACT or CQ in combination with primaquine are suggested to prolong the efficacy of CQ and prevent/reduce relapse in Ethiopia . In summary , P . falciparum and P . vivax revealed moderate levels of genetic diversity and similar population structure in Ethiopia . Human migrations may promote parasite gene flow while environmental heterogeneity and geographical distance did not appear to be a major gene flow barrier . To effectively reduce malaria burden in Ethiopia , control efforts should focus on seasonal migrant populations . Unconstrained parasite gene flow may partly explain similar patterns of resistance marker prevalence across the country . Our findings are paramount to monitoring the emergence and spread of antimalarial drug resistance and offer evidence-based guidelines to existing treatment regimes . | Sub-Saharan Africa is home to nearly 90% of malaria cases . In Ethiopia , two-thirds of the population lives in areas at risk of malaria infection . Malaria spread via human migrations and emergence of drug resistant parasites are major issues to malaria control in this country . Our study used microsatellite markers to determine gene flow patterns of P . falciparum and P . vivax in different parts of Ethiopia . We found that gene flow occurred across broad geographical distance and that environmental heterogeneity did not appear to influence gene flow . Unconstrained parasite gene flow may partly explain similar patterns of resistance marker prevalence across the country . While no mutation was detected in pfK13 that relates to artemisinin resistance in P . falciparum , over 50% of our P . vivax samples had pvmdr1 976F mutation that may relate to chloroquine resistance . This merits further clinical observations and/or in vitro testing . Our findings heighten the concern of chloroquine resistance for P . vivax malaria after more than a decade-application and suggests alternative treatment regime to alleviate the problem . Broadly , malaria control efforts should focus on seasonal migrant populations to effectively reduce malaria burden in Ethiopia . | [
"Abstract",
"Introduction",
"Materials",
"and",
"methods",
"Results",
"Discussion"
] | [
"parasite",
"groups",
"medicine",
"and",
"health",
"sciences",
"plasmodium",
"drugs",
"population",
"genetics",
"geographical",
"locations",
"microbiology",
"tropical",
"diseases",
"parasitic",
"diseases",
"parasitic",
"protozoans",
"parasitology",
"antimalarials",
"apicomp... | 2017 | Transmission dynamics of co-endemic Plasmodium vivax and P. falciparum in Ethiopia and prevalence of antimalarial resistant genotypes |
Biofilms are dense microbial communities . Although widely distributed and medically important , how biofilm cells interact with one another is poorly understood . Recently , we described a novel process whereby myxobacterial biofilm cells exchange their outer membrane ( OM ) lipoproteins . For the first time we report here the identification of two host proteins , TraAB , required for transfer . These proteins are predicted to localize in the cell envelope; and TraA encodes a distant PA14 lectin-like domain , a cysteine-rich tandem repeat region , and a putative C-terminal protein sorting tag named MYXO-CTERM , while TraB encodes an OmpA-like domain . Importantly , TraAB are required in donors and recipients , suggesting bidirectional transfer . By use of a lipophilic fluorescent dye , we also discovered that OM lipids are exchanged . Similar to lipoproteins , dye transfer requires TraAB function , gliding motility and a structured biofilm . Importantly , OM exchange was found to regulate swarming and development behaviors , suggesting a new role in cell–cell communication . A working model proposes TraA is a cell surface receptor that mediates cell–cell adhesion for OM fusion , in which lipoproteins/lipids are transferred by lateral diffusion . We further hypothesize that cell contact–dependent exchange helps myxobacteria to coordinate their social behaviors .
Biofilms are ubiquitous in nature . Within these structures microbes adhere to surfaces and each other in dense communities coated by an extracellular matrix . Although biofilms are of great medical and industrial interest [1] , little is known about how these cells interact . In some cases , cell-cell contacts likely promote communication and provide spatial cues about neighboring cells to direct biofilm maintenance and maturation [2] , [3] . Experimentally , biofilm research is hindered by limited knowledge and approaches to study their cellular dynamics [4] . Recently we described a novel biofilm dependent process whereby myxobacteria exchange their outer membrane ( OM ) lipoproteins [5] , [6] . This transfer process can result in phenotypic changes and may represent a unique mechanism in which biofilm cells communicate . Although OM lipoprotein exchange is an interesting phenomenon , little is known about the mechanism and protein components required for transfer . Myxobacteria are gram-negative soil dwelling microbes that exhibit complex multicellular behaviors . Central to these behaviors is gliding motility , which powers and coordinates swarm expansion , rippling , predation and fruiting body development on solid surfaces . Myxococcus xanthus has two distinct motility systems called A ( adventurous ) and S ( social ) motility , which served as the experimental backdrop for the discovery of OM lipoprotein exchange [7] , [8] . S-motility is powered by the retraction of type IV pili adhered to external surfaces , effectively pulling the cell forward [9] . The motor powering A-motility is beginning to be defined and may involve cell surface adhesins that translocate on tracks [10] . Nonmotile mutants ( A−S− ) thus typically contain two mutations . Of interest here is a small subset of motility mutants that can be complemented extracellularly when mixed with another strain that encodes the corresponding wild-type gene [8] , [11] . Historically , this process was called ‘stimulation’ as the recipient mutant transiently gains the ability to glide . Stimulation only involves phenotypic changes; there are no genotypic changes . Of the six stimulatable motility genes ( cglB/C/D/E/F and tgl ) [7] , [8] , only two have been previously identified; cglB ( A-motility ) and tgl ( S-motility ) [12] , [13] . Importantly , both of these genes encode type II signal sequences ( SS ) for lipoproteins . The mechanism of stimulation was determined to involve cell-to-cell transfer of either the CglB or Tgl lipoproteins from donor to recipient cells , thus restoring missing protein function to the respective mutant [5] . Strikingly , lipoprotein transfer is efficient as recipient cells accumulate approximately equal quantities of proteins as donors [5] , [6] . Recently , we described the identification of the cglC/D/E/F genes [14] . These genes encode either a type I or type II signal sequence . To determine the molecular mechanism of OM lipoprotein exchange ( stimulation ) we recently defined the cis factor requirement in the cargo protein [6] . Surprisingly , simply a type II signal sequence for OM localization is sufficient for heterologous transfer of the mCherry fluorescent protein . Cytoplasmic or inner membrane reporters were not transferred . Transfer also requires specific cell-cell contacts where motility is apparently required to align biofilm cells [6] , [15] . Here , we sought to identify trans or host genetic determinants required for lipoprotein transfer . In a prior study we screened known S-motility mutants for stimulation defects [16] . This resulted in the identification of a subset of pil mutants that were conditionally defective in tgl stimulation . However , these mutants were not further pursued because they are functional for cgl stimulation , and tgl stimulation occurs when cells are mixed on hard agar at low cell densities . This report identifies two gene products universally required for stimulation and lipoprotein transfer . In addition , we provide evidence , for the first time , that myxobacteria exchange their OM lipids , and that this process can regulate swarming and developmental behaviors .
To elucidate the mechanism of lipoprotein transfer we sought to identify mutants defective in stimulation . We reasoned that cgl and tgl stimulation occurs by a common mechanism , whereby OM proteins , and perhaps periplasmic proteins , are transferred from donor to recipient cells that lack a corresponding protein function . To avoid trivial or idiosyncratic mutants associated with particular cgl or tgl genes , we sought mutants universally defective in stimulation of the six known cgl/tgl complementation groups . We initiated these studies by first characterizing select mutants in the Dale Kaiser strain collection , the laboratory in which A- and S-motility and stimulation were discovered [7] , [8] . One such mutant ( DK396 ) , isolated by Jonathan Hodgkin , appeared to possess the desired phenotype . This strain was isolated by ultraviolet light mutagenesis on an A−S+ ( DK1211 ) strain and then screened for the loss of S-motility ( nonmotile A−S− ) . Serendipitously , this mutant was found to be donor defective for stimulation , a phenotype we verified for all cgl/tgl mutants . As the donor defect mutation was not known nor easily mapped , the DK396 genome was sequenced to identify the gene of interest . Upon >39X sequence coverage the DK396 genome was compared to the wild-type DK1622 genome to identify DNA changes [17] . Mutations in 20 gene candidates were identified ( Table S1 ) . The mutations responsible for the A- and S-motility defects , but not the stimulation defect , were easily found as they were in known motility genes ( Table S1; aglT and pilR , respectively ) [18] , [19] . Based on the severity of the mutations and predictions of gene function and subcellular localization , a prioritized list of 9 gene candidates was chosen . Assuming the phenotype was caused by a loss-of-function mutation , these genes were systematically tested for a role in stimulation by a rapid gene disruption method in a nonmotile donor strain . From these experiments one insertion mutation in mxan_6895 ( hereby named traA for transfer ) was found to recapitulate the donor defective phenotype observed in DK396 . Figure 1 shows that a disruption mutation in traA results in a complete block of stimulation for all the cglB , C , D , E , F and tgl mutants , as indicated by sharp colony edges ( Figure 1 row D ) . The traA+ isogenic control donor stimulates all cgl/tgl mutants for A or S-motility ( Figure 1 row C ) . The degree to which strains were stimulatable varied and only involved partial motility restoration ( Figure 1 compare row C to A ) . From these results it was concluded that traA was universally required for cgl/tgl stimulation . Next we tested whether TraA was required for SSOM-mCherry transfer [6] . This reporter has a type II SS for OM lipoprotein localization fused to a fluorescent protein . In this assay a nonmotile and non-stimulatable SSOM-mCherry donor was mixed with an A-motile GFP+ labeled recipient . The cell mixture was pipetted onto a TPM agarose pad and motile recipients were allowed to swarm . The swarm edge was then examined by epifluorescence microscopy to determine whether SSOM-mCherry was transferred from the nonmotile donor to motile recipients . As shown in Figure 2C controls , SSOM-mCherry was readily detected in GFP labeled recipient flares [6] . In contrast , an isogenic donor that contained the traA::km disruption exhibited no SSOM-mCherry transfer ( Figure 2F ) . To verify these results we conducted related experiments where the same strains were again mixed and spotted on agar , and after short incubations cells were harvested and microscopically examined on glass slides . Here transfer was directly tested by assessing whether GFP labeled recipients become red . As previously reported , control strains show transfer ( Figure 3 left green and red panels , see arrows ) , where typically >90% of recipients obtain detectable levels of SSOM-mCherry [6] . In contrast , when an isogenic traA− donor was used no SSOM-mCherry transfer was detected ( Figure 3 middle merged panel ) . We further note that replication of this experiment; under similar or different conditions/strain backgrounds , where thousands of cells were evaluated , never resulted in detection of SSOM-mCherry transfer from a traA− donor . We conclude that TraA is required for OM lipoprotein transfer and stimulation . The traA ORF and the downstream mxan_6898 ORF ( locus tag numbers are not consecutive ) overlap by four bases , suggesting they form an operon and their gene products may function in the same pathway ( Figure 4A ) . To test this we created an insertion mutation in mxan_6898 . This mutant exhibited a complete block in stimulation for all cgl and tgl mutants and was completely defective in SSOM-mCherry transfer ( Figures S1 and S2 ) . In addition , markerless in-frame deletions in traA and mxan_6898 were constructed and found to elicit the identical phenotypes reported here . Therefore mxan_6898 was named traB and its gene product is predicted to function in the same pathway as TraA . The mxan_6894 ORF is located 126 bps upstream of traA , suggesting it is not part of the traAB operon . To test for a possible role in stimulation/transfer an insertion mutation was again created . In contrast to traA and traB , the mxan_6894::km mutant showed no overt defect in simulation or SSOM-mCherry transfer . To test whether the stimulation/transfer defect of DK396 was solely caused by the traA mutation , the selectable mxan_6894::km mutation and the tightly linked traA+ allele were transduced into DK396 . All resulting Kmr transductants were fully competent for stimulation , thus the traA mutation in DK396 caused the stimulation/transfer defects found in this strain . Since the mutation in DK396 was a missense substitution ( Table S1; 227P→L ) , we tested whether it caused a dominant-negative phenotype by complementation analysis . Here , the wild-type traAB genes were cloned into a plasmid that directs site specific recombination into the Mx8 phage attachment site . Integration of this plasmid into the DK396 genome restored stimulation to the resulting strain , thus demonstrating the traA227P→L allele was recessive . In addition , this plasmid , which has traAB under the heterologous transcription control of the strong pilA promoter , was introduced into a tra+ strain that contains the SSOM-mCherry reporter . Strikingly , upon microscopic examination this TraAB overexpressing strain was found to dramatically cause cells to adhere to one another in both kinked end-to-end chains and side-by-side contacts ( Figure S3 ) . The implication of this observation is discussed below . Next , we tested whether TraAB plays a role in recipient cells for stimulation/transfer . Since traA and traB mutants are fully motile , one or more of these mutations were introduced into all the cgl/tgl mutants . Importantly , when recipient cells contain a traA or traB mutation and mixed with a Tra+ donor , no stimulation occurred ( Figure 1 row E and Figure S2 ) . We conclude that TraAB are required in both donor and recipient cells for stimulation . Next , defects in protein transfer were tested . As described above , when a nonmotile ( traA+ ) donor was mixed with a motile traA− recipient , SSOM-mCherry was not transferred ( Figure 2G–2I and Figure 3 right column ) . We conclude that TraAB are required in donor and recipients for stimulation and lipoprotein transfer . The finding that OM lipoproteins are efficiently and apparently non-specifically transferred suggests that OM lipids may also be exchanged . To test this , donor cells were stained with a fluorescent lipophilic dye called DiD oil . As shown , DiD specifically stained the cell envelope , which fluoresced red ( Figure S8 ) . Importantly , when stained cells were harvested , washed and mixed with GFP labeled recipients in solution , recipients did not fluoresce red , indicating the dye did not freely diffuse between cells . As transfer requires a hard surface , cell-cell contact and motility , we next tested , under these conditions , for DiD transfer [6] . As shown in Figure 6 ( left panels ) , DiD transfer readily occurred to GFP-labeled recipients . As controls , no DiD transfer occurred when isogenic recipients contained a traA mutation or when donor and recipients were both nonmotile ( Figure 6 , middle and right panels , respectively ) . In accordance with the above results , TraA was also required in donors , and similarly TraB in donors/recipients , for DiD transfer ( Figure S9 ) . These experiments show , similar to SSOM-mCherry transfer ( Figure 2 and Figure 3 ) [6] , that lipophilic dye and hence OM lipid , requires a hard surface , cell motility , and TraAB functions in donor and recipient cells for transfer . As noted above , the traA and traB mutants exhibited no overt defects in A or S-motility , suggesting that OM transfer was not required for motor functions . However , the exchange of OM lipids and proteins involves significant resource sharing between cells and therefore this process must involve physiological consequences . One such phenotypic consequence was the restoration of swarming defects to certain motility mutants ( Figure 1 ) . However , extracellular complementation might have little significance between wild-type cells as they contain a full complement of motility proteins . In strain-mixing experiments we discovered that tra+ , but not tra− strains , dramatically inhibited swarm expansion when a nonmotile strain was mixed with a motile strain . An example of how a nonmotile strain inhibits swarm expansion of an A+S− strain was illustrated in Figure 7A . In contrast , when identical mixing experiments were done between isogenic traA− strains , swarm expansion occurred ( Figure 7B and 7C ) . As was found for lipoprotein and lipid transfer , the relief of swarm inhibition occurred when the traA mutation was introduced into either the nonmotile or motile strains . However , we note , swarm expansion was consistently more robust when the motile strain , instead of the nonmotile strain , contained the traA mutation ( compare Figure 7B to 7C ) . An identical relief of swarm inhibition was again found when strains instead contained the traB mutation . Similarly , a Tra+ dependence for swarm inhibition of A+S+ motility was found when these strains were instead mixed with a nonmotile strain . In contrast , inhibition of A−S+ motility was minimal . To test whether swarm inhibition was specific to certain motility genes we test a variety of A−S− double mutants , including combinations of dsp/dif , pilA , pilM , pilT , pilQ , stk , aglB , aglR and aglM mutations , and in all cases these nonmotile strains inhibited swarm expansion of A+S− motile strains . We conclude that swarm inhibition was not dependent on specific motility genes , but instead was dependent on TraAB and thus OM exchange . Macroscopically swarm inhibition was apparent ( Figure 7A , 4 day incubation ) ; however swarm inhibition was not absolute as flares were initially observed emerging from inoculation mixtures ( Figure 7D , 15 hrs ) . Microscopically , the number and size of these early emerging flares were reduced compared to traA− mixtures ( Figure 7 , compare 7D to 7E and 7F ) . However , over longer incubations , e . g . 4 days , the strain mixtures that were Tra+ failed to swarm farther ( Figure 7 , compare 7A to 7B and 7C ) . To investigate this behavior time-lapse microscopy was used to track cell movements . Consistent with the above observations , for the first ≥1 day after plating the A-motile cells exhibited similar cell movements with respect to speed , reversal frequency and percent of cells moving , whether the mixtures contained tra+ or tra− cells . In contrast , by day 2 these same cell mixtures exhibited drastically different behaviors . That is mixtures containing traA+ cells exhibited a complete block in group movements , while isolated cells occasionally exhibited motility that was aberrant ( Video S1 ) . In sharp contrast , isogenic strain mixtures with traA− mutations in either the motile or nonmotile strain exhibited robust group and single cell motility ( Videos S2 and S3 ) . Swarm inhibition does not appear to depend on a diffusible signal , because when these identical tra+ strains were separated by a membrane ( nitrocellulose ) or soft agar overlay , no motility inhibition was observed . Hence , we hypothesize that nonmotile cells produce a time dependent ( ≥2 days ) physiological signal that was transferred by OM exchange to motile cells that blocked their motility . Myxobacteria are noted for the social behaviors and ability to form multicellular fruiting bodies in response to starvation . We thus tested whether Tra plays a role in development . A traA mutation was introduced into a wild-type strain , but no overt defects in fruiting body formation or sporulation was observed . To extend the above swarm inhibition findings , we next tested whether genetically distinct strain mixtures , as found in nature [33] , interfered with development in a Tra dependent manner . First , the traA mutation did not significantly alter the ability of A+S− strain to sporulate ( Figure 8 ) [34] . Second , as development is coupled to motility [35] , nonmotile strains cannot fruit or sporulate and a traA mutation does not alter this phenotype ( Figure 8 ) . Strikingly , however , when the A+S− strain was mixed in a 1∶1 ratio with a nonmotile strain no viable spores were detected ( ≥6-logs; Figure 8 ) . In contrast , when isogenic strains contained the traA mutation in either strain , the ability of the A-motile strain to sporulate was restored to control levels ( Figure 8 ) . Thus similar to swarm inhibition , a nonmotile strain can block development of a motile strain that depends on TraA and hence OM exchange .
To understand the mechanism of lipoprotein exchange we identified mutants universally defective in cgl/tgl stimulation and protein transfer . Interestingly , these TraAB proteins were required in both donor and recipient cells . To our knowledge , this is the first bacterial transfer system where the same gene products are required in both donor and recipient cells . This finding and the ability of M . xanthus cells to rapidly and homogeneously exchange lipoproteins [5] , [6] implies that lipoproteins are transferred in a bidirectional manner . A bidirectional transfer mechanism is distinct from known secretion and conjugative systems [36] , [37] , where proteins or DNA are transferred unidirectionally from donor to recipient cells . Since OM lipoprotein exchange occurs efficiently and involves a form of bulk transfer [5] , [6] , we hypothesized that OM lipids may also be exchanged . This hypothesis was supported by the finding that a lipophilic fluorescent dye was readily exchanged between cells . Importantly , transfer of lipophilic dye and hence membrane lipids , have the same stringent requirements in transfer as OM lipoproteins [6] . That is , dye transfer only occurred when cells were motile within structured biofilms; no detectable dye transfer occurred in liquid or between nonmotile ( non-stimulatable ) cells on a solid surface . In addition , dye transfer required the TraAB proteins in donor and recipient cells . We thus conclude that dye exchange does not occur by diffusion or by diffusible OM vesicles , but instead requires specific cell-cell contacts mediated by cell motility . Based on earlier observations that OM , but not IM , lipoproteins are transferred [6] , we surmise that only OM lipids are exchanged bidirectionally . Presumably transfer consists of the outer leaflet lipopolysaccharide ( LPS ) and the inner leaflet phospholipids . In this respect it is interesting to note that species of Borrelia have been directly observed to fuse their OMs , a process apparently mediated by cell motility [38] , and Bacillus subtilis reportedly transfers proteins in biofilms via membrane enclosed nanotubes [39] . Based on sequence , domain architecture and functional similarities to eukaryotic proteins , we propose that TraA serves as a cell surface receptor . In particular , TraA has similarities to the Saccharomyces cerevisiae FLO1 and FLO5 cell surface receptors/adhesions [21] , [24] , [40] ( Figure 4 ) . These FLO proteins have domain architecture consisting of a SS , N-terminal PA14 domain , a central tandem repeat region and a C-terminal protein sorting tag ( GPI site; glycosylphosphatidylinositol anchor ) for cell surface attachment [26] . Thus , by analogy , we suggest that in TraA the SS serves to transport the protein to the periplasm followed by SS cleavage . The processed N-terminal PA14 domain would serve as a receptor for ligand binding , presumably a glycan . The cysteine-rich tandem repeats could serve as a rigid stalk for PA14 presentation on the cell surface . The MYXO-CTERM motif could function , analogous to a GPI site , in protein sorting to the cell surface . Recent reports suggest the MYXO-CTERM and related C-terminal tags ( Figure 5 ) are widely distributed in bacteria and archaea , where they are proposed to be posttranslationally modified and direct protein sorting to the cell surface [29]–[31] . Although initial attempts to generate TraA antibodies or fluorescent protein fusions were unsuccessful , TraAB overexpression was found to dramatically increase the ability of cells to adhere to one another ( Figure S3 ) . This result is consistent with TraA serving as a cell surface adhesin . Furthermore , the identification of the traA227P→L missense mutation within PA14 highlights the importance of this domain for function ( Figure 4 ) . We also note that Dictyostelium discoideum , a eukaryotic soil slime mold that exhibits similar multicellular behaviors as M . xanthus [41] , produces two secreted signals , called DicA1 ( PsiF ) and PsiA , whose proteins contain PA14 domains followed by cysteine-rich repeats ( Pfam00526 ) of various lengths that show some resemblance to TIGR04201 [21] , [27] , [42] , [43] . Thus , M . xanthus and other microbes , including eukaryotes , appear to utilize PA14 encoding proteins as extracellular signaling and recognition molecules to mediate social interactions . Recent bioinformatic analysis suggests gram-negative bacteria encode C-terminal protein sorting tags that function analogously to the well-characterized gram-positive LPXTG/sortase system [29] . In the case of MYXO-CTERM , we postulate that this motif forms a transmembrane α-helix and anchors pre-TraA into the IM [29] , [31] . Here the Arg rich C-terminal tail would reside in the cytoplasm , while the remainder of the protein would be in the membrane or periplasm ( Figure 5 ) . Thus analogous to lipoprotein processing [44] , an acyl transferase could attach a lipid moiety via a thioether bond to the invariant Cys ( Figure 5 and Figure S7 ) . Subsequently , an endoprotease would cleave the TIGR03901 motif downstream of the aforementioned Cys residue . Once processed a system analogous to the Lol pathway could transport these proteins to the cell surface . As the traB gene overlaps in a bicistronic operon with traA ( Figure 4A ) and mutations in each gene elicit identical phenotypes , suggests that TraAB likely function in the same transport pathway . Since the C-terminal region of TraB contains an OmpA-like domain ( Pfam00691 ) , it likely binds non-covalently to the cell wall . The N-terminal region constitutes the majority of this protein ( ∼400 amino acids ) and has no ascribed function ( Figure 4 ) , but theoretically could interact with the OM and even traverse the OM to interact with TraA . It is also plausible that TraB may facilitate TraA's localization to the cell surface . A working model for the mechanism of cell contact-dependent exchange is outlined in Figure 9 . First , cell-cell recognition is postulated to be mediated by TraA serving as a cell surface receptor . We suggest that the distant PA14 domain may function in ligand binding to neighboring cell surfaces . Glycans found in LPS or glycoproteins are possible ligands . In a variation of this model TraA may function as a homophilic receptor . Similar to the FLO1 system , a key component of this model involves reciprocal TraA binding by both cells . A ‘donor’ cell was arbitrarily assigned and its OM ( mCherry ) lipoproteins were symbolized as red lollipops . Upon aligned cell-cell contact and docking the OM membranes of adjoining cells fuse . Although not directly depicted , TraAB may facilitate membrane fusion by bringing OMs into close proximity and perhaps causing local membrane perturbations that help catalyze OM fusion . Membrane fusion may also be facilitated at cell poles where the membranes have high tip curvatures and thus are more fusogenic [45] . Once cells are adhered cell motility could also stress the membrane . Upon OM fusion , lipids and lipoproteins rapidly exchange bidirectionally; a process presumably driven by lateral diffusion . Integral and associated OM proteins are also likely transferred as the CglE and CglF proteins encode type I signal sequences [14] . It is unknown whether soluble periplasmic proteins are transferred . Prior studies clearly indicate inner membrane lipoproteins and cytoplasmic proteins are not transferred [6] . Following fusion cells physically separate , a process likely facilitated by gliding motility . The exchange of OM lipoproteins has phenotypic consequences to the cell , including complementation of mutational defects ( Figure 1 ) . Whether the restoration of mutation defects is ecologically important is unknown; however population heterogeneity within biofilms , especially from an environmental setting are significant [4] , and consequently some individuals within a population are less fit . Thus , we hypothesize that the ability to exchange and share the OM proteome allows some individuals to gain fitness and for the population to establish OM homeostasis . In turn , homeostasis may increase population fitness by normalizing intercellular signal output and reception by reducing population heterogeneity . Thus community behaviors , such as swarming and development might be better coordinated . In this respect , our findings that a mixture of nonmotile cells with motile cells inhibits the latter cells from swarming in a TraAB and time dependent manner ( Figure 7 ) , suggests these cells are communicating and coordinating their behaviors via OM exchange . Similarly , OM exchange can regulate development behaviors between genetically distinct strains ( Figure 8 ) . The use of strain mixtures to study cell-cell interactions in motility and development is ecologically relevant , as diverse M . xanthus isolates are found in close proximity in nature [33] , [46] . The mechanism for developmental inhibition by nonmotile cells on motile cells is unknown , but may simply reflect a block in motility ( Figure 7 ) [47] . Alternatively or in addition , OM exchange with nonmotile cells may transmit a signal that blocks development . Currently , we are investigating the nature of these putative signals . Our results indicate that myxobacteria exchange and thus share a significant amount of their cellular resources . This has led us to hypothesize that cell contact-dependent OM exchange represents a form of cooperative social behavior that may involve kin recognition . A kin recognition mechanism avoids the theoretical and ecologically relevant concern that ‘cheater’ cells could exploit or disrupt this social behavior to obtain resources [48] . This problem is highlighted by observations that environmental M . xanthus populations arise from diverse origins [33] , [46] . Thus unlike artificial laboratory settings where multicellular behaviors are typically studied with a single homogenous culture , natural myxobacteria isolates must recognize kin from non-kin cells as they vacillate between single cell and multicellular life . The data presented here provide three lines of evidence that cell contact-dependent OM exchange involves kin recognition . First , TraAB proteins are required in both ‘donors’ and ‘recipients . ’ Thus if one cell does not express TraAB , exchange cannot occur . Second , exchange appears bidirectional , thus both cells are giving and receiving . Therefore , there is no inherent advantage one cell type has over another , unless one cell is starving and has depleted resources . Third , TraA contains a PA14 domain , with features resembling PA14 domains in yeast flocculin proteins involved in kin recognition and social behaviors . More specifically , flo1 and other genes within this group were classified as ‘greenbeard’ genes , which by molecular definition are cell surface receptors that recognize other cells carrying the same gene to provide social preferential treatment [49] , [50] . In the case of FLO1 the protein allows yeast cells to enter the protective domain of a floc , where cells are so tightly joined they become deformed . Within flocs cells are protected from environmental stresses and cheater cells ( flo1− ) cannot enter [40] . In another greenbeard example , the Dictyostelium csA gene , which encodes a homophilic cell surface receptor , plays a discrimination role in partitioning cells to desirable locations within fruiting bodies [51] . Current experiments are testing whether TraA plays such a role .
Bacterial strains and plasmids are listed in Table S2 [52] . M . xanthus was grown at 33°C in CTT medium ( 1% casitone , 1 mM KH2PO4 , 8 mM MgSO4 , 10 mM Tris-HCl , pH 7 . 6 ) in the dark and when necessary supplemented with kanamycin ( Km; 50 µg/ml ) , oxytetracycline ( Tc; 15 µg/ml ) , or streptomycin ( Sm; 600 µg/ml ) . For ½ CTT , casitone was reduced to 0 . 5% . On plates , agar concentration was 1 . 0 or 1 . 2% . TPM buffer contains 10 mM Tris , 1 mM KH2PO4 and 8 mM MgSO4 , pH 7 . 6 . Escherichia coli was grown at 37°C in LB medium and when necessary supplemented with Km ( 50 µg/ml ) , ampicillin ( 100 µg/ml ) or Sm ( 100 µg/ml ) . The DK396 genome was sequenced by using Illumina second generation DNA sequencing technology ( NCGR , Santa Fe , NM ) . Sequence reads were aligned and analyzed for mutations against the wild-type DK1622 reference genome within the Alpheus bioinformatic platform [53] . DNA cloning followed routine protocols [54] . Chromosomal and plasmid DNA was isolated with UltraClean Microbial DNA and Mini Plasmid isolation kits ( MO BIO Laboratories , Inc . ) , respectively , as described by the manufacture . All insertion mutations were created by PCR amplification of internal gene fragments with Taq 2X Master Mix ( New England BioLabs ) followed by direct cloning of products into pCR2 . 1 TOPO ( Invitrogen ) and then transformed into DH5α . To overexpress the traAB operon it was fused downstream of the strong pilA promoter with an optimally designed ribosomal binding site [55] . Specifically , the pilA promoter was amplified with Phusion High-Fidelity PCR Master Mix with HF Buffer ( New England Biolabs ) and cloned into pSWU19 at the EcoRI to XbaI restriction sites [18] . traAB was then similarly amplified and cloned into the XbaI and HindIII sites . Primers are listed in Table S3 . Plasmid constructs were confirmed by restriction digestion analysis or DNA sequencing . Verified plasmids were electroporated into M . xanthus and integrated into the genome by homologous recombination with antibiotic selection [34] . To identify the donor defect mutation from DK396 , insertion mutations were made in DK6204 [56] or DK8601 A−S− donor strains ( Table S1 ) . Mx4 or Mx8 bacteriophages were used for strain construction by generalized transduction [15] . Mutants were verified by phenotypes and molecular methods including PCR and sequencing . M . xanthus strains were grown to a Klett ∼100 ( ∼3×108 cfu ml−1 ) , concentrated by centrifugation and resuspended to a calculate Klett of 1000 in TPM buffer . For stimulation , donors and recipients were mixed at a 1∶1 ratio and 3 µl were pipetted onto ½ CTT 1% agar pads containing 3 mM CaCl2 ( added after autoclaving ) and incubated in a humid chamber for various times . Micrographs were taken with either an Olympus SZX10 stereo microscope ( whole colony ) or a Nikon E800 phase contrast/fluorescent microscope ( colony edge ) coupled to digital imaging systems . A heterologous fluorescent OM lipoprotein reporter , called SSOM-mCherry , was used to monitor protein transfer in live cells [6] . To clearly differentiate recipients from SSOM-mCherry expressing donors , the former cells expressed the green fluorescent protein ( GFP ) . Thus , in general terms , protein transfer was scored as the ability of green cells to become red . Lipoprotein transfer was microscopically determined by mixing donor and recipients ( 1∶3 or 1∶1 ratios ) and either ( i ) detected as motile recipient flares emerging from inoculum spots with nonmotile donors , or by ( ii ) harvesting cell mixtures and inspecting single cells on glass slides as previously described [6] . To reduce background fluorescence , the former cells were spotted on a thin TPM agarose ( 1% ) pads prepared on a glass slide . A sampler kit ( Invitrogen; cat# L7781 ) containing different lipophilic fluorescent dyes were evaluated for M . xanthus OM staining . According to the manufacture these dyes are not transferred from stained to unstained cells . DiD oil ( component B; DilC18 ( 5 ) oil ) was chosen for further studies where a Texas Red-4040B ( Semrock ) filter set was used to visualize staining . Cells were grown to Klett ∼100 , harvested by centrifugation and resuspended in TPM buffer to a calculated Klett of 250 . To stain cells , 1 µl of dye ( 1 mg/ml , dissolved in ethanol ) was added to 49 µl of cells and incubated for 1 to 2 hrs in the dark at 33°C . Cells were then pelleted by centrifugation , washed with 1 ml TPM and microscopically examined ( 100× objective ) . Similar to monitoring SSOM-mCherry transfer , dye transfer was also assayed by mixing stained donors with GFP labeled recipients ( 1∶1 ratio ) and spotted on a ½ CTT 1% agar . After 4 hrs incubation , cells were scraped from the agar surface , washed 2× in 1 ml TPM , placed on glass slide with cover-slip and inspected whether green cells also stained red . Log phase M . xanthus cultures were concentrated by centrifugation to a calculated Klett of 1000 and pipetted onto TPM starvation agar ( four 25 µl spots ) and incubated for 5 days at 33°C . Cells and spores were harvested and placed into a tube with 500 µl of TPM buffer , heated at 50°C for 2 hrs and then gently pulse sonicated to disperse spores . Spore suspensions were serial diluted and 10 µl samples spotted on CTT agar . After 7 days of incubation , viable spores were enumerated as CFUs . All developmental assays were done in triplicate and averaged . PA14 domain analysis and alignments are described in Results and Figure S4 . The cysteine-rich repeat of MXAN_6895 was identified by inspection . TIGRFAMs model TIGR04201 was developed by multiple sequence alignment of several repeats , HMM construction , search against a large collection of proteins from prokaryotic reference genomes , and iteratively refined . In proteins identified by TIGR04201 as having at least one copy of the repeat , additional , lower-scoring repeats are confirmed by manual inspection of HMM search results . Completed HMMs were added to the TIGRFAMs database , which uses the HMMER 3 . 0 software package [57] . A search was undertaken for candidate protein-sorting domains with architectural elements similar to the LPXTG-containing recognition sequence of sortase A [28] , the PEP-CTERM putative recognition sequence of exosortase and the PGF-CTERM putative recognition sequence of archaeosortase A [29] . The common architecture was; signature motif , hydrophobic predicted transmembrane helix , cluster of basic residues , positioned at the extreme C-terminus and found in protein regions lacking other homologies . A general purpose classifier , TIGRFAMs [58] HMM TIGR03901 , was constructed to model a candidate protein-sorting signal domain approximately thirty-three residues long , with an invariant Cys residue in its signature motif , universal in but restricted to the eight species of Myxococcales among 1460 prokaryotic reference genomes; scoring thresholds give no false-positive in any species . To identify atypically low-scoring instances of the domain in M . xanthus , a species-specific HMM was derived from TIGR03901 by HMM search , inspection of results , realignment , and repetition of the search through several iterations . Extensive biocuration of the similar but shorter GlyGly-CTERM motif found primarily in gammaproteobacteria , modeled by HMM TIGR03501 , improved the disambiguation of GlyGly-CTERM ( which does not occur in M . xanthus ) from MYXO-CTERM . Approximate atomic coordinates for the PA14Tra structure was automatically generated from the alignment of known PA14 domains ( Figure S4 , residues 62 to 259 ) . This was done by using a standard two-step template-based modeling protocol . The initial 3-D model was obtained using MODELLER 9 . 9 , with 2XJP ( FLO5 ) as template structure ( default parameters , best-scoring of 20 models ) [59] . To produce the final model , side-chain atoms were refined using SCWRL4 [60] . | All cells interact with their environment , including other cells , to elicit cellular responses . Cell–cell interactions between eukaryotic cells are widely appreciated as large multicellular organisms coordinate cell behaviors for tissue and organ functions . In bacteria cell–cell interactions are not widely appreciated , as these organisms are relatively simple and are often depicted as single-cell entities . However , over the past decade , the concept of bacteria living in microbial communities or biofilms has received broad acceptance as a major lifestyle . As biofilm cells are packed in tight physical contact , there is an opportunity for cell–cell signaling to provide spatial and physiological clues of neighboring cells to elicit cellular responses . Although much has been learned about diffusible signals through quorum sensing , little is known about cell contact–dependent signaling in bacteria . In this report we describe a new mechanism where bacterial cells within structured biofilms form contacts that allow cellular material to be exchanged . This exchange elicits phenotypic changes , including in cell movements and development . We hypothesize that OM exchange involves kin recognition that bestows social benefits to myxobacterial populations . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"bacteriology",
"prokaryotic",
"models",
"model",
"organisms",
"molecular",
"cell",
"biology",
"genetics",
"molecular",
"genetics",
"biology",
"microbiology",
"genetics",
"and",
"genomics"
] | 2012 | Cell Contact–Dependent Outer Membrane Exchange in Myxobacteria: Genetic Determinants and Mechanism |
Kaposi's Sarcoma ( KS ) , the most common tumor of AIDS patients , is a highly vascularized tumor supporting large amounts of angiogenesis . The main cell type of KS tumors is the spindle cell , a cell of endothelial origin , the primary cell type involved in angiogenesis . Kaposi's Sarcoma-associated herpesvirus ( KSHV ) is the etiologic agent of KS and is likely involved in both tumor formation and the induction of angiogenesis . Integrins , and specifically integrin αVβ3 , have known roles in both tumor induction and angiogenesis . αVβ3 is also important for KSHV infection as it has been shown to be involved in KSHV entry into cells . We found that during latent infection of endothelial cells KSHV induces the expression of integrin β3 leading to increased surface levels of αVβ3 . Signaling molecules downstream of integrins , including FAK and Src , are activated during viral latency . Integrin activation by KSHV is necessary for the KSHV-associated upregulation of a number of angiogenic phenotypes during latent infection including adhesion and motility . Additionally , KSHV-infected cells become more reliant on αVβ3 for capillary like formation in three dimensional culture . KSHV induction of integrin β3 , leading to induction of angiogenic and cancer cell phenotypes during latency , is likely to be important for KS tumor formation and potentially provides a novel target for treating KS tumors .
Kaposi's sarcoma-associated herpesvirus ( KSHV ) , a gamma herpesvirus , is the etiological agent for Kaposi's sarcoma ( KS ) . KS is the most common tumor in AIDS patients world-wide , and is the most commonly reported tumor in parts of central Africa [1] , [2] . KS tumors are highly vascularized , with abnormal , leaky vasculature , and excess inflammation and edema . The histopathology of KS tumors supports a role for angiogenesis in tumor formation . The primary cell type of KS lesions are spindle-shaped endothelium-derived cells aptly named spindle cells . Nearly all spindle cells support latent KSHV infection , although a low percentage of cells undergoing lytic reactivation are always present [3] . KSHV can infect many types of cells in culture including endothelial cells [4] , [5] . KSHV infection of endothelial cells in culture leads to predominantly latent infection with a similar low percentage of cells undergoing lytic replication as in the KS tumor [4] , [6] . KSHV infection of endothelial cells can promote angiogenesis related phenotypes , including increased stability of tubules formed by macrovascular endothelial cells , induction of angiogenesis and capillary morphogenesis in low growth factor conditions , and enhanced migration and invasion [7]–[11] . Furthermore , KSHV infection can induce increased expression and secretion of signaling factors involved in angiogenesis , such as vascular endothelial growth factor ( VEGF ) . Both VEGF-A and –C are expressed by KSHV-infected endothelial cells [12] , [13] . Interestingly , KSHV infection promotes the upregulation of both VEGF receptor 1 , a blood vasculature marker , and VEGF receptor 3 , a marker for lymphatic endothelium [13]–[17] . The upregulation of both VEGF receptors suggests KSHV-infected cells are more sensitive to the growth and migratory effects of VEGF than the surrounding uninfected endothelium . KSHV infection also leads to upregulation of other molecules with important roles in the regulation of angiogenesis . KSHV-induced expression of cyclooxygenase-2 ( COX-2 ) as well as angiogenin was shown to be important for the maintenance of latency , as well as inflammatory cytokine expression and capillary morphogenesis [18] . KSHV infection of endothelial cells upregulates several members of the angiopoietin family of growth factors , including angiopoietin-2 and angiopoietin-like 4 , which are involved in regulating angiogenic remodeling and vessel stabilization [19]-[21] . In addition to secretion of growth factors , KSHV infection promotes disruption of adherens junctions , allowing for increased vascular permeability and invasion [22]–[25] . Furthermore , there have been several studies examining the role of other molecules and signaling pathways which have been implicated in KSHV-induced angiogenesis [26]–[34] . Several KSHV genes have been shown to regulate expression of genes involved in angiogenesis . For example , both vIRF3 and the glycoprotein K1 promote VEGF expression [35] , [36] . Other genes were found to have angiogenic chemoattractant properties , such as the viral homologs to macrophage inflammatory proteins ( vMIPs I–III; [37] , [38] ) . The viral G-protein coupled receptor ( vGPCR ) is a constitutively active signaling receptor that has been linked to a variety of angiogenic signaling pathways [21] , [39]–[44] . However , vGPCR and the other genes mentioned are primarily expressed during lytic replication and they have not been shown to be necessary for the induction of angiogenic phenotypes in the context of viral infection . Since only a small percentage of the infected cells undergo lytic replication , it is unknown if these proteins are sufficient to promote an angiogenic phenotype in the more abundant latently infected endothelial cells . During latency in endothelial cells , only five viral genes are expressed , including the latency-associated nuclear antigen ( LANA-1 ) . LANA-1 has been shown to play a role in a host of processes including maintenance of the viral genome and host cell survival ( reviewed in [45] ) . In addition to these functions , LANA-1 may play several roles in promoting angiogenesis . Expression of LANA-1 in endothelial cells induces upregulation of angiogenin , which may aid in induction of angiogenesis by VEGF and basic fibroblast growth factor [10] . Furthermore , LANA-1 interaction with Daxx reduces its repression on Ets-1-dependent VEGF receptor expression [46] . While it is apparent that KSHV genes are capable of inducing angiogenic pathways , a better understanding of how viral infection alters the host cell to induce angiogenesis is still needed . Angiogenesis is a complex process that is tightly regulated by a delicate balance of pro- and anti-angiogenic factors . However , many pathogenic processes , such as tumor formation , shift this balance to promote continual vascular growth . How the different angiogenic signaling factors interact and are regulated is only partly understood . Along with secreted cytokines and growth factors , other signaling proteins such as integrins have been shown to regulate endothelial cell activation and angiogenesis . Integrins are cell surface proteins that link the extracellular matrix to the cytoskeleton . They form dimers of α and β subunits that recognize extracellular matrix ( ECM ) proteins and , upon ligand binding , undergo conformational changes and recruit intracellular adaptor and signaling molecules . Integrins can induce activation of focal adhesion kinase ( FAK ) leading to activation of Src kinase . Integrin signaling complexes play a role in a number of cellular processes such as adhesion to the ECM , migration , and cell survival during suspension , all of which are essential functions for endothelial cells during the process of angiogenesis ( reviewed in [47]–[49] ) . In particular , αVβ3 integrin has been shown to be upregulated in the vasculature associated with a number of tumor types and has been linked to regulation of angiogenesis [50]–[52] . Antagonists of integrin αVβ3 inhibit tumor angiogenesis and tumor growth in a variety of animal models of cancer [52]–[55] . Interestingly , αVβ3 integrin is a receptor for several viruses , including KSHV [56] . KSHV glycoprotein B ( gB ) can associate with integrins on the cell surface leading to increased signaling , adhesion , and to cytoskeletal rearrangements . However , previous studies have not examined the role of integrins during latent KSHV infection . We have identified a role for integrin β3 during latent infection of KSHV in endothelial cells . Infected cells upregulate integrin β3 expression leading to increased cell surface αVβ3 . Latently infected endothelial cells become more adherent to integrin ligands fibronectin and vitronectin , and are also more migratory than mock-infected cells . These induced phenotypes require RGD-binding integrins , specifically integrin β3 . Interestingly , although both uninfected and infected cells organize in three-dimensional culture , infected cells are more sensitive to inhibitors of integrin β3 and its downstream signaling molecules , such as Src kinase . This suggests that during latent KSHV infection there is a shift in endothelial cell signaling that results in a more angiogenic phenotype dependent on αVβ3 .
Our microarray analysis of mock- and KSHV-infected endothelial cells indicated that integrin β3 was significantly upregulated in all our KSHV-infected samples ( data not shown ) . To confirm this we utilized quantitative real-time RT-PCR to measure integrin β3 mRNA levels during KSHV latent infection of endothelial cells . TIME or primary human dermal microvascular endothelial cells were mock- or KSHV-infected and allowed to establish latent infection for 48 hours . RNA from these cells was extracted and subjected to real-time RT-PCR using primers specific to integrin β3 . Figure 1A shows that KSHV-infected TIME cells had a 4 . 6-fold increase in integrin β3 expression as compared to their mock counterparts . In primary endothelial cells ( ECs ) , KSHV induced a 3 . 5-fold increase in integrin β3 expression over mock infection . In the cultures analyzed , greater than 90% of the infected cells expressed LANA and less than 1% of the cells expressed ORF 59 , a marker of lytic infection , indicating that the increase is likely to be a result of latent infection . We next examined the expression of integrin β3 at the protein level , using western blot analysis and flow cytometry . In TIME cells latently infected with KSHV , total cellular integrin β3 was significantly increased as visualized by Western blot analysis with an antibody specific to integrin β3 ( Figure 1B ) . In endothelial cells αV is the only integrin α subunit that dimerizes with the β3 subunit . Flow cytometric analysis showed a specific increase in cell surface expression of αVβ3 during latent KSHV infection of TIME cells ( Figure 1C ) . The mean fluorescence intensity of mock-infected cells was approximately 22 . 3 , while KSHV infection increased the mean to 31 . 9 . Although there is only a modest increase in cell surface αVβ3 protein expression as compared to total cellular expression , this shift was highly reproducible upon multiple infections with different stocks of virus . Importantly , expression of αVβ3 increased in the majority of cells , indicating that the increased surface protein expression was a property of latently infected cells , not only the low percentage of lytically infected cells . Activation of integrins leads to recruitment of the intracellular protein focal adhesion kinase ( FAK ) and autophosphorylation of FAK on tyrosine 397 [47] . This autophosphorylation provides a binding site for src family kinases , which subsequently become activated . To determine if KSHV promotes integrin signaling through FAK and Src , we analyzed the phosphorylation state of these proteins at 48 hours post infection , after the establishment of latency . Figure 2A shows an increase in FAK phosphorylation on tyrosine 397 in KSHV-infected cells , compared to mock-infected cells . Furthermore , Src is also phosphorylated more heavily on tyrosine 416 in KSHV-infected cells . These experiments were also performed in primary HMVECs , which upon infection undergo a small but reproducible increase in Fak and Src activation ( Figure 2B ) . These data suggest that KSHV latent infection promotes signaling through FAK and Src . A key process during cell migration is the regulation of focal adhesion complexes , groups of signaling molecules that mediate integrin signaling to the actin cytoskeleton . In order to better understand the effect of KSHV induction of integrin β3 on endothelial cell morphology , we examined the formation of focal adhesion complexes in mock- or KSHV-infected cells . Figure 3A shows the localization of the focal adhesion component vinculin in mock-infected cells , which have many centrally localized focal adhesions . The cells were also stained with phalloidin to identify polymerized actin which is present throughout the cells . In contrast , at 48 hours post-infection , endothelial cells latently infected with KSHV have fewer focal adhesions but they are strongly localized to the periphery , suggesting changes in the formation and turnover of the focal adhesions ( Figure 3C ) . The polymerized actin also localizes to the periphery of the cell during latent infection . The fewer focal adhesions and their localization at the periphery suggest higher turnover and increased migration in KSHV-infected cells . To eliminate the potential effects of virus binding and entry , UV-inactivated virus was used to infect cells . UV-irradiated virus can bind and enter cells but there is no viral gene expression , as determined by immunofluorescence with anti-LANA antibodies . UV-inactivated virus did not alter the localization of focal adhesions at 48 hours post-infection ( Figure 3B ) . Interestingly , in a mixed population of infected and uninfected cells , only infected cells ( as determined by staining with LANA-1 antibody ) had peripheral localization of focal adhesions . Uninfected cells in the same population had more centrally localized focal adhesions , similar to mock-infected cells ( data not shown ) . The cells were similarly stained with phalloidin and antibodies to integrin β3 . Integrins are known components of focal adhesion complexes and have previously been shown to localize with vinculin in these complexes [47] , [48] , [57] . As expected , integrin β3 had a very similar pattern of relocalization to the periphery as was seen with antibodies to vinculin ( Figure 3D–F ) . These data suggest that latent KSHV infection promotes formation and turnover of focal adhesions and that integrin αVβ3 is activated and localized to the newly formed focal adhesions . These observations are visible throughout the culture , as can be seen in Figure 3G–I , where multiple cells are shown stained for the latent antigen LANA ( red ) and the actin cytoskeleton ( green ) . Latently infected cells ( Figure 3I , arrows ) show peripheral organization of actin filaments , compared to the neighboring uninfected cells in the same culture ( Figure 3I , arrow heads ) , which have actin organization similar to mock-infected or UV-inactivated virus ( Figure 3G–H ) . Furthermore , peripheral focal adhesions can be detected in greater than 90% of the highly infected cells . In contrast , ORF59 is detected in less than 1% of the cells ( data not shown ) . In our endothelial infections the KSHV late lytic protein K8 . 1 , is detected in an even lower percentage of the cells than ORF59 and we could only detect gB in a very low percentage of cells as well ( data not shown ) . Importantly , gB has previously been shown to be detected only in the low percentage of cells undergoing lytic replication [58] . Together , these data indicate that the vast majority of cells with peripheral focal adhesions are latently infected and , while lytic replication could play some paracrine role , it is unlikely to play a primary role in organization of focal adhesion complexes during KSHV infection as uninfected cells would be altered as well in this scenario . We next wanted to determine whether KSHV latent infection promotes an angiogenic phenotype in endothelial cells through the activation of αVβ3 . Adhesion and migration of endothelial cells are both critical for angiogenesis . Figure 4 shows the adhesion of mock- or KSHV-infected cells to various concentrations of the extracellular matrix proteins fibronectin and vitronectin , both of which contain the arginine-glycine-aspartic acid ( RGD ) motif that is recognized by a subset of integrins , including αVβ3 . Latent infection by KSHV enhanced the adhesion of endothelial cells to both fibronectin ( Figure 4A ) and vitronectin ( Figure 4B ) compared to mock-infected cells . None of the cells adhered to either collagen I , which lacks an RGD motif , or laminin , suggesting the increased adhesion is specific to fibronectin and vitronectin ( data not shown ) . Short RGD-containing peptides will compete with vitronectin and fibronectin binding to RGD-dependent integrins , like αVβ3 , and inhibit integrin-dependent adherence to these substrates . Interestingly , an RGD-containing short peptide , but not a control RAD- or RGE-containing peptide , inhibited KSHV-induced adhesion to fibronectin or vitronectin . The same peptides had little effect on mock-infected cells ( Figure 4C–D ) . This indicates that KSHV-infected cells require RGD-binding integrins for increased adherence while mock-infected cells are relatively unaffected by the concentrations of RGD peptides used . In addition to adhesion , integrin activation can promote migration of endothelial cells through formation and turnover of focal adhesions . In Figure 3 we demonstrated a strong reorganization of focal adhesions during KSHV latency in endothelial cells . We therefore examined the effect of KSHV infection on endothelial cell migration . 48 hours post-infection , endothelial cells were seeded on a transwell membrane with 8 µm pore size and allowed to migrate for 1 . 5 hours . Approximately three times more KSHV-infected endothelial cells migrated through the pores as compared to mock-infected cells ( Figure 5A–C ) . This increase in migration was not due to increased proliferation , since KSHV infection does not promote proliferation in TIME cells as determined by uptake of BrdU ( data not shown ) . To determine whether this enhanced migration was due to increased integrin signaling , we used a Src kinase inhibitor to block downstream integrin signaling . The Src kinase inhibitor led to a decrease in the mean number of mock- and KSHV-infected cells that migrated through the pores ( Figure 5D ) . However , the lower dose of the Src kinase inhibitor decreased the migration of mock-infected cells by approximately 50% while migration of KSHV-infected cells was reduced more than 80% ( Figure 5E ) . Therefore , KSHV-infected cell migration is sensitive to Src inhibitors to a greater degree than migration of mock-infected cells . KSHV latent infection leads to small increases in the ability of endothelial cells to organize into capillary-like structures in three-dimensional culture at 6 hours after seeding . However , these effects are not consistently statistically significant ( our unpublished observations and Figure 6A ) . Due to the increased integrin activity and signaling , we hypothesized that KSHV infection may promote capillary morphogenesis through a different pathway than uninfected cells . Therefore , we examined the role of integrin β3 in the ability of KSHV-infected cells to organize in three-dimensional culture on Matrigel . We first tested whether RGD-binding integrins were involved by using an RGD-containing peptide to block integrin activity . Interestingly , in the presence of the RGD-containing peptide , KSHV-infected cells formed approximately 85% fewer capillary-like structures while mock-infected cells lost only 40% of their activity ( Figure 6A and B ) . This indicates that KSHV-infected cells are more sensitive to inhibition of integrin activation as compared to mock-infected cells and suggests that KSHV-infected cells become reliant on integrin activity for this angiogenic phenotype . In order to determine if this effect was due specifically to integrin β3 we used siRNA to knock down integrin β3 protein expression . Cells were transfected with siRNA 24 hours post-infection with KSHV and harvested for analysis 48 hours later ( at 72 hours post-infection ) . Figure 6 ( C and D ) shows that the expression of αVβ3 integrin complexes is significantly reduced upon transfection of integrin β3 siRNA . Interestingly , αVβ3 expression is reduced more strongly in mock-infected cells ( Figure 6C ) than in KSHV-infected cells ( Figure 6D ) . Mock-infected cells with control siRNA have a geometric mean fluorescence of 22 . 3 , which drops to 3 . 4 in cells expressing integrin β3 siRNA , a greater than 6 . 5 fold decrease in expression . In contrast , in KSHV-infected cells , the geometric mean fluorescence of integrin β3 goes from 31 . 9 in control siRNA-expressing cells to 6 . 9 in β3 siRNA-expressing cells , which is still above background fluorescence levels and is a 4 . 6 fold decrease . This is likely due to the increased integrin β3 transcript levels in KSHV-infected cells . Integrin β3 specific siRNA impaired KSHV-induced capillary morphogenesis ( Figure 6E and F , p-value = 0 . 003 ) but had little effect on mock-infected cells ( a 33% versus 12% decrease in capillary formation respectively ) , despite the small amount of residual integrin β3 expression in KSHV-infected cells with the β3 specific siRNA . This suggests that the KSHV-induced angiogenic phenotype specifically requires integrin β3 expression . To further confirm that integrin activation and signaling during KSHV latent infection is required for the KSHV-induced angiogenic phenotype , we next determined whether Src kinase activity is required for KSHV-induced capillary morphogenesis . Mock- and KSHV-infected cells were treated with the indicated concentrations of Src kinase inhibitor SKI-1 or DMSO alone and plated on Matrigel matrix . Similar to the effects of inhibition by the RGD-containing peptide or integrin β3 siRNA , inhibition of Src kinase had a greater impact on KSHV-infected cells than on mock-infected cells ( Figure 6G and H ) . KSHV-infected cells treated with 5 nM SKI-1 had an approximate 5-fold decrease in the number of capillaries formed compared to DMSO-treated cells . In contrast , mock-infected cells retained nearly complete ability to organize on Matrigel . At higher concentrations of SKI-1 , capillary formation of KSHV-infected endothelial cells was completely eliminated while mock-infected cells had only modest decreases in capillary formation . To further demonstrate that the effects seen were due to inhibition of Src , another Src kinase inhibitor , su6656 , was used . Su6656 had a similar effect on KSHV-infected cells as SKI-1 ( Figure 6G and H ) . Similar experiments with primary HMVECS yielded comparable results ( data not shown ) . Thus , three different methods of inhibiting integrin signaling – RGD peptides , siRNA to integrin β3 , and chemical inhibition of downstream integrin signaling – blocked capillary formation by KSHV infected endothelial cells in a three dimensional matrix more significantly than in mock infected cells .
During angiogenesis , activated endothelial cells disrupt the extracellular matrix , proliferate and migrate towards angiogenic stimuli , gain traction into new areas through increased adhesion to specific substrates , and organize into preliminary vasculature . These activities are necessary for vascularization of solid tumors and contribute to the promotion of tumor formation and metastasis . As has been described by others , KSHV is likely to promote angiogenesis in a paracrine fashion to promote new blood vessel growth in the KS tumor [10] , [12] , [19]–[21] , [28] , [41] , [59] . Here we found that KSHV also directly induces angiogenic phenotypes in latently infected cells . KSHV latent infection of endothelial cells increased cell motility and cell adhesion . Therefore , angiogenic phenotypes induced in latently infected endothelial cells may promote angiogenesis in KS tumors but may also directly be involved in the activation of the endothelial tumor cells themselves . Integrins are involved in many processes of both angiogenesis and oncogenesis . Integrins have been shown to be critical for angiogenic phenotypes such as adhesion , migration , and anchorage-independent survival ( reviewed in [47] , [48] ) . Integrins have also been shown to be important in oncogenesis of many cell types ( reviewed in [60] ) . In particular , integrin αVβ3 is expressed on blood vessels in human tumor biopsy samples , but not on vessels in normal human tissues , suggesting a role in tumor-associated angiogenesis [50] , [51] . Furthermore , inhibition of integrins , and in particular αVβ3 , can inhibit angiogenesis and promote regression of tumors by inducing apoptosis of endothelial cells [52]–[55] . We found that KSHV not only induces the expression of integrin β3 leading to increased surface expression of αVβ3 , but latent infection also leads to the activation of αVβ3 and downstream activation of FAK and Src . Importantly , integrin signaling , specifically αVβ3 , is necessary for KSHV induction of angiogenic phenotypes including cell motility and adhesion . Regulation of angiogenesis is a complex process that involves cross talk between multiple signaling pathways . Integrin signaling complexes have been shown to interact with angiogenic growth factor receptors , such as VEGF and basic fibroblast growth factor [48] . Specifically , interaction between αVβ3 and VEGF receptor 2 can lead to phosphorylation of the cytoplasmic tail of β3 as well as increased phosphorylation of VEGF receptor 2 in a Src-dependent fashion [61] . Interestingly , β3 knockout mice exhibit increased pathological angiogenesis , due to increased expression and signaling through VEGF receptor 2 [62] , [63] . However , knock-in expression of a mutant β3 that lacks phosphorylation sites blocks VEGF-induced pathological angiogenesis in these mice [64] . Integrin αVβ3 also plays a role in regulating the expression of VEGF; its activation can increase VEGF secretion by tumor cells [65] . Thus , upregulation of VEGF and its receptors by KSHV likely acts in concert with increased αVβ3 signaling to promote angiogenesis . Interestingly , KSHV-infected cells become dependent upon αVβ3 for capillary formation . Inhibition of integrin signaling and αVβ3 had small effects on the ability of uninfected endothelial cells to form capillary-like structures on Matrigel . However , during KSHV latency , capillary formation is extremely sensitive to inhibitors of αVβ3 and integrin signaling . In uninfected cells αVβ3 is localized with focal adhesions all over the cell , while in infected cells αVβ3 is localized to the focal adhesions at the periphery . It is possible that the dramatic change in focal adhesions and polymerized actin in the infected cells leads to increased reliance on αVβ3 . However , whether the increased sensitivity is due to a switch in the infected cell solely to integrin signaling for capillary formation or if there is a decrease in a compensatory pathway is unknown . In either event , this provides a good target for therapy , as the latently infected cells are more sensitive than mock-infected cells to inhibitors of integrin signaling . Integrins are also used for KSHV binding and entry into cells . It was first demonstrated that α3β1 was a receptor for KSHV on endothelial cells [66] . Subsequently it was shown that gB binds to αVβ3 and that αVβ3 may be the dominant integrin for binding and entry [56] . Interestingly , both of these integrins together with CD98 may form an entry complex [67] . During binding and entry , gB binding to integrins leads to activation of the integrins and subsequent activation of FAK and focal adhesions [68]–[70] . While this phenomenon is similar to what we describe , we found that activation of FAK and Src also occurs during latency and , at the time points we examined , this is not due to initial binding and entry of the virus . Additionally , gB has previously been shown to only be expressed in cells undergoing lytic replication and cannot be detected at the late times post-infection that we examined [58] . While cells undergoing lytic replication are present in our cultures , less than 1% of the cells express ORF 59 , a common marker of lytic infection . Using an immunofluorescence assay we find that most of the infected cells have altered focal adhesions at the periphery . Furthermore , the bulk of the cells shift expression of αVβ3 and the bulk of the population is more motile and more adherent , indicating that the phenomenon we are examining is occurring in latently infected cells . Importantly , in cultures where we have a mixed population of uninfected and KSHV-infected cells , only the cells that have LANA expression , i . e . are latently infected , have relocalized actin to the periphery ( Figure 3I ) and have focal adhesions at the periphery as well ( not shown ) . Taken together , these data all indicate that there is a factor in the latent cells that is necessary for activation of integrin signaling and induction of the angiogenic phenotypes . While paracrine factors from the low percentage of lytically infected cells could play some role in the angiogenic phenotypes described , a latent factor is still required . The viral gene or genes upregulating integrin β3 and activating αVβ3 and FAK are currently unknown and will be the focus of future studies . Inhibition of integrins and integrin signaling has been proposed as a target for tumor therapy through inhibition of tumor cell growth itself and through inhibition of neo-angiogenesis . A number of inhibitors of integrin signaling are in clinical trials for tumors that have increased levels of integrins and specifically αVβ3 ( reviewed in [49] , [60] ) . Importantly , an inhibitor of αVβ3 is in phase3 trials for inhibition of tumor formation . Based on the studies presented here , we believe that αVβ3 may be a good target for therapy to treat KS tumors . It is difficult to target herpesvirus latency due to the limited viral gene expression . However , through a better understanding of the host cell requirements for latency new therapeutic targets can be identified . Integrin β3 induction and activation by KSHV during latency is critical for a number of phenotypes important for tumor growth and latency , making it an attractive target for therapy .
Primary human dermal microvascular endothelial cells ( hDMVEC ) and TIME cells [71] were maintained as monolayer cultures in EBM-2 medium ( Lonza ) supplemented with 5% fetal bovine serum , vascular endothelial growth factor , basic fibroblast growth factor , insulin-like growth factor 3 , epidermal growth factor , and hydrocortisone ( EGM-2 media ) . BCBL-1 [72] and BJAB cells [73] were maintained in RPMI 1640 medium ( Celgro; Mediatech , Inc . ) supplemented with 10% fetal bovine serum , penicillin , streptomycin , glutamine , and β-mercaptoethanol . KSHV inoculum was obtained from BCBL-1 cells ( 5×105 cells/ml ) induced with 20 ng of TPA ( 12-O-tetradecanoylphorbol-13-acetate; Sigma ) /ml . After 5 days , cells were pelleted , and the supernatant was run through a 0 . 45-µm-pore-size filter ( Whatman ) . Virions were pelleted at 30 , 000xg for 2 h in a JA-14 rotor , Avanti-J-25 centrifuge ( Beckman Coulter ) . The viral pellet was resuspended in EGM-2 without supplements . KSHV infections of TIME and primary hDMVEC were performed in serum-free EBM-2 supplemented with 8 ug/ml polybrene for 3 h , after which the medium was replaced with complete EGM-2 . Mock infections were performed identically except that concentrated virus was omitted from the inoculum . For all experiments , infection rates were assessed by immunofluorescence using antibodies against the latency-associated nuclear antigen ( LANA ) and the lytic protein ORF59 . In all infections performed with wild-type ( wt ) KSHV >85% of the cells were LANA-positive and <1% were ORF59-positive . UV inactivation of KSHV viral stocks ( 5×1 , 200 µJ ) was performed in a UV Stratalinker 1800 ( Stratagene ) . Total RNA was isolated from TIME cells using the RNeasy Plus Minikit ( Qiagen ) . One hundred or 500 ng of total RNA was used in a SuperScript III , Platinum SYBR green , one-step , quantitative reverse transcription PCR ( RT-PCR; Invitrogen ) according to manufacturer's protocols with the primers for either GAPDH ( glyceraldehyde-3-phosphate dehydrogenase ) ( forward , 5′-AAG GTG AAG GTC GGA GTC AAC G-3′; reverse , 5′-TGG AAG ATG GTG ATG GGA TTT C-3′ ) or integrin β3 ( forward , 5′-GCA AGG ATG CAG TGA ATT GT-3′; reverse , 5′-CTT GGG ACA CTC TGG CTC TT-3′ ) . Relative abundances of integrin β3 mRNA were normalized by the delta threshold cycle method to the abundance of GAPDH , with mock-infected TIME cells set to 1 . Error bars reflect standard errors of the means ( four experiments ) . Cells were harvested with a cell scraper and pelleted using the Sorvall RT7 Plus centrifuge at 4°C . An aliquot of the cells was seeded onto chamber slides for immunofluorescence analysis . Cell pellets were washed once in cold phosphate-buffered saline and then resuspended in lysis buffer ( 20 mM Tris [pH 7 . 0] , 2 mM EGTA , 5 mM EDTA , phosphatase inhibitors , protease inhibitors , and 1% Triton X-100 ) . Samples were sonicated , rocked at 4°C for 30 minutes , and then spun at 6 , 000 x g at 4°C . Cell extracts were fractionated on a sodium dodecyl sulfate-polyacrylamide gel electrophoresis gel , and the proteins were transferred electrophoretically to Immobilon P polyvinylidene difluoride membranes ( Millipore ) in Tris-glycine buffer ( 25 mM Tris , 192 mM glycine , 20% methanol ) . Blots were incubated with the indicated antibody ( dilutions: 1∶1000 for anti-integrin β3 , 1∶1000 for anti-phosphoFAK and anti-phosphoSrc; 1∶20 , 000 for anti-actin ) and subsequently with horseradish peroxidase-conjugated goat anti-mouse or rabbit immunoglobulin G ( 1∶10 , 000; IgG; Jackson ImmunoResearch ) . Immunoreactive proteins were visualized by chemiluminescence using the Amersham ECL Plus Western blotting detection reagents ( GE Healthcare ) . Differences in band intensity were quantified by densitometric methods . Mock- or KSHV-infected TIME cells were seeded on LabTek Permanox four-well chamber slides ( Fisher Scientific ) and fixed with 4% ( wt/vol ) paraformaldehyde in phosphate-buffered saline . Immunofluorescence was performed as described previously [4] . Briefly , cells were incubated in Tris-Buffered Saline ( 20 mM Tris , 150 mM NaCl , pH 7 . 6; TBS ) containing 1% normal goat serum followed by incubation with primary antisera at a dilution of 1∶100 ( anti-integrin β3; anti-vinculin ) or 1∶1 , 000 ( rabbit or rat anti-LANA; mouse anti-ORF59 ) diluted in TBS containing 1% BSA overnight . Cells were then incubated with fluor-conjugated secondary antibodies ( goat anti-rabbit Alexa Fluor 488 , goat anti-mouse Alexa Fluor 594 , or goat anti-rat Alexa Fluor 488; Molecular Probes/Invitrogen; Fluorescein-conjugated phalloidin; Sigma ) for 2 hours . Cells were mounted in medium containing DAPI ( 4′ , 6′-diamidino-2-phenylindole ) before being viewed under a Zeiss LSM 510 Meta confocal microscope . Images were analyzed using the Zeiss Zen 2009 LE imaging software . siRNA specific to integrin β3 and negative-control oligonucleotides were designed and synthesized by Ambion ( Austin , TX ) . The following oligonucleotide sequences were used: integrin β3 ( Ambion identification [ID] no . 112581; sense , 5′-GCU AAU UCU UUG ACC UGU UdTdT-3′ ) and negative-control siRNA ( sense , 5′-AGU ACU GCU UAC GAU ACG GdTdT-3′ ) . At 24 h post-infection , mock- or KSHV-infected TIME cells were transfected with 3 µg siRNA using Amaxa's Nucleofector kit ( Cologne , Germany ) according to the manufacturer's protocol . The transfection efficiency for siRNA was approximately 90% when it was assessed with 6-carboxyfluorescein-labeled negative-control siRNA . Transfected cells were harvested for analysis after an additional 2 days of incubation at 37°C . Monolayers of cells grown in 60-mm tissue culture dishes were washed once with phosphate-buffered saline ( PBS ) containing 0 . 04% EDTA and incubated with 2 ml of cell dissociation solution ( Sigma , St . Louis , MO ) to remove cells from plates . Cells were washed once with DMEM containing 10% FBS and fixed in 4% paraformaldehyde in TBS for 10 minutes on ice . Cells were then pelleted and resuspended in 0 . 5 ml TBS with 1% goat serum for 20 min on ice . Cells were pelleted and incubated with mouse anti-αVβ3 ( 2 µg/ml; LM609; Millipore ) or mouse IGG control prepared in 0 . 5 ml of TBS with 1% BSA for 30 min on ice . Following incubation , cells were washed twice with TBS/1% BSA and then incubated with anti-mouse secondary antibody conjugated to Alexafluor 488 ( Invitrogen; diluted 1∶200 in 0 . 5 ml of TBS containing 1% BSA ) for 30 min on ice . The stained cells were washed twice with TBS containing 1% BSA , resuspended in 0 . 5 ml of TBS containing 1% BSA , and analyzed by FACScan caliber flow cytometer ( Becton-Dickinson , Franklin Lakes , NJ ) . Data was analyzed using FloJo flow cytometry analysis software ( Tree Star , Inc . ; Ashland , OR ) . Cell adhesion assays were performed using Nunc 96-well Maxisorp plates as described previously [74] . Briefly , wells were coated with different concentrations of collagen , fibronectin , laminin , or vitronectin in TBS containing 2 mM each of CaCl2 and MgCl2 overnight at 4°C . The next day , plates were washed with TBS and blocked with 200 µl of TBS Ca/Mg containing 1% BSA for 1 hour at room temperature . Cells were resuspended in 20 mM HEPES , 150 mM NaCl , 4 mg/ml BSA ( pH 7 . 4 ) and plated at 5×104 cells/well . Cells were allowed to adhere at 37°C in a humidified incubator for 1 . 5 hours . Nonadherent cells were gently washed off with TBS containing Ca/Mg , and the number of adherent cells was determined by measuring the standard intracellular acid phosphatase activity . For inhibition studies on fibronectin , cells were treated with GRADSPK or GRGDSPK peptides ( Sigma ) in HEPES buffer for 15 minutes on ice prior to plating . For inhibition studies on vitronectin , cells were treated with GRGES or GRGDS peptides ( Peptides International ) in HEPES buffer for 15 minutes on ice prior to plating . For transwell assays , transwell filters ( Costar 3422 ) were placed in a 24-well dish and coated on the bottom side with 0 . 5 ml of 2 µg/ml of fibronectin ( BD Biosciences ) in PBS overnight at 4°C . The filter was rinsed with PBS and then blocked with 0 . 5 ml of 2% bovine serum albumin ( BSA ) prepared in PBS for 1 h at room temperature . Following blocking , the filter was rinsed with PBS , 0 . 5 ml of serum-free DMEM medium was added to the bottom of each well . Cells were resuspended at 1×105 cells/ml in serum free medium and 0 . 1 ml was added to the top of each well . Each condition was done in duplicate . Following 1 . 5 hours in a 37°C tissue culture incubator , the cells and medium were aspirated and the upper side of the membrane wiped with a cotton swab . The cells that had migrated through the membrane and attached to the bottom of the filter were fixed with 4% paraformaldehyde and stained with crystal violet . The mean number of cells migrated through the filter was determined by counting ten high power fields ( ×100 ) . For inhibition studies , cells were incubated with SKI-1 ( Calbiochem ) for 15 minutes on ice prior to addition to the transwell . Matrigel ( 10 mg/ml; BD Biosciences , Bedford , MA ) was applied at 0 . 5 ml/35 mm in a tissue culture dish and incubated at 37°C for at least 30 min to harden . Mock- or KSHV-infected cells were removed using trypsin-EDTA , washed with growth medium once , and resuspended at 1 . 5×105 cells per ml in growth medium . Cells ( 1 ml ) were gently added to the Matrigel-coated plates , incubated at 37°C , monitored for 6 h , and photographed in digital format using a Nikon microscope . Capillaries were defined as cellular processes connecting two bodies of cells . Ten fields of cells were counted for each condition and the mean and standard deviations were determined . For inhibition studies , GRADSPK or GRGDSPK peptides , SKI-1 ( Calbiochem ) , or su6656 ( Sigma ) were added at the time of plating on Matrigel . Statistical differences between groups were evaluated with Student's t-test ( two-tailed ) . Mean±SD was shown and a p-value of ≤0 . 05 was considered significant and indicated by asterisk . | Kaposi's Sarcoma ( KS ) is the most common tumor of AIDS patients world-wide and is characterized by very high vascularization . The main KS tumor cell type is the spindle cell , a cell of endothelial origin . Kaposi's Sarcoma-associated herpesvirus ( KSHV ) , the etiologic agent of KS , is found predominantly in the latent state in spindle cells . In this study we examined how KSHV alters endothelial cells to induce phenotypes common to angiogenesis and tumor formation . Integrins are cell surface adhesion and signaling proteins that can be involved in tumor growth and tumor angiogenesis . We found that KSHV infection of endothelial cells leads to increased expression of integrin β3 , a molecule that , when paired with its cognate α subunit , αV , has been shown to be critical for tumor-associated angiogenesis . KSHV infection promotes angiogenic phenotypes in endothelial cells including adhesion , motility and capillary morphogenesis , and these phenotypes require expression and signaling through integrin β3 . Therefore , KSHV induction of integrin beta3 and downstream signaling is required for the induction of phenotypes thought to be critical for KS tumor formation . αVβ3 inhibitors are in clinical trials for inhibition of tumors and we propose that these inhibitors may be clinically relevant for treatment of KS tumors . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"medicine",
"viruses",
"and",
"cancer",
"endothelial",
"cells",
"cancer",
"treatment",
"microbiology",
"basic",
"cancer",
"research",
"cancers",
"and",
"neoplasms",
"cardiovascular",
"oncology",
"integrins",
"tumor",
"physiology",
"cell",
"movement",
"signaling",
"signa... | 2011 | Latent KSHV Infection of Endothelial Cells Induces Integrin Beta3 to Activate Angiogenic Phenotypes |
An unexpectedly high seroprevalence and pathogenic potential of human parvovirus B19 ( B19V ) have been observed in certain malaria-endemic countries in parallel with local use of chloroquine ( CQ ) as first-line treatment for malaria . The aims of this study were to assess the effect of CQ and other common antimalarial drugs on B19V infection in vitro and the possible epidemiological consequences for children from Papua New Guinea ( PNG ) . Viral RNA , DNA and proteins were analyzed in different cell types following infection with B19V in the presence of a range of antimalarial drugs . Relationships between B19V infection status , prior 4-aminoquinoline use and anemia were assessed in 200 PNG children <10 years of age participating in a case-control study of severe infections . In CQ-treated cells , the synthesis of viral RNA , DNA and proteins was significantly higher and occurred earlier than in control cells . CQ facilitates B19V infection by minimizing intracellular degradation of incoming particles . Only amodiaquine amongst other antimalarial drugs had a similar effect . B19V IgM seropositivity was more frequent in 111 children with severe anemia ( hemoglobin <50 g/L ) than in 89 healthy controls ( 15 . 3% vs 3 . 4%; P = 0 . 008 ) . In children who were either B19V IgM or PCR positive , 4-aminoquinoline use was associated with a significantly lower admission hemoglobin concentration . Our data strongly suggest that 4-aminoquinoline drugs and their metabolites exacerbate B19V-associated anemia by promoting B19V replication . Consideration should be given for choosing a non-4-aminoquinoline drug to partner artemisinin compounds in combination antimalarial therapy .
Human parvovirus B19 ( B19V ) is a nonenveloped icosahedral virus with a single-stranded DNA genome which has been classified within the Erythrovirus genus of the Parvoviridae family . The virus is readily transmitted via the respiratory route and has a worldwide distribution . Seroprevalence increases with age and 50%–80% of adults have detectable B19-specific antibody . Since its discovery in 1975 [1] , B19V has been associated with an expanding range of clinical disorders that reflect the patient's immunologic and hematologic status . In healthy individuals , B19V typically causes a mild childhood febrile illness known as erythema infectiosum or fifth disease . More severe manifestations of B19V infection are arthropathies , aplastic anemia , hydrops fetalis and fetal death [2] . Viremia occurs during the first week of infection . The virus has a predilection for bone marrow erythroid progenitor cells . At the height of viremia , there is an abrupt fall in the reticulocyte count and anemia can supervene . Although this is rarely apparent in healthy patients , it can have serious clinical consequences where there is pre-existing anemia [2] . A case in point is malaria . B19V co-infection has been considered a significant risk for severe anemia in children living in malaria-endemic regions [3] . Studies examining the inter-relationship between malaria , B19V infection and anemia have , however , produced inconsistent results . In a retrospective study from Papua New Guinea ( PNG ) [4] , 60% of children <2 years of age and 90% of 6 year-olds were B19V seropositive . B19V infection was significantly associated with severe anemia even in the absence of risk factors including malaria . Similar results were obtained in children from the Republic of Niger [5] . However , studies from Malawi and Kenya did not show evidence that B19V infection contributes to anemia in children and the seroprevalence of B19V was relatively low [6] , [7] . At the time the studies were performed in PNG [4] and Niger [5] , chloroquine ( CQ ) was used as first-line treatment for malaria in these countries . In addition , a serosurvey performed in Eritrea at a time when CQ was the first-line agent also revealed an unusually high B19V seroprevalence [8] . By contrast , CQ had been discontinued in Malawi because of resistance of local strains of Plasmodium falciparum and its use was declining in Kenya when the B19V seroprevalence studies were conducted [6] , [7] . Although CQ has , in addition to antimalarial efficacy , broad antiviral activity [9] , [10] , it also enhances Semliki Forest virus ( SFV ) and encephalomyocarditis virus ( EMCV ) infection in mice [11] , and Epstein-Barr virus ( EBV ) expression [12] . The latter effect is thought to play a role in the higher incidence of EBV-induced Burkitt's lymphoma in malaria-endemic areas where CQ is in common use [13] . We hypothesize , therefore , that geo-epidemiological differences in B19V seroprevalence and pathogenic potential result from CQ-associated enhanced replication . To test this hypothesis , we examined the effect of CQ and other commonly-used antimalarial drugs on B19V replication in three different cultured cell lines . In addition , we examined the relationship between B19V infection and use of 4-aminoquinoline drugs in a sample of children from PNG who were hospitalized with severe anemia . The results provide evidence that CQ and AQ aggravate B19V-associated anemia by promoting B19V replication .
All patients were participants in a prospective observational and genetic study of severe pediatric infections ( http://www . malariagen . net/home/ ) . Written informed consent was obtained from each parent/guardian . Ethical approval for the study was obtained from both the PNG Institute of Medical Research Institutional Review Board and the Medical Research Advisory Committee of the PNG Department of Health . The study was conducted in accordance with the Helsinki Declaration . A B19V-infected plasma sample was obtained from our donation center ( Genotype 1; CSL Behring AG , Charlotte , NC ) and was concentrated by ultracentrifugation through 20% ( w/v ) sucrose . UT7/Epo cells were cultured in RPMI , 10% FCS and 2 U/ml of recombinant human erythropoietin ( Epo; Janssen-Cilag , Midrand , South Africa ) . HepG2 cells were cultured in MEM supplemented with 10% FCS . Bone marrow mononuclear cells ( BMMCs ) were obtained as frozen stocks from Stemcell Technologies ( Vancouver , BC , Canada ) and were cultured in IMDM , 10% FCS and 2 U/ml of Epo . All cells were incubated at 37°C and in an atmosphere of 7 . 5% CO2 . All drugs were purchased from Sigma ( St . Louis , Miss ) . Chloroquine diphosphate ( CQ ) , primaquine diphosphate ( PQ ) and amodiaquine dihydrochloride dihydrate ( AQ ) were dissolved in water , piperaquine ( PPQ ) in 5% lactic acid , mefloquine hydrochloride ( MQ ) in DMSO , lumefantrine ( LFT ) in dimethylformamide and artesunate ( AT ) and pyrimethamine ( PM ) in ethanol . The final drug concentration ranges used in the B19V infectivity assay were 5–100 µM for PQ and LFT , AQ and PM , 0 . 05–20 µM for MQ , 2 . 5–100 µM for AT , 1–100 µM for PPQ and 0 . 05–100 µM for CQ . The highest concentration of the drugs did not exceed 0 . 2% of total culture volume . B19V infectivity was assessed in two different cell lines , namely megakaryoblastic leukemia UT7/Epo cells and the human hepatocellular liver carcinoma cell line HepG2 , and in primary bone marrow mononuclear cells ( BMMCs ) . UT7/Epo cells are the most susceptible cell line to B19V infection [14] . Although viral RNA , DNA and proteins can be detected in B19V-infected UT7/Epo cells , viral replication is restricted to a level that does not normally allow production of virus progeny . The HepG2 cell line was chosen because it allows virus binding and probably internalization , but it is non-permissive for B19V infection [15] . BMMCs have been shown to support B19V infection , although only a minor subset of these cells is permissive for B19V [16] . UT7/Epo , HepG2 and bone marrow mononuclear cells ( BMMCs ) ( 3×105 ) were infected with 20 , 000 DNA-containing B19V particles per cell ( 5 , 000 for BMMCs ) , corresponding to an MOI of approximately 20 ( 5 for BMMCs ) in the presence of the pre-determined concentrations of the selected drugs . For viral RNA and DNA analysis , cells were collected at different post-infection times as indicated in the figure legends . Total poly ( A ) + mRNA was isolated and viral NS1 mRNA quantified as previously described [17] . Total DNA was extracted and viral DNA was quantified using established methods [17] . UT7/Epo cells were infected as specified above in the presence of 0 or 25 µM CQ . At increasing post-infection times ( see figure 1C ) , cells were lysed in protein loading buffer and total proteins were resolved by sodium dodecyl sulfate ( SDS ) -10% polyacrylamide gel electrophoresis ( PAGE ) . After transfer to a PVDF membrane , the blot was probed with a mouse antibody against B19V structural proteins ( 1∶2 , 000 dilution; US Biologicals , Swampscott , MA ) , followed by a horseradish peroxidase-conjugated secondary antibody ( 1∶20 , 000 dilution ) . The viral structural proteins were visualized with a chemiluminescence system ( Pierce , Rockford , IL ) . Additionally , viral protein expression was examined by immunofluorescence , as previously described [17] . UT7/Epo cells ( 3×105 ) were infected with 20 , 000 viral particles per cell ( MOI 20 ) at 4°C for 2 h . The cells were washed 8 times with PBS to remove unbound virus and incubated at 37°C in the presence or absence of antimalarials . At increasing post-internalization times from 1 to 7 h , the cells were washed 2 times with PBS and the amount of intact viral DNA was quantified as specified above . We studied 111 children <10 years of age with severe anemia ( hemoglobin <50 g/L ) and 89 community-based age and sex-matched healthy control children with a hemoglobin >100 g/L . Those with severe anemia represented a subset ( 15 . 9% ) of all 697 children admitted to Modilon Hospital , Madang Province on the north coast of PNG with any severe illness during the period of study . Modilon Hospital is a referral hospital and the only provincial facility able to manage severely ill children . All such children were given treatment as recommended under PNG national treatment guidelines including intramuscular artemether for malaria infection . The healthy controls were recruited from the same villages as the patients and were slide-negative for malaria . As well as a hemoglobin concentration ( HaemoCue® , Angelholm , Sweden ) at presentation , plasma was assayed for B19V IgM by EIA kit ( Biotrin International ) and , in those with severe anemia , for viral DNA using two specific oligonucleotide primers [4] . We did both tests because viremia starts to decline once specific IgM is produced around day 9 after inoculation , while virus-induced marrow suppression can last for another 2–3 weeks [4] . Thus , although the simultaneous detection of B19V IgM and DNA is strongly indicative of acute infection , we did not want to exclude children with evidence of recent but resolving infection as a contributor to severe anemia . In those who were B19V IgM or PCR positive , plasma was assayed for chloroquine and amodiaquine and their respective active desethyl metabolites using a validated high performance liquid chromatography assay [18] . The assay had a limit of quantitation of 1 µg/L for each analyte .
CQ increased the production of B19V NS1 gene transcription after 24 h incubation . At CQ concentrations ranging from 10 to 75 µM , NS1 RNA increased up to 1 , 170% ( Figure 1A ) . Similarly , kinetic studies in the presence of 25 µM of CQ showed that viral DNA synthesis was more rapid and extensive than in untreated cells ( Figure 1B ) . The expression of structural viral proteins in extracts of infected UT7/Epo cells was also increased in the presence of 25 µM CQ ( Figure 1C ) . Viral protein expression was detectable by 34 h in untreated cells and by 24 h in CQ-treated cells . Immunofluorescence experiments showed , that in the presence of CQ a larger number of cells were infected by B19V ( Fig . 1D ) . In the absence of CQ , only a minor amount of viral DNA synthesis was observed starting at 120 h post-infection . No viral RNA could be detected , confirming the poor permissiveness of this cell line for B19V infection . However , in the presence of increasing concentrations of CQ , viral DNA synthesis increased progressively reaching 2 , 290% at CQ concentrations of 60 µM ( Figure 2A ) . Kinetic studies showed that , in the presence of CQ ( 25 µM ) , viral DNA was detected earlier than in untreated cells ( Figure 2B ) . Viral NS1 RNA was only detectable in CQ-treated cells ( Figure 2C ) . The presence of CQ ( 25 µM ) accelerated B19V RNA synthesis . However , in CQ-treated cells , viral RNA transcription ceased abruptly and was followed by progressive degradation resembling apoptosis ( Figure 2D ) . Detection of phosphatidyl serine–anexin V complexes by fluorescence microscopy confirmed that the infected BMMCs entered the apoptotic pathway ( data not shown ) . Therefore , the effect of CQ in cultured BMMCs could not be evaluated at stages later that 10 h post-infection . The apoptotic effects were not observed in the cell lines UT7/Epo and HepG2 at concentrations up to 60 µM ( data not shown ) . With the exception of the CQ-analogue AQ which enhanced B19V infection at concentrations above 5 µM , no other antimalarial drug had a significant effect on B19V infection ( Figure 3 ) . Mild inhibition was observed in the presence of AT , while MQ inhibited the infection at concentrations >10 µM . These effects were also observed when the drugs were added 4 to 7 h post-infection ( data not shown ) , raising the possibility that B19V infectivity was reduced by a drug-specific cytotoxicity . The enhancement of B19V infection by CQ decreased progressively with increases in the time at which CQ was added , with no detectable effect at 8–9 h post-infection ( Figure 4A ) . These data indicate that CQ acts early in B19V infection . At progressive times after internalization of B19V , the cells were washed and the viral DNA was quantified . In untreated cells , a progressive degradation of the incoming viral DNA was evident . However , in the presence of CQ ( 25 µM ) or AQ ( 20 µM ) , degradation of incoming particles was prevented or minimized ( Figure 4B ) . Serological screening revealed that 3 of 89 healthy control children ( 3 . 4% ) and 18 of the 111 with severe anemia ( 16 . 2% ) were IgM positive ( P = 0 . 004 by Fisher's exact test ) . A further 6 B19V IgM-negative children with severe anemia were positive by PCR and 5 of the 18 IgM-positive children were also PCR positive . In 22 of the 24 IgM and/or PCR positive children with severe anemia and plasma available for assay , only 5 did not have detectable 4-aminoquinoline or metabolite concentrations . Based on the respective pharmacokinetic profiles [18] , [19] , this suggests that most of these children had been treated with either CQ or AQ within the previous 6 weeks . Hemoglobin concentrations by B19V IgM/PCR and 4-aminoquinoline status are shown in Figure 5 . The lowest concentrations were in the 5 children who were both IgM and PCR positive ( i . e . had acute B19 infections ) . These children had a similar mean age ( 53 vs 57 months ) , body weight ( 15 kg in both groups ) and spleen size ( 5 vs 7 cm ) to those children who were either IgM or PCR positive ( P>0 . 37 by Mann-Whitney U test ) and the percentages with malaria were similar ( 40 . 0 vs 44 . 4%; P = 0 . 63 by Fisher's exact test ) . In patients who were IgM or PCR positive ( indicating a recent but not necessarily acute infection or one which was acute but early in its course ) , 4-aminoquinoline use was associated with a significantly lower admission hemoglobin concentration ( P = 0 . 037 ) . Although the number of patients treated with AQ was small and restricted to children who were IgM positive but PCR negative , they had some of the highest hemoglobin concentrations in this subgroup . This suggests that , consistent with the in vitro data , CQ had a greater suppressive effect on bone marrow than AQ in our patients .
Severe anemia is a common and life-threatening complication of malaria in children living in endemic areas [20] . B19V co-infection has been identified as a major factor in its pathogenesis [3] , but there are significant regional differences in its seroprevalence and resulting clinical impact [4]–[8] despite the fact that B19V infection is a common childhood illness . Because B19V-associated severe anemia appears to parallel local use of CQ as first-line treatment for malaria , we hypothesized that CQ promotes B19V replication and that , as a consequence , it contributes indirectly to severe anemia . Although a more profound clinical study would be necessary , the present results provide already a strong evidence for this hypothesis . Apart from its antimalarial effects , CQ has a wide antiviral activity . One of the most important mechanisms of action against viruses is the alkalinization of the endosomal vesicles . In this way , CQ is active against viruses that require a low pH step for cell entry , such as flavivirus , retrovirus and coronavirus [9] . All parvoviruses studied to date also depend on endosomal acidification for cell entry because it facilitates capsid structural transitions [21] , [22] , and in particular the externalization of the N-terminal region of VP1 which is required for endosomal escape and nuclear targeting [23] . Accordingly , CQ inhibits parvovirus infections . However , we have previously shown that B19V is unique among parvoviruses in that N-VP1 is already externalized on receptor binding [17] and thus not dependent on a low endosomal pH for this critical conformational change . B19V is also unique among parvoviruses for its higher sensitivity to acid degradation [24] . Consequently , CQ-associated alkalinization of endosomal vesicles would be expected to minimize the acidic degradation of incoming B19V particles . Our data confirm that the intracellular degradation of B19V is prevented or minimized in the presence of CQ or AQ . However , other lysosomotropic drugs such as ammonium chloride or bafilomycin A1 , which also raise the endosomal pH , had an inhibitory effect on B19V infection in our in vitro system ( data not shown ) . Therefore , the mechanism underlying the stabilization of B19V by CQ or AQ is likely to extend beyond pH-neutralizing activity to destabilization of endosome/lysosome membranes typically observed in CQ-treated cells . In this way , CQ would facilitate the endosomal escape of B19V before it reaches the degradative lysosomal compartment and increase the number of particles that can target the nuclei for replication . This is of particular importance since nuclear targeting has been identified as a major limiting factor in parvovirus infections [22] , [25] . The plausibility of our in vitro observations as an explanation of epidemiological data depends on the pharmacological properties of CQ , especially tissue concentrations . The in vitro enhancement of B19V infection by CQ was achieved at concentrations ( 10–75 µM ) that were well above those achieved in plasma after therapeutic doses in children ( typically <5 µM ) [18] . However , B19V does not replicate in plasma but in tissues , primarily the bone marrow . CQ concentrations in bone marrow are substantially higher than in plasma [26] and have been measured at approximately 100 µM in animal studies [27] . This could reflect , in part , concentration of the drug within precursor cells such as has been observed in circulating erythrocytes [28] . The long terminal elimination half-life of CQ ( around 10 days in children ) [18] means that conditions favorable to B19V viral replication in bone marrow may persist for several weeks after dosing . In many malaria-endemic regions , antimalarial therapy is given empirically to febrile children without blood smear confirmation . Ironically this might include fever due to B19V itself . The administration of frequent courses of CQ may mean that a child spends long periods of each year at risk of CQ-associated enhanced B19V viremia and its consequences such as anemia . AQ is a long half-life 4-aminoquinoline compound like CQ and also promoted B19V replication in our in vitro experiments . Other drugs tested , including primaquine ( an 8-aminoquinoline ) and mefloquine ( a methanol quinoline ) , did not influence B19V infection in vitro , suggesting that the effect is specific to 4-aminoquinoline compounds . Some other viral infections ( SFV , EMCV and EBV ) [11] , [12] are enhanced by CQ but not by other 4-aminoquinoline antimalarial drugs . We were able to obtain preliminary human data that are consistent with our laboratory findings . In our 200 unselected PNG children who were participants in a case-control study of severe pediatric infections , we confirmed previous reports that B19V seropositivity is associated with severe anemia and that the lowest hemoglobin concentrations are in those children who had acute infections ( i . e . both IgM and PCR positive ) [4] . Although there were limited numbers , there was some evidence that prior 4-aminoquinoline , especially CQ , use exacerbates B19V-associated severe anemia apart from in those IgM- and PCR-positive cases who were presumably at the stage of maximal viral replication and consequent bone marrow suppression . Properly designed epidemiological studies in larger , non-convenience samples are , however , needed to confirm these findings . Although CQ is a safe and inexpensive antimalarial drug , the increasing emergence of resistant P . falciparum and P . vivax has seen its use decline throughout the tropics . Our data suggest that the prevalence of severe malarial anemia should also fall as a result . However , when an effective B19V vaccine becomes available , this should be considered a priority intervention where pediatric B19V seroprevalence rates are high and other causes of anemia such as nutritional deficiency and intestinal parasitic infection are present . Artemisinin-based combination therapy ( ACT ) is the current WHO-recommended first-line treatment for uncomplicated malaria [29] . Our data suggest that , pending more definitive in vivo data including appropriately designed clinical trials , a non-4-aminoquinoline drug should be preferred to partner the artemisinin derivative so that the contribution of B19V to severe anemia is minimized . | Human parvovirus B19 ( B19V ) is typically associated with a childhood febrile illness known as erythema infectiosum . The infection usually resolves without consequence in healthy individuals . However , in patients with immunologic and/or hematologic disorders , B19V can cause a significant pathology . The virus infects and kills red cell precursors but anemia rarely supervenes unless there is pre-existing anemia such as in children living in malaria-endemic regions . The link between B19V infection and severe anemia has , however , only been confirmed in certain malaria-endemic countries in parallel with chloroquine ( CQ ) usage . This raises the possibility that CQ may increase the risk of severe anemia by promoting B19V infection . To test this hypothesis , we examined the direct effect of CQ and other commonly used antimalarial drugs on B19V infection in cultured cell lines . Additionally , we examined the correlation between B19V infection , hemoglobin levels and use of CQ in children from Papua New Guinea hospitalized with severe anemia . The results suggest strongly that CQ and its derivatives aggravate B19V-associated anemia by promoting B19V replication . Hence , careful consideration should be given in choosing the drug partnering artemisinin compounds in combination antimalarial therapy in order to minimize contribution of B19V to severe anemia . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"virology",
"infectious",
"diseases/viral",
"infections",
"public",
"health",
"and",
"epidemiology/infectious",
"diseases",
"public",
"health",
"and",
"epidemiology/preventive",
"medicine",
"hematology/anemias",
"microbiology/parasitology",
"infectious",
"diseases/protozoal",
"in... | 2010 | Chloroquine and Its Derivatives Exacerbate B19V-Associated Anemia by Promoting Viral Replication |
Cytosine DNA methylation is a stable epigenetic mark that is frequently associated with the silencing of genes and transposable elements ( TEs ) . In Arabidopsis , the establishment of DNA methylation is through the RNA-directed DNA methylation ( RdDM ) pathway . Here , we report the identification and characterization of RDM16 , a new factor in the RdDM pathway . Mutation of RDM16 reduced the DNA methylation levels and partially released the silencing of a reporter gene as well as some endogenous genomic loci in the DNA demethylase ros1-1 mutant background . The rdm16 mutant had morphological defects and was hypersensitive to salt stress and abscisic acid ( ABA ) . Map-based cloning and complementation test led to the identification of RDM16 , which encodes a pre-mRNA-splicing factor 3 , a component of the U4/U6 snRNP . RNA-seq analysis showed that 308 intron retention events occurred in rdm16 , confirming that RDM16 is involved in pre-mRNA splicing in planta . RNA-seq and mRNA expression analysis also revealed that the RDM16 mutation did not affect the pre-mRNA splicing of known RdDM genes , suggesting that RDM16 might be directly involved in RdDM . Small RNA expression analysis on loci showing RDM16-dependent DNA methylation suggested that unlike the previously reported putative splicing factor mutants , rdm16 did not affect small RNA levels; instead , the rdm16 mutation caused a decrease in the levels of Pol V transcripts . ChIP assays revealed that RDM16 was enriched at some Pol V target loci . Our results suggest that RDM16 regulates DNA methylation through influencing Pol V transcript levels . Finally , our genome-wide DNA methylation analysis indicated that RDM16 regulates the overall methylation of TEs and gene-surrounding regions , and preferentially targets Pol IV-dependent DNA methylation loci and the ROS1 target loci . Our work thus contributes to the understanding of RdDM and its interactions with active DNA demethylation .
Cytosine methylation in eukaryotic cells is an epigenetic mark that plays important roles in diverse biological processes , such as the silencing of genes and transposons [1] , [2] , X inactivation [3] , paramutation [4] , and imprinting [5] . In plants , cytosine methylation can occur in all three sequence contexts: CG , CHG and CHH ( H = A , T , or C ) . The Arabidopsis genome has 24% of CG , 6 . 7% of CHG and 1 . 7% of CHH sites methylated at the cytosine [6] . Maintenance of CG , CHG and CHH methylation is catalyzed by MET1 , CMT3 and DRM2 enzymes , respectively [7]–[10] . Nevertheless , de novo cytosine methylation in all three sequence contexts can be catalyzed by DRM2 [10] in a pathway known as RNA-directed DNA methylation ( RdDM ) [11] , [12] . In this pathway , a plant-specific RNA polymerase IV is recruited to transcribe transposons and some endogenous repeat loci and the transcripts are copied into double-stranded RNAs ( dsRNAs ) by RNA-DEPENDENT RNA POLYMERASE2 ( RDR2 ) [13]–[16] . The dsRNAs are then processed into 24-nucleotide ( nt ) siRNA duplexes by DICER-LIKE 3 ( DCL3 ) and the siRNAs were subsequently methylated at their 3′ ends by the RNA methylase HEN1 for stabilization [17] , [18] . PolIV , RDR2 , DCL3 and HEN1 are the key components for siRNA biogenesis and stability . In addition , the SNF2-like putative chromatin remodeling protein CLSY1 and the homeodomain transcription factor-like SHH1/DTF1 , which interacts with Pol IV , assists in the Pol IV and RDR2-dependent siRNA biogenesis [19]–[21] . Following the methylation of the siRNA duplexes by HEN1 , one strand of the siRNAs is loaded into AGO4 [22] . AGO4 interacts with the nascent transcript produced by Pol V , another plant-specific RNA polymerase , through base-pairing between the siRNA and nascent transcript [23] . Pol V transcription is facilitated by a complex formed by DRD1 , DMS3 and RDM1 ( termed DDR complex ) , and the transcripts serve as scaffolds for recruiting RdDM effector complex [24] , [25] . AGO4 also interacts with the largest subunit NRPE1 of Pol V and KTF1 , a homolog of yeast transcription elongation factor Spt5 , to stabilize the association of AGO4 with the scaffold transcripts [26]–[28] . The RDM1 protein of DDR complex is associated with AGO4 and DRM2 and thus may help to recruit DRM2 to the region being transcribed by Pol V to catalyze DNA methylation [25] , [29] . In addition to Pol V transcripts , Pol II-generated non-coding transcripts are also involved in the RdDM through recruiting the AGO4-containing effector complex [25] , [30] . Besides AGO4 , AGO6 and AGO9 are also involved in the RdDM , acting in a partially redundant manner with AGO4 [31] , [32] . Recently , the DRM2 paralog DRM3 that is catalytically mutated was reported to play a role in the RdDM through promoting the activity of DRM2 [33] . More recently , a GHKL ATPase domain-containing protein DMS11 was identified as a new component of the RdDM machinery and was proposed to cooperate with DMS3 in the RdDM pathway by providing the missing ATPase function for DMS3 [34] or to regulate chromatin architecture [35] . Together , these results suggest that there is great complexity to the RdDM pathway and more components are likely required for modulating this important pathway . Moreover , the mechanisms through which Pol IV , Pol V and DRM2 are targeted to specific loci are still not fully understood . In an attempt to identify genes involved in the RdDM , Ausin et al ( 2012 ) carried out a screen on T-DNA insertion lines by using FWA transgene silencing as a reporter system [36] . As a result , a splicing factor SR45 was identified and demonstrated to be required for the RdDM . Splicing factors are well known to be involved in the removal of introns from pre-mRNAs . In eukaryotic cells , pre-mRNA splicing takes place in a large multicomponent complex , called the splicesome , which is formed by ordered interactions of four small ribonucleoprotein particles ( snRNPs ) , U1 , U2 , U4/U6 and U5 snRNPs , and numerous snRNP-associated proteins [37] . The U1 snRNP assembles with the 5′ splice site of pre-mRNA and recruits several splicing factors to form the commitment complex , and then the U2 snRNP interacts with the branch point of the introns to form the pre-spliceosome . Subsequently , the U5 snRNP associates with the U4/U6 snRNP to form a U4/U6 . U5 tri-snRNP that assembles with the pre-spliceosome to form the mature spliceosome [37] , [38] . SR45 encodes a serine/arginine-rich ( SR ) protein belonging to a conserved family of structurally and functionally related non-snRNP proteins . SR45 is suggested to help with the formation of the bridge between 5′ and 3′ splice sites in the early stage of spliceosome assembly through interacting with U1-70K and U2AF35b proteins [39] , [40] . The mechanism of SR45 function in RdDM is not understood . Although the abundance of Pol IV-dependent siRNAs is decreased in sr45 mutant , it is still not known how SR45 is involved in the siRNA accumulation and it is possible that SR45 may play an indirect role through the splicing of genes that encode RdDM pathway components [36] . In this study , we report a U4/U6 snRNP associated protein RDM16 , which is required for the RdDM . RDM16 encodes a pre-mRNA-splicing factor 3 ( PRP3 ) and is involved in the pre-mRNA splicing in planta . The rdm16 mutation did not influence the pre-mRNA splicing of known RdDM genes , suggesting that RDM16 involvement in the RdDM might be direct . We also show that rdm16 did not affect the siRNA levels but decrease the Pol V transcripts , which suggested that RDM16 functions in a later step of RdDM through a different mechanism from that of SR45 . In addition , we performed genome-wide DNA methylation analysis and found that the majority of loci whose methylation is RDM16-dependent overlapped with Pol IV- and ROS1-targeted loci . Together , our results shed new light on the RdDM pathway and the dynamic balance between RdDM and active DNA demethylation .
To identify components involved in the RdDM machinery , we carried out a forward genetics screen on a T-DNA-mutagenized population in the ros1 background , which contains the RD29A promoter-driven luciferase reporter gene ( RD29A-LUC ) as well as the 35S promoter-driven NPTII transgene ( 35S-NPTII ) [41] . In this system , the RD29A-LUC transgene , endogenous RD29A and 35S-NPTII transgene are well expressed in wild-type plants under stress conditions , while mutations in ROS1 , the 5-methylcytosine DNA glycosylase/DNA demethylase gene , led to the silencing of all the three genes . Based on reactivation of the RD29A-LUC in ros1 , we have identified a number of genes required for RdDM [28] , [29] , [42]–[44] . In this study , we identified a new mutant , rdm16 , as a suppressor of ros1 ( Figure 1A ) . Mutation of RDM16 in the ros1 background caused a partial release of the silencing of RD29A-LUC under stress treatment . Nevertheless , the silencing of the RdDM-independent 35S-NPTII transgene was not released in rdm16ros1 compared to the ros1 single mutant ( Figure S1A ) . We also performed quantitative RT-PCR for the expression analysis of RD29A-LUC and endogenous RD29A under various stress conditions . The results showed that mutation of RDM16 reactivated the expression of both RD29A-LUC and endogenous RD29A under salt , ABA and cold treatments ( Figures 1B and 1C ) . To examine the cause for the reactivation of RD29A-LUC expression in the rdm16ros1 mutant , we carried out DNA methylation analysis by bisulfite sequencing of WT , ros1 and rdm16ros1 mutants . Consistent with previous reports , the DNA methylation of both RD29A-LUC and endogenous RD29A promoters was detected at low levels in WT , while heavy cytosine methylation at all sequence contexts ( CG , CHG and CHH ) was found in ros1 ( Figures 2A and 2B ) . In comparison with ros1 , the high methylation of both RD29A-LUC and endogenous RD29A promoters was substantially decreased in the rdm16ros1 double mutant at all the three cytosine contexts CG , CHG and CHH . We also measured the DNA methylation in the 35S promoter and the results showed that the DNA methylation level of 35S promoter in rdm16ros1 was similar to that in ros1 ( Figure S1B ) , which is consistent with the phenotype that mutation of RDM16 did not affect the silencing of 35S-NPTII transgene in ros1 ( Figure S1A ) . Additionally , we also performed DNA methylation analysis on a ROS1-targeted endogenous locus , At4g18650 [45] , and RdDM-targeted repeat loci , AtSN1 , SoloLTR and MEA-ISR . Results showed that DNA methylation of At4g18650 promoter and AtSN1 was reduced in rdm16ros1 at all three cytosine contexts ( CG , CHG and CHH ) in comparison with that in ros1 ( Figures 2C and 2D ) . The decreased level of DNA methylation in At4g18650 promoter was comparable to or even stronger in rdm16ros1 than that in nrpd1ros1 ( Figure 2C ) . Mutation of RDM16 also reduced the DNA methylation at CHG sites of SoloLTR , while its effect on CG and CHH was not observed ( Figure S2A ) . rdm16ros1 did not affect the DNA methylation at MEA-ISR locus ( Figure S2B ) . Previous reports revealed that the expression of ROS1 is sensitive to RdDM mutations [29] , [42]–[44] . Our expression analysis showed that like previously reported RdDM mutants , the expression of ROS1 was also reduced in rdm16ros1 ( Figure 1D ) , which is consistent with a role of RDM16 in the RdDM pathway . Mutation of RDM16 also caused morphological defects , including dwarf stature , smaller , rounded and wrinkled leaves , and smaller siliques ( Figures 3A–3C ) . Furthermore , seed germination of rdm16ros1 mutant was hypersensitive to salt stress and abscisic acid ( ABA ) ( Figures 3D–3F ) . Nevertheless , at the young seedling stage , rdm16ros1 did not show higher sensitivity to salt and ABA than ros1 and WT ( Figures 3G and 3H ) . To perform genetic analysis for rdm16ros1 , we generated an F2 population from a cross between rdm16ros1 and ros1 and then observed the LUC signal and morphological defects . We found that reactivation of LUC expression was tightly linked to the morphological defects . Among 235 F2 plants examined , 59 plants showed both increased LUC signal and morphological defects , while the rest 176 plants showed low LUC signal and normal morphology . The ratio of plants with high LUC expression to plants with low LUC expression fitted to 1∶3 ( P>0 . 95 ) , suggesting that the increased LUC expression and morphological defects was controlled by a recessive mutation in a single nuclear gene . To map the responsible gene , we used an F2 population derived from a cross between rdm16ros1 ( C24 background ) and ros1-4 ( Col background ) and selected plants with defective morphology and increased LUC signal for mapping . By using 25 indel markers across the genome and 74 plants , we were able to find the marker At124 at 11 . 3 Mb of chromosome 1 was linked to the RDM16 gene ( Figure S3A ) . By developing three more polymorphic markers around At124 , we mapped the gene between At120 and At124 on chromosome 1 with a recombination rate of 8 . 7% and 4 . 7% from the RDM16 gene , respectively . To further map the gene , we used a population of 632 plants and developed additional 6 polymorphic markers between the At120 and At124 markers . Through linkage analysis , we further mapped the RDM16 gene between the At126 and At130 markers , with 13 and 1 recombinants , respectively ( Figure S3B ) . Three markers within the two markers At126 and At130 were tightly linked to the gene . As a result , the candidate region for RDM16 was defined to about 1 . 39 Mb ( Figure S3B ) . To clone the gene , we sequenced the whole genome of rdm16ros1 mutant by second-generation high throughput DNA sequencing . In the 1 . 39 Mb mapping interval , we found a gap in the promoter of At1g28060 in rdm16ros1 mutant , suggesting a deletion or insertion occurred in the region ( Figure S4A ) . PCR analysis showed that the DNA fragment spanning the gap could not be amplified in rdm16ros1 , whereas it was well amplified in WT and ros1 mutant ( Figure S4B ) . These results suggested that there was a DNA fragment inserted into the At1g28060 promoter of rdm16ros1 mutant . To determine the DNA sequence inserted into the promoter , we performed TAIL-PCR analysis on both sides of the insertion . We detected partial sequence of At1g24590 from this analysis , although the full sequence of the insertion was not known ( Figure S4C ) . In addition , we found that the insertion also caused a 45-bp deletion ( -154 to -109 bp from ATG ) in the promoter of At1g28060 . To examine the effect of the mutation on the expression of At1g28060 , we compared mRNA expression level of the gene between rdm16ros1 and ros1 . In comparison with ros1 , mRNA expression of the coding sequence of the gene was decreased by about 2 fold in rdm16ros1 , and the expression of 5′UTR was greatly reduced ( Figure S5 ) . These results suggest that decreased expression of At1g28060 might be responsible for the observed phenotypes in rdm16ros1 . To confirm that At1g28060 is RDM16 , two T-DNA insertion lines ( rdm16-2 and rdm16-3 ) were obtained and characterized . rdm16-2 has a T-DNA insertion in the fifth intron of At1g28060 ( Figures S6A and S6B ) . The T-DNA insertion did not abolish the expression of the full mRNA of the gene , but reduced its expression ( Figures S6C and S6D ) . rdm16-2 also showed similar defective leaf phenotype to rdm16-1 , including smaller , rounded and wrinkled leaves , although the morphological defects in rdm16-2 were stronger than those in rdm16-1 ( Figure S6E ) . To examine whether rdm16-2 affected DNA methylation , we compared the DNA methylation level of AtSN1 between WT and rdm16-2 through bisulfite sequencing . Like rdm16-1 , rdm16-2 also decreased DNA methylation at all three cytosine contexts ( CG , CHG and CHH ) in comparison with its WT ( Figure S6F ) . Furthermore , the expression of ROS1 was reduced in rdm16-2 as well ( Figure S6G ) . The rdm16-3 mutant contains a T-DNA insertion in the fifth exon of At1g28060 ( Figure S6A . Genomic PCR analysis with selfed progeny of a heterozygous plant of rdm16-3 showed that no homozygous rdm16-3 plants were found among 107 selfed progeny , suggesting that homozygous rdm16-3 mutant was lethal . In addition , we found wild-type and heterozygous plants at a 68∶39 ratio , distorting from the expected ratio of 1∶2 , which suggested that the viability of male and/or female gametes of the heterozygous plants was affected . To test this hypothesis , we made reciprocal crosses between heterozygous rdm16-3/+ and SALK_057447C insertion line , and then analyzed the frequency of rdm16-3 haploid in the resultant F1 plants . The SALK_057447C line did not show morphological defects and its T-DNA insertion site could be used to determine whether the F1 plants are from real crosses . Of 107 real F1 plants with SALK_057447C line as the pollen donor , 17 plants were heterozygous for rdm16-3 , whereas 90 plants showed wild-type RDM16 . Among 82 F1 plants with rdm16-3/+ as the pollen donor , there were 24 heterozygous and 58 wild-type plants . These results indicated that knockout of RDM16 in both female and male gametes reduced their viability and the effect on female gametes was more severe . To further confirm that the mutation in At1g28060 is responsible for the observed phenotype in rdm16ros1 , we conducted a complementation test on rdm16ros1 by introducing a wild-type RDM16 with 1 . 4 kb promoter and the full genomic sequence of the gene into the mutant . In the complementation lines , the morphological defects including dwarf stature , smaller and deformed siliques , and defective leaves were fully complemented in the T2 generation ( Figure 4A ) . We also analyzed the RD29A-LUC expression by luminescence imaging and DNA methylation level of both RD29A-LUC and endogenous RD29A promoters by bisulfite sequencing in the complementation lines . Results showed that the increased expression of RD29A-LUC and the reduced DNA methylation level of both promoters in rdm16ros1 were rescued in the complementation lines ( Figures 4B–4D ) . Taken together , these results show that At1g28060 is RDM16 and the mutation in At1g28060 is responsible for the altered expression and DNA methylation of reporter gene and endogenous loci , and the defective plant development in rdm16ros1 . RDM16 has six exons and five introns , encoding a peptide of 786 amino acids with a predicted molecular mass of 88 . 6 kD . BLAST searches indicated that RDM16 encodes a pre-mRNA-splicing factor 3 , a component of U4/U6 snRNP protein complex . Domain analysis predicted that RDM16 has a pre-mRNA processing factor 3 ( PRP3 ) domain and a DUF1115 domain with unknown function ( Figure S7 ) . RDM16 is conserved in eukaryotes and has putative orthologs in rice , human and yeast ( Figure S7 ) . For example , RDM16 exhibits 26 . 8% identity and 46 . 0% similarity at the amino acid level to human HPRP3 ( accession number NP_004689 ) , whose mutation results in autosomal dominant retinitis pigmentosa in humans [46] . RDM16 has 30 . 2% similarity to yeast Prp3 ( accession number NP_010761 ) , which is a U4/U6 snRNP protein necessary for the integrity of U4/U6 snRNP and U4/U6 . U5 tri-snRNP [47] . In addition , there is a homologous protein , At3g55930 , sharing 40 . 2% identity with RDM16 , encoded in the Arabidopsis genome . To examine whether RDM16 is indeed involved in pre-mRNA splicing , we carried out Illumina paired-end RNA-seq in WT ( Col ) and rdm16-2 mutant seedlings . A total of 51 . 5 million and 53 . 2 million reads for WT and rdm16-2 with the average length of 90 nucleotides were generated , respectively . Among them , there were 48 . 3 million and 49 . 6 million unique reads for WT and rdm16-2 , respectively , mapped to the Arabidopsis genome . Analysis of intron-retention events in WT and rdm16-2 showed that 308 intron-retention events in 258 genes occurred in rdm16-2 mutant in comparison with WT ( FDR<0 . 01 ) ( Table S1 ) . This result indicated that RDM16 is involved in pre-mRNA splicing in planta . Our RNA-seq analysis also showed that there were 689 and 152 genes significantly upregulated ( >2 fold ) and downregulated ( <2 fold ) in the rdm16-2 mutant , respectively ( Tables S2 and S3 ) . Although the expression of so many genes was altered , no genes involved in the RdDM pathway were found in the list . Additionally , we did not find any genes involved in the RdDM pathway showing splicing defects in the mutant ( Table S1 ) . Therefore , it is unlikely that the altered DNA methylation in the rdm16 mutants was an indirect consequence of reduced expression or splicing defect of genes encoding RdDM components . To further exclude the possibility of RDM16 affecting the expression or splicing of the RdDM components , we carried out real-time RT-PCR and regular RT-PCR to examine the expression of the RdDM components among WT , ros1 and rdm16ros1 by using primers spanning introns of the genes . The real-time RT-PCR analysis showed that the expression of the genes involved in the RdDM pathway was not different in the three lines ( Figure S8A ) . No splicing defects in any RdDM pathway genes were found in the rdm16ros1 mutant by the RT-PCR analysis ( Figure S8B ) . Together , these results indicated that the involvement of RDM16 in RdDM was not an indirect effect of RDM16 on the expression or splicing of RdDM components . To examine the effect of RDM16 on the Arabidopsis methylome , we performed whole genome bisulfite sequencing in WT , ros1 , rdm16ros1 and nrpd1ros1 ( 4-week-old leaves ) . The average depth of sequenced methylomes is 15–18 with 0 . 3–0 . 5% error rates , which indicated a high quality of the data ( Table S4 ) . We analyzed the methylation levels of transposable elements ( TEs ) , genes , and 2 kb upstream or downstream of the genes or TEs . Transposable elements are heavily methylated and show higher methylation levels than surrounding regions at all sequence contexts ( Figure 5A , Figure S9A ) . In ros1 mutant , the overall CG and CHH methylation of TEs and surrounding regions were slightly increased in comparison with the wild-type . Mutation of RDM16 did not affect the CG and CHG methylation ( Figure S9A ) . However , CHH methylation of both TEs and surrounding regions was substantially decreased in rdm16ros1 in comparison with ros1 or WT ( Figure 5A ) . To investigate whether RDM16 has preference for TEs of different sizes , we divided the TEs into five groups based on the TE size and calculated average DNA methylation levels . The DNA methylation level of TEs was gradually elevated as the TE size increased . Mutation of RDM16 reduced the CHH methylation of all five groups of TEs , but did not influence CG and CHG methylation ( Figure 5A , Figure S9A ) . This result indicates that RDM16 targets all TEs and does not have a preference for TEs of different sizes . The methylation pattern of TEs in rdm16ros1 is similar to nrpd1ros1 , although the decrease in the level of DNA methylation was less dramatic in rdm16ros1 ( Figure 5A ) . In contrast to TEs , gene body methylation was mainly in the CG sequence context and the CG methylation is depleted towards both the 5′ and 3′ ends of genes ( Figure S9B ) , which suggested that CG methylation at the 5′ and 3′ ends of the genes might not be compatible with Pol II-mediated transcription . Mutation of RDM16 did not affect overall DNA methylation of gene bodies in any sequence context ( Figure 5B , Figure S9B ) . The methylation level of gene body was also not reduced in the nrpd1ros1 mutant , suggesting that the gene body methylation is largely independent of RdDM . Nevertheless , when we divided the genes into 5 groups with different gene sizes for the methylation analysis , we found that CHH methylation of short genes ( <1 kb ) was reduced in rdm16ros1 as well as nrpd1ros1 compared to ros1 mutant ( Figure 5B , Figure S9B ) . Short genes have low CG methylation and increased CHH methylation , which suggested that RdDM pathway preferentially targets short genes , which results in elevated CHH methylation , and RDM16 is involved in the regulation of the DNA methylation of short genes . The CHG and CHH methylation levels of gene surrounding regions were greater than that of gene body , and 5′ promoter regions had higher methylation level than 3′ downstream region of genes in the wild-type ( Figure 5B , Figure S9B ) . In ros1 mutant , the DNA methylation level was increased in all sequence contexts and the effect on CHH methylation was greater , suggesting that ROS1 targeted the promoters and downstream sequences of genes . Additionally , as gene size increased , the difference in CHH methylation in 3′ downstream regions of genes was decreased between WT and ros1 mutant ( Figure 5B ) , which suggested that ROS1 preferentially targeted 3′ downstream regions of short genes . Mutation of RDM16 did not affect the CG methylation , but slightly reduced the CHG methylation and notably decreased the CHH methylation in both 5′ and 3′ regions of genes ( Figure 5B , Figure S9B ) . In the rdm16ros1 mutant , the CHH methylation of 5′ and 3′ regions of genes of all sizes was reduced to a similar level as in WT , while in the nrpd1ros1 mutant the CHH methylation was further decreased ( Figure 5B ) . These results suggest that ROS1 actively counteracts the effect of RDM16 and the RdDM pathway on the DNA methylation of 5′ promoters and 3′ downstream regions of genes . We next identified differentially methylated regions ( DMRs ) in rdm16ros1 compared to ros1 mutant . There were 747 loci with decreased DNA methylation and 468 loci with increased DNA methylation in rdm16ros1 in comparison with ros1 mutant ( Tables S5 and S6 ) , while the number of hypomethylated and hypermethylated loci in nrpd1ros1 was 3929 and 1417 , respectively ( Tables S7 and S8 ) . Of the 747 hypomethylated loci in rdm16ros1 , 347 loci were located in intergenic regions , and most of the 347 loci were located within 1 . 5 kb promoter regions ( Figures 6A and 6B ) , which suggested that RDM16 was preferentially involved in the regulation of DNA methylation of intergenic regions and potentially involved in regulating gene transcription . In contrast to RDM16 , transposable elements ( TE ) are the main targets of NRPD1 ( Figure 6A ) . Despite their different preferences , the hypomethylated loci of rdm16ros1 largely overlapped ( 77% ) with those of nrpd1ros1 ( Figure 6C ) . Moreover , the non-overlapping hypomethylated loci of rdm16ros1 also showed reduced DNA methylation levels in nrpd1ros1 , especially at the CHH context . This result suggests that all RDM16 targets are influenced by the dysfunction of NRPD1 , although some RDM16 targets were excluded from the list of NRPD1 targets due to the stringent cutoffs used to define differentially methylated regions . Interestingly , we noticed that the DNA demethylation enzyme ROS1 also preferentially targeted intergenic regions ( Figure 6A ) . When comparing the 747 hypomethylated loci in rdm16ros1 with the hypermethylated loci in ros1 , we found a striking overlap between them ( Figure 6D ) . Similar to the case in nrpd1ros1 , the nonoverlapping hypomethylated loci of rdm16ros1 also showed altered DNA methylation in ros1 . Together , these results suggest that RDM16 is involved in the RdDM pathway and preferentially regulates the DNA methylation of loci that are targeted for demethylation by ROS1 . To validate the whole-genome DNA methylation results , we selected 3 from the 747 hypomethylated loci for individual locus bisulfite sequencing . All three loci showed low levels of DNA methylation in WT , but became heavily methylated in ros1 at all sequence contexts ( Figure 7 ) . In the rdm16ros1 double mutant , the high DNA methylation was reduced compared to the ros1 single mutant , especially at the CHG and CHH contexts , although the decreased level of DNA methylation in rdm16ros1 was less prominent than in nrpd1ros1 ( Figure 7 ) . To investigate whether rdm16 may alter DNA methylation through influencing small RNA levels , we performed small RNA Northern blot analysis in WT , ros1 , rdm16ros1 , nrpd1ros1 and nrpe1ros1 . The results showed that the rdm16 mutation did not affect the accumulation of any of the tested siRNAs compared to WT and ros1 controls , although the abundance of these tested siRNAs was greatly reduced in both nrpd1ros1 and nrpelros1 mutants ( Figure 8 ) . The targets of these siRNAs include the RD29A promoter , Solo-LTR and AtSN1 whose methylation levels were decreased in the rdm16ros1 mutant ( Figure 2 , Figure S2A ) . Therefore , these results suggest that RDM16 regulates DNA methylation not through influencing the siRNA abundance . The lack of effect of the rdm16 mutation on siRNA abundance suggests that RDM16 might function at a later step in the RdDM pathway . Therefore , we examined whether the rdm16 mutation may affect DNA methylation by influencing Pol V transcript levels . We selected 13 Pol V-dependent loci [48] including the RD29A-LUC promoter for analysis . Pol V-dependent transcripts of 8 of the 13 loci were detected in the C24 wild type under our conditions . Five of the 8 loci including the RD29A-LUC promoter displayed a decreased expression in rdm16ros1 compared to the WT and ros1 controls , although the reduction in expression level was less in rdm16ros1 than in nrpe1ros1 ( Figure 9 ) . All of the five loci with decreased Pol V transcripts in rdm16ros1 showed a reduced DNA methylation level to some extent in rdm16ros1 compared to ros1 ( Figures 2A and 2B , Figure S10 ) . We also compared the levels of the Pol V-dependent transcripts in Col WT , rdm16-2 and nrpe1-11 . The results show that 8 loci had a decreased expression in rdm16-2 compared to WT , although the reduction in rdm16-2 was not as dramatic as in nrpe1-11 ( Figure S11 ) . These data suggest that RDM16 regulates DNA methylation through influencing Pol V transcripts . To investigate whether RDM16 may be associated with its target loci , we performed chromatin immunoprecipitation ( ChIP ) assays on Pol V target loci by using native promoter-driven RDM16-3xFlag or RDM16-3xHA transgenic lines . The RDM16-3xFlag or RDM16-3xHA construct complemented the defects of rdm16 mutant and we were able to detect tagged RDM16 protein in the transgenic lines by Western blot analysis ( Figure S12 ) , indicating that the tagged proteins are expressed and functional in vivo . The ChIP results show that RDM16 was enriched at all of the target loci tested in both RDM16-3xFlag and RDM16-3xHA transgenic lines ( Figure 10 ) , which indicated that RDM16 is associated with its target loci . To examine the subnuclear localization patterns of RDM16 , we used the RDM16-3xFlag transgenic line to perform immunolocalization assays . The results show that the RDM16 protein was dispersed throughout the nucleoplasm without any preferential accumulation at the Cajal body ( Figure S13 ) , based on the co-localization analysis between RDM16 and U2B , a maker for the Cajal body [26] . This localization pattern of RDM16 differs from that of ZOP1 , a splicing factor recently reported to be involved in the RdDM pathway through affecting Pol IV-dependent siRNA accumulation [49] . In addition to the dispersed nucloplasmic localization , ZOP1 also preferentially accumulates in the Cajal body . The presence of ZOP1 but not RDM16 in the Cajal body is consistent with our hypothesis that RDM16 regulates DNA methylation through a different mechanism from those of reported splicing factors . ”
In this study , we isolated a new factor , RDM16 , that functions in the RdDM pathway through a forward genetic screen . RDM16 is required for the transcriptional silencing and the increased DNA methylation of transgenic RD29A-LUC in ros1 mutant . Our genome-wide DNA methylation analysis showed that RDM16 also affects DNA methylation of TEs and gene surrounding regions globally and RDM16 preferentially influences NRPD1- and ROS1-targted loci . RDM16 encodes a homolog of yeast Prp3 protein , which is a component of U4/U6 snRNP-associated protein and is involved in pre-mRNA-splicing in yeast [47] . Our RNA-seq data indicated that RDM16 is involved in the pre-mRNA-splicing in Arabidopsis ( Table S1 ) . None of the mis-spliced genes had changes in DNA methylation in the rdm16 mutant , suggesting that splicing and regulation of DNA methylation are two separate functions of RDM16 . However , there is a possibility that RDM16 regulates DNA methylation by influencing expression and/or pre-mRNA splicing of genes involved in the RdDM pathway . Our RNA-seq and RT-PCR analysis indicated that RDM16 did not affect the expression or pre-mRNA splicing of genes encoding known RdDM components ( Tables S1 , S2 , S3 , Figure S7 ) . These suggest that RDM16 involvement in the regulation of DNA methylation might be quite direct . To examine the mechanism of RDM16 in the RdDM , we compared the expression level of small RNAs and Pol V transcripts . However , unlike previously reported putative splicing factor sr45 , mutation of RDM16 did not influence the small RNA accumulation but reduced the expression level of Pol V transcripts ( Figures 8 and 9 , Figure S11 ) , which suggested that RDM16 regulates DNA methylation through a different mechanism from SR45 . RDM16 is critical for normal plant development , which is likely due to its role in the splicing of developmentally important genes . Knockout of RDM16 in both female and male gametes reduced their viability and knockout of both maternal and paternal RDM16 alleles in the embryo is lethal . Knockdown of RDM16 caused a series of developmental defects , including reduced plant stature , smaller leaves and siliques , rounded and wrinkled leaves ( Figures 3A–C , Figure S6E ) . Furthermore , Knockdown of RDM16 led to an increased sensitivity to salt stress and ABA in seed germination ( Figures 3D–F ) . Dysfunction of STA1 , the U5 snRNP-associated pre-mRNA-splicing factor , also caused an increased sensitivity to ABA [50] . These suggest that hypersensitivity to ABA might be a common feature in the splicing factor-defective mutants . Nevertheless , there are differences in the physiological defects between the rdm16 and sta1 mutants . The hypersensitivity to chilling in sta1 mutant was not observed in the rdm16 mutant . On the other hand , sta1 did not have the phenotype of hypersensitivity to salt stress as rdm16 did . Moreover , unlike in sta1 , the hypersensitivity to salt stress and ABA was not observed in the rdm16 mutant at the seedling stage ( Figures 3G and 3H ) . Since both rdm16-1 and sta1-1 are weak mutant alleles but are not nulls , it is not clear whether the different physiological phenotypes between rdm16-1 and sta1-1 were due to the different mutations or due to the distinct targets of RDM16 and STA1 . In fission yeast , the formation of heterochromatin requires the RNAi machinery whose core components consist of Dicer ( Dcr1 ) , Argonaute ( Ago1 ) and RNA-dependent RNA polymerase ( Rdp1 ) [51] , [52] . The pathway of RNA-induced heterochromatin formation in fission yeast parallels the RdDM pathway in Arabidopsis . In the fission yeast , Pol II transcribes the dg and dh repeats of centromeric regions and then Rdp1-containing RNA-dependent RNA polymerase complex ( RDRC ) and Dcr1 process the centromeric transcripts into siRNAs [51] , [52] . The siRNA is loaded into Ago1 to form the RNA-induced transcriptional gene-silencing ( RITS ) complex that also contains the chromodomain protein Chp1 and GW-motif-containing protein Tas3 . Through base-pairing of the siRNA with Pol II-produced nascent repeat transcripts , the RITS is specifically recruited to the target region and facilitate the recruitment of histone-modifying enzymes such as the H3K9 methyltransferase Clr4 to induce H3K9 methylation and the formation of heterochromatin [51] , [52] . Recently , several splicing factors were reported to be involved in the RNAi-directed silencing process in fission yeast [53] , [54] . Mutation of these splicing factors , but not the splicing itself , affects the accumulation of centromeric siRNAs and consequently the integrity of centromeric heterochromatin . Splicing factors are found to be co-purified with affinity-selected Cid12 , a component of RDRC , and therefore it was proposed that splicing factors might be recruited to centromeric noncoding RNA through the recognition of dg intron sequence and then interact with RDRC to provide a platform for siRNA generation and finally facilitate the centromere repeat silencing [53] , [54] . Among these splicing factors involved in the centromere repeat silencing , Prp3 , the homolog of RDM16 in fission yeast , is not included . Nevertheless , Prp3 was co-immunoprecipitated with FLAG-tagged Cid12 and mutation of Prp3 led to the increased accumulation of dg transcripts [53] , [54] , which suggest that Prp3 is also involved in the processing of the centromeric transcripts into siRNAs , but the influence may not be strong enough to have an observable effect on the centromere repeat silencing in the mutant . In this report , we showed that RDM16 , a homolog of yeast Prp3 , is involved in the RdDM pathway in Arabidopsis . Mutation of RDM16 reduced the DNA methylation and released the transcriptional silencing of transgene and endogenous targets . However , rdm16 mutation did not influence the siRNA accumulation at DNA methylation-affected loci . These results suggest that RDM16 function in the RdDM pathway might adopt a different mechanism from the yeast splicing factors in the pathway of centromere repeat silencing . Mutation of SR45 causes a decreased accumulation of siRNAs and the AGO4 protein level was also reduced , which suggested that SR45 acts at an early step in the RdDM pathway in siRNA generation [36] . Recently , Zhang et al . ( 2013 ) reported that ZOP1 and several other splicing factors are also involved in regulating DNA methylation through influencing siRNA abundance [49] . Since the rdm16 mutation did not cause a reduction in siRNA abundance , this suggests that RDM16 likely functions in a later step in RdDM . Indeed , mutation of RDM16 caused a reduction in the levels of Pol V transcripts . Dysfunction of Pol V not only abolishes the Pol V transcripts but also influences the siRNA abundance . However , the rdm16 mutation caused a decrease in Pol V transcript levels but did not affect the siRNA abundance , suggesting that the reduction in the levels of Pol V transcripts in rdm16 may not be strong enough to cause a decrease in the siRNA accumulation or that Pol V may function in the siRNA accumulation independently of its transcripts . As a splicing factor , RDM16 has a known role in RNA processing , so it is possible that RDM16 function in RdDM may involve an interaction with scaffold RNAs generated by Pol V or Pol II . Consistent with this hypothesis , our ChIP assays indicated that RDM16 is associated with Pol V target loci . The association of RDM16 with Pol V target loci might be mediated through the interaction of RDM16-containing complex with Pol V or nascent Pol V transcripts . Further work is required to determine whether RDM16 is associated with Pol V and/or can interact with Pol V transcripts . Regardless , the fact that SR45 [36] and ZOP1 [49] but not RDM16 are involved in siRNA accumulation suggests that the various splicing factors function at different steps in the RdDM pathway . This is consistent with our notion that RDM16 functions directly in RdDM , and further argues against the model that splicing factors function indirectly in RdDM by affecting the pre-mRNA splicing of genes encoding RdDM components . Our analysis of the methylome of the rdm16 mutant showed that dysfunction of RDM16 affected the overall CHH methylation of TEs ( Figure 5A ) , indicating an important role of RDM16 on the silencing of parasitic sequences . The overall methylation of gene body was not influenced by the mutation of any of the genes , ROS1 , RDM16 or NRPD1 ( Figure 5B ) . However , in short genes ( <1 kb ) , mutation of all three genes affected the genic CHH methylation levels , and the methylation level was similar between rdm16ros1 and WT . These suggest that DNA methylation of short genes is largely dependent on the RdDM pathway and ROS1 demethylates the DNA methylation that is mainly contributed by RDM16 through the RdDM pathway . In the gene surrounding regions , mutation of ROS1 caused increased DNA methylation levels of both 5′ promoters and 3′ downstream regions of the genes , with the methylation peaks close to gene ends ( Figure 5B ) , which suggested that ROS1 might be important for the regulation of DNA methylation of gene regulatory elements and thus likely affects gene transcription . Mutation of RDM16 in the ros1 background reduced the CHH methylation to the wild-type level . Furthermore , RDM16 target loci were largely overlapping with ROS1 target loci . The active DNA demethylation machinery may recognize features resulting from RdDM and thus preferentially demethylate RdDM target sequences . It would be interesting to investigate whether RDM16 or its associated proteins may be involved in somehow marking the RdDM target sequences for demethylation .
The wild-type C24 and ros1 mutant plants carry a homozygous RD29A promoter-driven luciferase transgene and a 35S promoter-driven NPTII transgene [41] . A T-DNA mutagenized ros1 population was generated and screened for suppressors of ros1 as described previously [43] . The rdm16ros1 mutant was obtained from this screening . rdm16-2 ( CS861738 ) , rdm16-3 ( CS861738 ) and another T-DNA insertion line ( SALK_057447C ) have the Columbia-0 ( Col ) genetic background . Plants were grown in a growth chamber or controlled room at 23°C with 16 h of light and 8 h of darkness . The rdm16ros1 was crossed to ros1 ( C24 background ) or ros1-4 ( Col background , Salk_045303 ) for genetic analysis or molecular mapping . The selfed F2 plants were subjected to the observation of luminescence emission and morphological alterations . For the mapping of RDM16 , 30 F2 plants with low luminescence signal and defective morphology were used to determine the location of RDM16 in the genome by using 25 indel polymorphic markers evenly distributed over the genome and then 74 F2 mutants and three more markers were used to map the RDM16 on the chromosome 1 . For the fine mapping of RDM16 , a total of 632 F2 plants were used and 6 additional markers between At120 and At124 markers were developed . As a result , RDM16 was finally mapped between At126 and At130 makers with a physical candidate region of 1 . 39 Mb . The second-generation high throughput DNA sequencing ( Illumina ) was carried out to search the mutation occurred in the rdm16ros1 mutant . There was a DNA fragment inserted in the promoter of At1g28060 . TAIL-PCR technique was used to determine the inserted position and the sequence of the DNA fragment [55] . For the complementation test of rdm16ros1 mutant , a DNA fragment harboring 1 . 4 kb promoter , the gene and 0 . 7 kb downstream of stop codon of At1g28060 was amplified and finally cloned into a binary vector pORE-O2 . The resultant vector was transformed into rdm16ros1 mutant through Agrobacterium-mediated floral dip method . T2 plants were genotyped by genomic PCR and plants harboring the transgene were subjected for morphological observations , luminescence imaging and bisulfite sequencing analysis . For the luminescence imaging , leaves of 4-week-old plants were detached and treated with 2% NaCl for 3 h in the light at 23°C , and then applied for the luminescence imaging . For the sensitivity assay at germination stage , seeds of WT , ros1 and rdm16ros1 were sowed on a 1/2 MS plate containing 0 , 75 mM NaCl or 0 . 5 µM ABA . The seed-containing plates were stratified at 4°C for 3 d and then exposed to a long day condition at 23°C . After 15 d treatment , the seedlings were photographed and compared . For the sensitivity assay at seedling stage , seeds were germinated and grown on a 1/2 MS plate for 6 d and then the seedlings were transferred to a 1/2 MS plate containing different concentrations of salt ( 0 , 75 , 100 or 125 mM NaCl ) or ABA ( 0 , 0 . 5 , 1 or 2 µM ABA ) . After 7 d treatment , the root elongation was measured and compared . To examine the expression of RD29A-LUC and endogenous RD29A , 12-day-old seedlings were exposed to various stress conditions: 0 , 150 mM NaCl for 3 h , 50 µM ABA for 3 h or cold ( 4°C ) for 2 d . After the treatment , the whole seedlings were sampled for RNA isolation . For the expression analysis of ROS1 , RdDM genes and Pol V transcripts , leaves of plants with three weeks old were collected for RNA isolation . Total RNA was extracted using the RNeasy Mini Kit ( Qiagen ) and then treated with Turbo DNase ( Ambion ) for 45 min to remove the contaminated DNA . Total RNA was used for first strand cDNA synthesis using a SuperScript II kit ( Invitrogen ) , following the manufacturer's instructions with an oligo ( dT ) 12–18 or random primer . The derived cDNA was used as template for semiquantitative or real-time RT-PCR analysis . For small RNA accumulation analysis , 2-week-old seedlings of the wild-type , ros1 , rdm16ros1 , nrpd1ros1 and nrpe1ros1 were harvested for the small RNA extraction . The extraction method and northern blot analysis of small RNA have been described previously [43] . The primers or probes for the expression analysis are listed in Table S9 . Leaves of 4-week-old plants were collected for genomic DNA isolation . The DNA was extracted using the DNeasy Plant Mini Kit ( Qiagen ) . To perform bisulfite sequencing , 80–100 ng of DNA was sodium-bisulfite converted and purified by using the BisulFlash DNA Modification Kit ( EIPGENTEK ) . For the DNA methylation analysis , at least 15 clones at each sample were sequenced . The primers for the bisulfite sequencing are listed in Table S9 . Leaves of the wild-type ( Col ) and rdm16-2 plants with 4 weeks old were collected for RNA isolation . Total RNA was extracted using the RNeasy Mini Kit ( Qiagen ) . The extracted RNA was sent to BGI ( Shenzhen , China ) for RNA-seq library preparation and whole transcriptome sequencing . The raw reads were aligned to the Arabidopsis genome ( TAIR10 , www . arabidopsis . org ) by using TopHat program ( http://tophat . cbcb . umd . edu ) . The assembling of the reads and the calculation of transcript abundance were performed by Cufflinks ( http://cufflinks . cbcb . umd . edu ) . Transcripts that were differentially expressed in WT and rdm16-2 were identified by Cuffdiff , a part of the Cufflinks package . For the intron-retention analysis , reads located in intron regions were calculated in WT and rdm16-2 separately , and then degrees of differential expression were measured according to the method described by Audic et al . ( 1997 ) [56] , which was constructed based on Poisson distribution eliminating the influence of sequencing depth . The introns with more than 95% read coverage and false discovery rate ( FDR ) <0 . 01 were regarded as intron-retention events . ” DNA was extracted from leaves of 4-week-old plant and sent to BGI ( Shenzhen , China ) for bisulfite treatment , library preparation , and sequencing . For data analysis , adapter and low quality sequences ( q<20 ) were trimmed and clean reads were mapped to a pseudo-C24 genome using BRAT-BW [57] allowing two mismatches . The pseudo-C24 genome was generated through the replacement of SNPs in the Col-0 genome with C24 variants ( http://1001genomes . org/data/MPI/MPISchneeberger2011/releases/current//C24/Marker/C24 . SNPs . TAIR9 . txt ) . The method for identification of differentially methylated regions ( DMRs ) was according to Qian et al . ( 2012 ) with minor modifications [58] . In brief , differentially methylated cytosine ( DMC ) was identified if the p-value from the two-tailed Fisher's exact test was less than 0 . 05 . DMRs were discovered using a sliding-window approach with 200 bp-window sliding at 50 bp intervals . A region with more than 2 DMCs was selected as an anchor region . The boundary of each anchor region was defined based on the locations of first and last DMCs in the region . If the distance between two anchor regions is less than 100 bp , they will be merged into one large region . DMRs were finally identified based on the regions with ≥100 bp , ≥5 DMCs , and absolute methylation difference of 0 . 3 for CG , 0 . 15 for CHG or 0 . 10 for CHH . Twelve-day-old seedlings of C24 and rdm16 mutant complementation lines with native promoter-driven RDM16-3xFLAG or RDM16-3xHA transgene were used for chromatin immunoprecipitation ( ChIP ) assays . The ChIP assays were performed according to a published protocol [59] . ChIP products were eluted into 50 µl of TE buffer , and a 2 µl aliquot was used for each qPCR reaction . Nuclei were isolated from protoplasts of Arabidopsis young leaves according to a published protocol [60] . Nuclei were then fixed in 4% formaldehyde and applied to slides for immunolocalization assay as previously described [61] . After being treated with the blocking solution ( 3% BSA in PBS ) , the nuclei were then incubated with primary antibodies overnight at 4°C . Each primary antibody was properly diluted in the blocking solution . After washing the slides , secondary anti-mouse TRITC ( Invitrogen ) and anti-rabbit FITC ( Invitrogen ) were added and incubated at 37°C . Chromatin was counterstained with DAPI . Images were acquired by SPINNING DISK confocal microscopy and then analyzed with Volocity software . The RNA-seq and whole-genome bisulfite sequencing data used in this paper have been deposited in the National Center for Biotechnology Information Gene Expression Omnibus ( NCBI GEO ) ( http://www . ncbi . nlm . nih . gov/geo/ ) under accession numbers GSE44635 and GSE44417 , respectively . | Both plants and animals utilize cytosine DNA methylation as an important epigenetic mark to suppress transposable elements ( TEs ) , repeat sequences and genes , which is crucial for the genome integrity and development . In plants , de novo DNA methylation can be mediated by the RNA-directed DNA methylation ( RdDM ) pathway . Plants have also evolved a pathway for active DNA demethylation that is initiated by the ROS1 subfamily of 5-methylcytosine DNA glycosylases , to counteract the RdDM pathway to prevent undesirable silencing . In this study , we identified RDM16 , a new factor in the RdDM pathway . We show that RDM16 is a pre-mRNA splicing factor and its function in the regulation of DNA methylation and gene silencing is not through influencing siRNA levels or the expression or splicing of genes encoding known RdDM components , but likely through affecting Pol V transcripts . We also show that RDM16 preferentially affects ROS1 target loci . Together , our findings contribute to the understanding of RdDM and its interactions with ROS1-mediated DNA demethylation . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [] | 2013 | A Pre-mRNA-Splicing Factor Is Required for RNA-Directed DNA Methylation in Arabidopsis |
Eukaryotes regulate gene expression and other nuclear processes through the posttranslational modification of histones . In S . cerevisiae , the mono-ubiquitylation of histone H2B on lysine 123 ( H2B K123ub ) affects nucleosome stability , broadly influences gene expression and other DNA-templated processes , and is a prerequisite for additional conserved histone modifications that are associated with active transcription , namely the methylation of lysine residues in H3 . While the enzymes that promote these chromatin marks are known , regions of the nucleosome required for the recruitment of these enzymes are undefined . To identify histone residues required for H2B K123ub , we exploited a functional interaction between the ubiquitin-protein ligase , Rkr1/Ltn1 , and H2B K123ub in S . cerevisiae . Specifically , we performed a synthetic lethal screen with cells lacking RKR1 and a comprehensive library of H2A and H2B residue substitutions , and identified H2A residues that are required for H2B K123ub . Many of these residues map to the nucleosome acidic patch . The substitutions in the acidic patch confer varying histone modification defects downstream of H2B K123ub , indicating that this region contributes differentially to multiple histone modifications . Interestingly , substitutions in the acidic patch result in decreased recruitment of H2B K123ub machinery to active genes and defects in transcription elongation and termination . Together , our findings reveal a role for the nucleosome acidic patch in recruitment of histone modification machinery and maintenance of transcriptional integrity .
In eukaryotes , transcription and other nuclear processes take place in the context of chromatin . The basic unit of chromatin is the nucleosome , which consists of approximately 147 base pairs of DNA wrapped around a histone octamer , containing two copies of each of the four core histone proteins: H2A , H2B , H3 , and H4 [1] . Histones are decorated with posttranslational modifications , which can alter chromatin architecture and recruit a wide range of proteins to the genome , thus regulating all chromatin transactions [2] . In addition to their intrinsic effects on modulating the chromatin template , certain histone modifications can promote other histone modifications , either on the same histone ( cis-regulation ) or on a different histone ( trans-regulation ) in a process termed “histone crosstalk” [3] . The monoubiquitylation of H2B on lysine 123 ( H2B K123ub ) in S . cerevisiae is associated with active gene transcription , impacts global nucleosome occupancy and plays important roles in transcription elongation , telomeric silencing , DNA replication , and DNA repair [4] . In yeast , this modification is catalyzed by the ubiquitin-protein ligase Bre1 and the ubiquitin-conjugating enzyme Rad6 [5–7] . In humans , the analogous lysine , H2B K120 , is ubiquitylated by RNF20/RNF40 and RAD6A/RAD6B [8 , 9] . In one of the best-studied examples of histone crosstalk , H2B K123ub is required for other histone modifications associated with active transcription: H3 K4 and H3 K79 di- and tri-methylation [10–12] . H3 K4 dimethylation , which is enriched at the 5'-ends of coding regions , and H3 K4 trimethylation , which is associated with active promoters , regulate histone acetylation patterns on genes by directing the recruitment of histone acetyltransferases and histone deacetylases [13] . H3 K79 methylation occurs across active genes , and dimethylation of this residue locally alters the nucleosome surface [14 , 15] . All of these histone modifications are conserved in higher eukaryotes , and disruption of these modifications can result in a range of human diseases , including cancer [16] . In addition to Rad6 and Bre1 , several protein complexes that regulate transcription elongation and nucleosome dynamics are required for wild-type levels of H2B K123ub . These include the Bur1-Bur2 cyclin-dependent kinase complex and the FACT histone chaperone complex [17–19] . Additionally , the Polymerase Associated Factor 1 complex ( Paf1C ) , which travels with RNA pol II and Spt5 during transcription elongation , promotes H2B K123ub through the Rtf1 subunit of the complex [20–23] . While protein complexes that promote H2B K123ub have been identified , little is known about how the nucleosome itself promotes H2B K123ub . We previously reported that the ubiquitin-protein ligase Rkr1/Ltn1 is required for the viability of yeast cells that lack the RTF1 gene or harbor an amino acid substitution for H2B K123 that prevents ubiquitylation ( H2B-K123R ) [24] . Rkr1/Ltn1 associates with ribosomes and degrades nonstop proteins [25 , 26] . The genetic interactions between rtf1∆ , H2B-K123R , and rkr1∆ suggest a requirement for the quality control functions of Rkr1 in the absence of an intact H2B ubiquitylation pathway . We reasoned that the negative genetic interactions between rkr1∆ and H2B-K123R could be exploited to identify histone residues that are required for H2B K123ub . Using a genetic screen , we identified H2A and H2B residues required for proper H2B K123ub and downstream histone modifications . Many of these residues map to the acidic patch on the surface of H2A . We found that amino acid substitutions in the acidic patch cause defects in the recruitment of the H2B K123ub machinery to active genes , an accumulation of read-through transcripts , and altered transcription elongation efficiency in vivo . Interestingly , the substitutions differentially impact histone modifications downstream of H2B K123ub . Therefore , while the H2A acidic patch residues functionally converge in regulating H2B K123ub , they diverge in regulating downstream histone modifications . Our data reveal a requirement for the nucleosome acidic patch in H2B K123ub and argue that this exposed nucleosome surface serves as an important protein docking site in which individual residues uniquely contribute to the regulation of histone modifications and gene expression .
To identify histone residues required for H2B K123ub in S . cerevisiae , we screened a comprehensive histone mutant library [27] for alanine substitutions in H2A and H2B that cause synthetic lethality or sickness when combined with a deletion of the RKR1 gene . We previously showed that rkr1∆ is synthetically lethal in strains carrying H2B-K123R as the only form of H2B [24] . Using a plasmid shuffle strategy , HIS3-marked hta1-HTB1 or HTA1-htb1 plasmids from the library were transformed into a rkr1∆ deletion strain , replacing a URA3-marked plasmid carrying wild-type copies of HTA1 and HTB1 . The URA3-marked wild-type plasmid was counter-selected on medium containing 5-fluoroorotic acid ( 5-FOA ) . Relative to their effects on a strain containing a wild-type RKR1 gene , nine histone mutant plasmids caused enhanced growth defects in the rkr1∆ background ( Fig 1A ) . Eight of the amino acid substitutions were located in H2A , and one was H2B-K123A ( Fig 1A ) . Identification of H2B-K123A served as a validation of our screen . Many of the residues identified in our screen cluster within the nucleosome acidic patch ( Fig 1B and 1C ) . The acidic patch serves as a binding site for several proteins , including the H4 tail of neighboring nucleosomes [1 , 28–30] . In addition to those in the acidic patch , two residues , L86 and H113 , reside near the docking domain of H2A [1] . To test if the amino acid substitutions in H2A cause H2B K123ub defects , we assessed global H2B K123ub levels by western blot analysis . Because the plasmids in the H2A and H2B mutant libraries encode FLAG-tagged H2B [27] , we initially used anti-FLAG western blots to distinguish H2B K123ub from unmodified H2B as a super-shifted band . We subsequently turned to a commercial antibody against human H2B K120ub , which can specifically detect yeast H2B K123ub [31] ( Fig 2 ) . Surprisingly , this antibody did not recognize FLAG-tagged H2B K123ub to the same degree as untagged H2B K123ub in our strains , raising concerns that the FLAG tag could influence H2B ubiquitylation or our ability to detect this modification ( S1 Fig ) . Therefore , we removed the FLAG tag from all of the plasmids carrying hta1-HTB1 mutations identified in our screen , and we continued with these constructs for all experiments in this study . The western analysis revealed that all of the H2A mutants have reduced global H2B K123ub levels compared to the wild-type control strain; however , the different substitutions affect H2B K123ub levels to varying degrees ( Fig 2A ) . For example , there is a striking difference in H2B K123ub levels in strains harboring substitutions of the neighboring residues H2A-E65 and H2A-L66 ( Fig 2A , lanes 4 and 5 ) . Our result reflects the H2B K123ub defect previously observed for an H2A-L66A mutant [27]; however , with removal of the FLAG tag , we now detect a defect in H2B K123ub in the H2A-E65A mutant as well . To measure chromatin-associated levels of H2B K123ub , we performed chromatin immunoprecipitation ( ChIP ) analysis of H2B K123ub and total H2B at active genes ( PYK1 and PMA1 ) and , as a control , at a non-transcribed region ( TELVI ) . We normalized levels of H2B K123ub to levels of total H2B to correct for any defects in H2B occupancy ( Fig 2B ) . For these ChIP analyses , and most other experiments in this study , we focused our efforts on H2A residues E57 , E65 , L66 , L86 , and E93 , because residues F26 and L94 are buried within the protein core of the nucleosome and could be impacting H2B K123ub levels indirectly ( Fig 1A ) . We also chose not to focus on H113 , because it is not conserved in higher eukaryotes . In agreement with the western analyses , the ChIP assays revealed reduced H2B K123ub levels on active genes in the H2A mutant strains ( Fig 2B ) . However , gene-specific defects are evident . For example , the E57A substitution causes an H2B K123ub defect at PYK1 but not ADH1 or PMA1 ( Figs 2B and S2 ) . Together these data demonstrate that the nucleosome acidic patch promotes H2B K123ub globally and at specific genes . Previous studies have shown that H2B K123ub is required for proper histone occupancy [32 , 33] , that the docking domain of H2A is important for the association of H2A and H2B with H3 and H4 [33–35] , and that the acidic patch lies at the interface of H2A and H2B [1 , 35] . Therefore , we examined global and local levels of histones by western analysis and ChIP , respectively ( Fig 3 ) . Global levels of H2B , H3 , and H2A were unaffected in the mutants , with two exceptions ( Fig 3A ) . The two exceptions , H2A-E93A and H2A-L94A , were detected at levels that were lower than wild-type H2A , indicating a potential defect in the expression , stability , or antibody recognition of these H2A mutant proteins . H2B , H2A , and H3 occupancy levels were assessed at both the highly transcribed gene PYK1 and a non-transcribed telomeric region using ChIP analysis ( Fig 3B–3D ) . Four of the alanine substitutions in H2A resulted in lower occupancy levels of H2B at PYK1 ( Fig 3B ) . H2A occupancy was not as drastically affected in the mutant strains; however , the signals for H2A-E57A and H2A-E93A enrichment were reduced at all loci tested ( Fig 3C ) . For H2A-E93A , this could be due to reduced H2A protein levels or reduced immunoreactivity ( Fig 3A ) . H3 occupancy levels at PYK1 were also slightly affected in some of the mutant strains , particularly at the 5’ end of the gene ( Fig 3D ) . Importantly , the reduced histone occupancy levels do not account for the reduced H2B K123ub levels in the H2A mutant strains , as we have normalized the H2B K123ub levels to total histone levels in our assays ( Fig 2 ) . As an alternative measure of chromatin integrity in the histone mutant strains , we used northern analysis to monitor transcription of the SER3 and FLO8 genes , which can serve as sensitive reporters of defects in chromatin structure [36–38] . In rich media , SER3 expression is repressed by transcription-coupled nucleosome assembly over its promoter via transcription of a noncoding RNA , SRG1 [39 , 40] . Mutations in the genes encoding the histone chaperones Spt6 and Spt16 lead to strong derepression of SER3 without decreasing SRG1 transcription [39] . Relative to the temperature-sensitive alleles spt6-1004 and spt16-197 , the H2A substitutions identified in our screen do not cause strong derepression of SER3 , suggesting that transcription-coupled nucleosome occupancy is largely intact over SRG1 ( Fig 3E ) . Cryptic initiation can occur when cryptic promoters within coding regions are unveiled by perturbations in nucleosome occupancy or histone modifications [36–38] . To assess cryptic initiation in the H2A mutants , we performed northern analysis of the FLO8 gene , using spt6-1004 and spt16-197 as positive controls for cryptic initiation ( Fig 3E ) . Relative to these control strains , the H2A mutants generate only very low levels of cryptic transcripts at FLO8 ( Fig 3E ) . Together , these data suggest that , although histone occupancy defects can be detected , chromatin structure is not grossly impaired in the H2A mutants . H2B K123ub is required for downstream histone modifications , including H3 K4 di- and tri-methylation ( H3 K4me2/3 ) , catalyzed by Set1 , and H3 K79 di- and tri-methylation ( H3 K79me2/3 ) , catalyzed by Dot1 [10 , 11 , 13] . We therefore asked whether the H2A substitutions also cause defects in modifications downstream of H2B K123ub , using western analysis . Surprisingly , although all of the H2A mutants identified in our screen have reduced H2B K123ub levels , we observed a range of defects in H3 methylation ( Fig 4 ) . For example , two substitutions , F26A and H113A , cause no apparent defects in global H3 K4 or K79 methylation , despite dramatically reducing H2B K123ub levels ( Figs 2A and 4A ) . In contrast , the E65A and L66A substitutions greatly reduce H3 K4 methylation and partially reduce H3 K79 methylation even though their effects on H2B K123ub levels are quite different ( Figs 2A and 4A ) . Substitution of residues E93 and L94 to alanine resulted in a strong H3 K79 methylation defect and only slight defects in H3 K4 methylation ( Fig 4A ) . Thus , E93 and L94 appear to selectively impact H3 K79 methylation . To determine the levels of H3 methylation on chromatin , we performed ChIP analysis of H3 K79me2/3 and H3 K4me3 at PYK1 , PMA1 and TELVI in the H2A mutant cells and normalized the data to total H3 occupancy levels ( Fig 4B ) . The modification defects observed by ChIP mirror the global H3 methylation defects visualized by western analysis with slight differences being likely due to differences in histone occupancy levels , which were taken into account in the ChIP assay . Our results indicate that the H2A residues play unique roles in regulating histone modifications dependent on H2B K123ub . To test whether the H2A substitutions confer other histone modification defects potentially through a general change in nucleosome structure , we performed western analysis of Set2-catalyzed H3 K36me2 and K36me3 , modifications that are not strongly dependent on H2B K123ub [12 , 20] . None of the H2A mutants exhibited defects in H3 K36 methylation ( Fig 4A ) . This is in agreement with previous work , which identified a distinct nucleosome surface required for H3 K36 methylation [41] , and the idea that the H2A substitutions identified in our screen are largely specific to the H2B K123ub cascade . Previous studies have shown that H3 K4me3 and H2B K123ub are required for proper transcription termination of small nucleolar RNAs ( snoRNAs ) through the Nrd1-Nab3-Sen1 pathway [42–44] . However , little is known about how these histone modifications or other nucleosome residues affect transcription termination . To assess transcription termination in our mutants , we performed RT-qPCR analysis on four snoRNA genes that are affected by histone modifications [44] . For these assays , we used probes that hybridize to the intergenic region between the snoRNA gene and the downstream gene . Detection of a PCR product is a measure of transcription in the region downstream of the snoRNA terminator ( S3 Fig ) . The RT-qPCR analysis indicates that the H2A acidic patch residues are required for proper transcription termination at the four snoRNA loci ( S3 Fig ) . Previous work described snoRNA termination to be differentially sensitive to disruption of H2B K123ub: SNR47 required H2BK123ub for proper termination whereas SNR48 was relatively insensitive to the absence of this mark [44] . The mutants identified in our screen , which all have abrogated H2B K123ub , have termination defects at both loci , indicating that the mechanistic basis for read-through of these terminators could be downstream of H2B K123ub ( S3 Fig ) . H2B K123ub is a transient histone modification; therefore one possible explanation for reduced H2B K123ub levels in the H2A mutants could be through decreased stability of the mark through the enhanced action of a ubiquitin-specific protease . The removal of H2B K123ub is due to the actions of two ubiquitin-specific proteases Ubp8 and Ubp10 [45–48] . To test whether the H2B K123ub deficiency observed in the H2A mutants is through decreased stability of the modification , we performed western blot analysis of H2B K123ub levels in strains that contain the H2A substitutions and are deleted for UBP8 . Upon deletion of UBP8 , the fold recovery of H2B K123ub levels was comparable to that of wild-type cells for the H2A-L86A and H2A-E93A mutants , suggesting that the H2B K123ub defect in these mutants is at least partially due to decreased stability of the mark ( Fig 5A and 5B ) . For the H2A-E57A , H2A-E65A , and H2A-L66A mutants , deletion of UBP8 did not fully rescue H2B K123ub levels ( Fig 5A and 5B ) . The most drastic effect was that of H2A-L66A , where little to no H2B K123ub was restored . Therefore , for these mutants , and especially H2A-L66A , the defect in H2B K123ub is likely due to a failure of the ubiquitylation machinery to fully establish the mark . An alternative , but not mutually exclusive explanation , is that E57 , E65 , and L66 could form a surface required for Ubp8 function or recruitment , as these three residues reside near each other on the nucleosome structure ( Fig 1B ) . To test the extent to which H2B K123ub and downstream H3 methylation events are coupled in the H2A mutant strains , we measured H3 K4me3 and H3 K79me2/3 levels in the presence and absence of UBP8 . Upon deletion of UBP8 , H3 K4me3 and H3 K79me2/3 increased in the wild-type strain and in the H2A-L86A mutant ( Fig 5C ) . When normalized to total H3 levels , an increase in H3 K4me3 and K79me2/3 levels was not detected in the H2A-E57A mutant upon deletion of UBP8 , which is consistent with the poor recovery in H2B K123ub in this strain ( Fig 5 ) . This result suggests there is a correlation between K123ub and downstream marks in the H2A-E57A mutant when UBP8 is deleted . Similarly , for the H2A-L66A mutant , no recovery of the methyl marks was observed in the ubp8∆ background , which corresponds to the severe defect in K123ub in this mutant . This observation is consistent with the idea that the establishment of H2B K123ub is the primary defect in this mutant . For the E65A mutant , H3 K4me3 levels were extremely low in both the presence and absence of Ubp8 , even though H2B K123ub levels were substantially recovered in the ubp8∆ background . This observation suggests that the E65A substitution prevents proper H3 methylation possibly by disrupting a functional interaction with the Set1/COMPASS complex . Finally , in agreement with our western and ChIP results ( Fig 4 ) , the E93A mutant appears most defective in supporting H3 K79 methylation , as deleting UBP8 elevated H3 K4me3 levels to a greater extent than H3 K79me2/3 levels in this strain . In addition to decreased stability of the ubiquitylation mark conferred by Ubp8 , the reduction in histone modification levels in the H2A mutants could be due to impaired recruitment of the modification enzymes required for the H2B K123ub cascade , such as the ubiquitin-protein ligase Bre1 . To analyze the effects of the H2A substitutions on recruitment of Bre1 to actively transcribed genes , we performed ChIP analysis of HSV-tagged Bre1 ( Fig 6A ) . All five of the H2A mutants tested showed reduced recruitment of HSV-Bre1 to PYK1 and PMA1 , particularly at their 5’ ends ( Fig 6A ) . With the exception of the H2A-E57A mutant , Bre1 occupancy was also reduced at ADH1 ( S2 Fig ) . As expected , HSV-Bre1 levels at the non-transcribed TELVI region were similar to those of the untagged control strain . Also in agreement with previous observations [7] , Bre1 levels at the 5’ ends of PMA1 and PYK1 were higher than those at the 3’ ends of the genes . To determine whether reduced levels of HSV-Bre1 could account for the reduced HSV-Bre1 occupancy in the H2A mutant strains , we performed western analysis . Our results show that total HSV-Bre1 levels in the H2A mutants are similar to those in a wild-type strain ( S4 Fig ) . These results indicate that residues in the H2A acidic patch are required for proper Bre1 recruitment to active genes . The Paf1C subunit , Rtf1 , has been implicated in the recruitment of the H2B ubiquitylation machinery during transcription [49] . We therefore used ChIP analysis to test whether the H2A residues that are important for Bre1 recruitment are also important for Rtf1 occupancy on active genes . Our ChIP results demonstrate a significant reduction in Rtf1 levels at PYK1 , PMA1 , and ADH1 in the H2A mutant strains ( Figs 6B and S2 ) . To rule out the possibility that the reduced Rtf1 occupancy is a result of lower protein levels , we measured global Rtf1 levels by western analysis . This analysis showed that Rtf1 levels are unaffected in the H2A mutants , indicating that reduced Rtf1 expression is not the cause of the H2B K123ub defect ( S4 Fig ) . Overall , the occupancy levels of HSV-Bre1 and Rtf1 correlated with H2B K123ub levels in some cases but not others . For example , the H2A-E57A mutant shows reduced HSV-Bre1 and Rtf1 occupancy but normal levels of H2B K123ub at the PMA1 locus . It is possible that small levels of Bre1/Rad6 and Rtf1 are sufficient to promote H2B K123ub at PMA1 in this mutant . Alternatively , decreased Ubp8 levels or activity could compensate for reduced Bre1 recruitment . We attempted to test this idea by ChIP but were unable to reliably measure Ubp8 occupancy in our strains . We previously demonstrated that recruitment of Paf1C to coding regions is mediated through a direct physical interaction between Rtf1 and the elongation factor Spt5 [50 , 51] . Therefore , it is possible that the lower Rtf1 and Bre1 occupancy levels in the H2A mutant strains reflect impaired recruitment of the transcription elongation machinery . To test this idea , we performed ChIP analysis of Spt5 , Spt16 , and Pol II occupancy at PYK1 , PMA1 , and TELVI in the histone mutant strains ( Figs 6C and 6D and S5 ) . We observed gene and allele specific defects in Spt5 occupancy , with the E57A , E65A , and L66A substitutions causing reduced Spt5 occupancy particularly at PYK1 . However , the levels of Spt5 occupancy largely mirrored Pol II occupancy levels , suggesting that the effects of the H2A substitutions on Spt5 recruitment are likely to be indirect . We also assessed the effects of the H2A substitutions on recruitment of the FACT complex member Spt16 ( Fig 6D ) , which is required for proper histone occupancy and H2B K123ub [18 , 19 , 52] . Interestingly the substitution within the docking domain , H2A-L86A , of the nucleosome exhibited increased Spt16 occupancy at all tested loci . In contrast , substitutions E57A and E93A led to reduced Spt16 occupancy , suggesting that , for these H2A mutants , a defect in Spt16 recruitment may be a contributing factor to the reduced H2B K123ub levels and lower histone occupancy levels ( Fig 3B and 3C ) . Global levels of Spt5 and Spt16 were not strongly affected , as judged by western analysis ( S5 Fig ) . Because the histone mutants have defects in H3 K4 methylation ( Figs 4 and 5C ) , the acidic patch residues may be required for recruitment of the H3 K4 methyltransferase Set1 . To test this , we performed ChIP analysis of HSV-tagged Set1 in the H2A mutants ( Fig 6E ) . With the exception of E57A , all of the substitutions affect occupancy of HSV-Set1 . However , after normalizing the H3 K4me3 occupancy levels to H3 occupancy levels , only the E65A and L66A substitutions cause a strong defect in H3 K4me3 ( Fig 4C ) . We thus conclude that HSV-Set1 recruitment may be impacted by the occupancy levels of H2B K123ub and H3 . For the E65A mutant , the severity of the H3 methylation defect and lack of restoration of H3 K4me3 upon deletion of UBP8 suggests that E65 may play a more direct role in promoting H3 K4 methylation . We did not observe a reduction in HSV-Set1 levels in the H2A mutants ( S6 Fig ) , as has been reported to occur when H3 K4 cannot be methylated [53 , 54] . It is possible that the H2A mutants lack the ability to regulate Set1 levels . Because the H2A mutants exhibit reduced levels of transcription elongation-coupled histone modifications , we asked whether the acidic patch substitutions alter the efficiency of transcription elongation . To assess transcription elongation efficiency in vivo we used a well-established galactose-controlled system to shut off transcription of a gene and measure occupancy of Pol II during the last wave of transcription [55–57] . This system incorporates the GAL1 promoter upstream of the non-essential gene YLR454W . Cells were grown in 2% galactose to activate the gene and 2% glucose was added to the cultures to prevent further initiation events . Samples were taken at different time points to determine "snap-shots" of Pol II density at four regions of YLR454W by ChIP ( Fig 7A ) . In wild-type cells , Pol II rapidly cleared the YLR454W coding region , as previously described [55–57] ( Fig 7B ) . In the H2A mutants , however , the rate and/or processivity of Pol II elongation was reduced . The most dramatic effect was observed with the H2A-L66A mutant , where Pol II density persisted at the 4 Kb and 8 Kb locations relative to the wild-type kinetics ( Fig 7C ) . The H2A-E65A mutant also exhibited a delay in Pol II passage , with occupancy persisting at multiple locations throughout the time course ( Fig 7D ) . The H2A-E93A mutant exhibited a slightly different and more modest elongation defect ( Fig 7E ) . Collectively these data reveal an important role for the nucleosome acidic patch in promoting efficient transcription elongation . Because the H2A acidic patch mutants have defects in H2B K123ub , we wanted to determine whether the in vivo elongation defects correlated with the loss of H2B K123ub . To begin to address this , we performed a similar analysis on H2B-K123R cells ( Fig 7F ) . Interestingly , Pol II elongation efficiency was reduced in the H2B-K123R mutant , as indicated by persistent enrichment toward the 3’ end of the gene . These data indicate that residues within the acidic patch , at least partly through their role in promoting H2B K123ub , are important for transcription elongation efficiency in vivo .
In this study , we exploited a genetic interaction between the H2B ubiquitylation pathway and the protein quality control factor Rkr1 to identify residues in H2A and H2B that are required for H2B K123ub . We identified eight residues in H2A that , when changed to alanine , cause defects in H2B K123ub ( Fig 2 ) . Most of these residues map to the acidic patch on the nucleosome ( Fig 1B ) , which plays critical roles in several important nuclear processes . Indeed , as shown through structural studies , the acidic patch serves as a direct binding platform on the nucleosome for a variety of proteins that affect transcription , chromatin structure , and chromosome segregation . These proteins include the Latency-Associated Nuclear Antigen ( LANA ) peptide from Kaposi’s sarcoma virus , the Regulator of Chromatin Condensation 1 protein ( RCC1 ) , the Bromo-Associated Homology ( BAH ) domain of Sir3 , and the centromere binding protein CENP-C [28 , 30] . Additionally , as shown through functional studies and a recently published structure of the Polycomb Repressive Complex 1 ubiquitylation module in complex with a nucleosome , the acidic patch interacts with ubiquitin-protein ligases that target H2A [58–60] . Despite the importance of H2B K123ub in regulating gene expression , nucleosome stability , and genic patterns of histone methylation and acetylation , little is known about how the enzymatic machinery for H2B K123ub interfaces with the nucleosome . In a recent study , a basic region of the RING domain of Bre1 was shown to be important for interacting with the nucleosome [61] . Here , we show that nucleosome acidic patch mutants have impaired chromatin occupancy of the ubiquitin-protein ligase Bre1 and the Paf1C subunit Rtf1 . The mechanism by which Rtf1 is required for H2B K123ub is largely undefined , although a recent study indicated a role for Rtf1 in stabilizing Bre1 protein levels [31] . In our H2A mutant strains , global protein levels of Bre1 are similar to those in a wild-type strain . This observation , together with our ChIP studies on Bre1 and Rtf1 , suggests that the nucleosome acidic patch plays an active role in promoting H2B K123ub . A previous study found that the N-terminus of H2A , the H2A repression ( HAR ) domain , is also required for H2B K123ub . However , recruitment of the H2B K123ub machinery was not affected in the H2A N-terminal tail mutant [62] . It is possible , then , that the acidic patch could recruit the H2B K123ub machinery to chromatin , potentially through a direct interaction with Bre1 and/or Rtf1 , while the HAR domain stimulates enzyme activity . In light of previous work showing that Paf1C recruitment is governed by a direct physical interaction between Rtf1 and the phosphorylated C-terminal region of the elongation factor Spt5 [50 , 51 , 63 , 64] , we were surprised that the H2A substitutions identified in our screen caused a loss in Rtf1 occupancy without a corresponding loss in Spt5 recruitment . However , it was recently shown that the human homolog of Bre1 , RNF20/40 , promotes recruitment of PAF1 to chromatin in human cells [65] . In addition , binding of human Paf1 to histone-like proteins and nucleosomes has been reported [66 , 67] . These observations align with our results and indicate that multiple interactions can mediate or stabilize the interaction between Paf1C and chromatin . Alternatively , given the importance of Spt5 phosphorylation in mediating the interaction between Rtf1 and Spt5 [50 , 51 , 63 , 64] , it is also possible that the H2A mutants are indirectly affecting Spt5 phosphorylation . Finally , we also note that Rtf1 recruitment defects could be due to the combined effect of the individual , and relatively modest , defects in Pol II , Spt5 , and Spt16 occupancy ( Figs 6 and S5 ) . The function of the ubiquitin-specific protease Ubp8 also appears to be affected by substitutions within the acidic patch ( Fig 5 ) . A recent study suggested that the acidic patch residue H2A-Y58 promotes H2B K123ub through regulating Ubp8 , as deletion of UBP8 rescued H2B K123ub in an H2A-Y58F mutant [68] . The H2A-Y58A is a lethal substitution in yeast and could not be isolated in our screen [27] . In our study , deletion of UBP8 rescued H2B K123ub to some degree in most of our mutants , which suggests that these mutants have defects in both ubiquitylating H2B-K123 and in stabilizing the mark ( Fig 5A ) . For the H2A-L66A mutant , the nearly complete absence of H2B K123ub in the presence or absence of UBP8 suggests that little ubiquitin is placed on H2B-K123 such that removal of UBP8 makes little to no difference in this mutant . The H2A residues we identified are required for H2B K123ub-dependent H3 methylation ( Fig 4 ) . Interestingly , some mutants exhibited defects in only H3 K4 methylation or H3 K79 methylation , while others had defects in both , despite all having reduced H2B K123ub levels . These data suggest that individual residues within the acidic patch promote methylation through separate mechanisms . Substitution of neighboring residues , H2A-E65A and H2A-L66A , differentially impacted H2B K123ub levels , but both mutants had undetectable levels of H3 K4 methylation ( Fig 4 ) [27] . It is possible that the methylation defects caused by the L66A substitution are largely due to a severe defect in the establishment of H2B K123ub in this mutant , similar to the effect of the H2B K123R mutant [69] . In contrast , the H3 K4 methylation defect of the H2A-E65A mutant may stem primarily from the reduced recruitment and/or activation of Set1 . For the H2A-E65A mutant , we noted a lack of recovery of H3 K4me3 and H3 K79me2/3 when H2B K123ub levels were increased through the deletion of UBP8 ( Fig 5C ) . This observation suggests that E65 is important for coupling H2B K123ub to downstream H3 methylation events . Interestingly , substitution of other residues near H2B K123 has been shown to uncouple H3 methylation from H2B K123ub . For example , H2B R119 and T112 , when mutated , increase H2B K123ub levels but decrease H3 K4me3 levels [70] . The severe deficiency in H3 K79me2/3 observed in the H2A-E93A mutant ( Figs 4 and 5C ) presents the intriguing possibility that this residue may interact with Dot1 to promote H3 K79 methylation . It is unlikely that the H3 K79 methylation defect detected in the H2A-E93A mutant is solely due to its defect in H2B K123ub , because when H2B K123ub levels are increased in the absence of UBP8 , the increase in H3 K79 methylation is very slight ( Fig 5C ) . Interestingly , the basic patch in the H4 tail is required for Dot1 methylase activity , but not for Dot1 recruitment [71] . Since the H4 tail interacts with the acidic patch of the nucleosome [1 , 29] , one explanation for the H3 K79me2/3 defect could be that E93 is required for recruitment of Dot1 , while the H4 tail stimulates Dot1 activity . Further supporting growing indications that chromatin structure is important for proper transcription termination through the NNS pathway , the H2A mutants tested exhibited transcriptional readthrough at four SNR genes ( S3 Fig ) . The magnitude of the transcriptional defect does not correlate strictly with the loss of any particular histone modification , suggesting that this phenotype may be sensitive to the combinatorial loss of several modifications and possibly other factors , such as histone occupancy and Spt16 recruitment ( Figs 3 and 6D ) . Regardless of the mechanism , the increased levels of aberrant transcripts in the H2A mutants could provide a rationale for the synthetic growth defects observed in H2A mutants lacking RKR1 . Rkr1 is a protein quality control factor that is involved in the degradation of aberrant proteins , including those that extend past stop codons [25 , 26] . The elevated synthesis of aberrant proteins , potentially as a consequence of improper transcription in the H2A mutants , could have lethal consequences for the cell [72] . The negative genetic interaction between rkr1∆ and the histone mutants suggests that the consequences of disrupting the acidic patch extend beyond chromatin and transcription . We assayed the effects of the H2A substitutions on transcription elongation through analysis of Pol II density during the last wave of transcription across GAL1-YLR454W . In this assay , the H2A-L66A mutant exhibited a strong defect in elongation efficiency and most closely mimicked the behavior of the H2B-K123R mutant . These data support the view that H2A-L66A phenocopies H2B-K123R for loss of H2B K123ub and its consequences . The H2A-E93A and H2A-E65A mutants also exhibited impaired elongation , although not to the same degree as the H2A-L66A and H2B-K123R mutants . Given the differential effects of the H2A substitutions on the histone modification levels in the cells , differences in Pol II elongation efficiency were not unexpected . Taken together , these data indicate that H2B K123ub and its effects on downstream histone modifications and nucleosome stability are important for efficient Pol II passage through chromatin . Combined , our data support a new role for the nucleosome acidic patch in transcription , specifically through the proper recruitment and/or activation of proteins that control H2B K123ub and downstream methylation events on H3 . The mutations that disrupt this patch impair several transcription-related processes , including the modification of histones , recruitment of transcriptional machinery , the efficient passage of Pol II through chromatin , and transcription termination ( Fig 8 ) . Many of these transcriptional defects likely stem from the pleiotropic effects of losing the critical H2B K123ub mark . Together with recent structural studies , our results strongly suggest that the acidic patch is an interaction platform for proteins that modulate numerous chromatin transactions in eukaryotic cells . An exciting goal for future studies will be to understand how cells regulate access to this important region of the nucleosome .
The S . cerevisiae strains used in this study are listed in S1 Table and are isogenic to the strain FY2 , which is a GAL2+ derivative of S288C [73] . Yeast transformations were performed as previously described [74] ) . With noted exceptions , experiments were performed using the strain KY943 transformed with histone mutant plasmids . To replace wild-type histone plasmids with HIS3-marked mutant histone plasmids , transformants were sequentially passaged three times on SC-His medium containing 2% dextrose and 0 . 1% 5-FOA . Unless otherwise noted , for all experiments , yeast strains were grown in SC-His medium containing 2% dextrose . HSV-Bre1 and HSV-Set1 strains contain three chromosomally located HSV tags on the N-termini of the proteins [75] . These proteins were confirmed to have proper function and expression . Cells were grown to saturation at 30°C and washed with sterile water . Beginning with a cell suspension at a concentration of 1 X 108 cells/mL , cells were diluted serially four times by a factor of ten in water . Two microliters of each dilution were spotted on SC-His medium and SC-His medium containing 5-FOA . Plates were incubated at 30°C for three days . Site-directed mutagenesis ( Agilent ) with primers listed in S2 Table was performed to remove the sequence encoding the FLAG tag from plasmids obtained from the H2A and H2B mutant library [27] . Plasmid sequences were confirmed by DNA sequencing . Plasmid names are given in S3 Table . For western analyses other than those that measure H2B K123ub , yeast cells were grown to log phase ( 2–3 X 107 cells/mL ) and lysed by bead beating in trichloroacetic acid ( TCA ) , as described previously [76] . To make whole cell extracts for H2B K123ub analysis , cells were lysed in SUTEB buffer ( 10 mM Tris-HCl , pH 8 . 0 , 1% SDS , 8 M urea , 10 mM EDTA , pH 8 . 0 , and 0 . 01% bromophenol blue ) [43] . Proteins were resolved on SDS-polyacrylamide gels ( 15% polyacrylamide for histone westerns , 10% polyacrylamide for Rtf1 and HSV-Bre1 , and 8% polyacrylamide for HSV-Set1 , Spt5 , and Spt16 westerns ) and transferred to nitrocellulose membranes . For H2B K123ub western blot analysis , proteins were transferred to PVDF membranes . Membranes were incubated with primary antibodies and then with anti-mouse or anti-rabbit secondary antibodies ( GE Healthcare 1:5 , 000 dilution ) . Antibodies that recognize the following proteins or histone modifications were used: total histone H3 ( 1:30 , 000 dilution ) [43] , trimethylated H3 K4 ( H3 K4me3 ) ( Active Motif 39159 , 1:2 , 000 dilution ) , H3 K4me2 ( Millipore 07–030 , 1:2000 dilution ) , H3 K79me3 ( note: this antibody has been reported by the manufacturer to cross-react with H3 K79me2 , Abcam ab2621 , 1:2 , 000 dilution ) , H3 K36me2 ( Millipore 07–369 , 1:1000 dilution ) , H3 K36me3 ( Abcam ab9050 , 1:1000 dilution ) , H2A ( Active Motif , 39235 , 1:5 , 000 dilution ) , H2B ( Active Motif , 39237 , 1:5 , 000 dilution ) , HSV ( Sigma-Aldrich H6030 , 1:350 dilution ) , Spt5 ( gift from Grant Hartzog , 1:1000 dilution ) , Spt16 ( gift from Tim Formosa , 1:500 dilution ) , Rtf1 ( 1:5 , 000 dilution ) [77] , and glucose-6-phosphate dehydrogenase ( G6PDH ) ( Sigma-Aldrich A9521 , 1:30 , 000 dilution ) . An antibody against a human H2B K120ub-containing peptide ( Cell Signaling 5546 , 1:1000 dilution ) was used to detect the analogous modification in S . cerevisiae , H2B K123ub . Proteins were visualized using enhanced chemiluminescence substrate ( PerkinElmer ) and either a 440 CF digital imaging station ( Kodak ) or a ChemiDoc XRS digital imaging station ( BioRad ) . For western blot analysis , signals were quantified using ImageJ software and normalized to the loading control specified in the figure legend . The relative signal from the wild-type strain was set equal to one . Error bars represent standard error of the mean for three biological replicates ( SEM ) . Chromatin immunoprecipitation ( ChIP ) assays were performed with 250 mL of log-phase yeast cultures ( 1–2 X 107 cells/mL ) as previously described [78] . For histone ChIPs , sheared chromatin was incubated overnight at 4°C with antibodies specific to H2B , ( 0 . 5 μl , Active Motif , 39237 ) , human H2B K120ub ( 2 . 5 μl , Cell Signaling 5546 ) , H3 K4me3 ( 2 . 5 μl , Abcam ab8580 ) , H3 K79me2/3 ( 2 . 5 μl , Abcam ab2621 ) , or total H3 ( 5 μl ) [43] . Chromatin prepared from an H2B-K123R strain served as a specificity control for the human H2B K120ub antibody ( S7 Fig ) . For other ChIPs , chromatin was incubated overnight at 4°C with antibodies specific to Spt16 ( 1 μl , gift from Tim Formosa ) , Spt5 ( 1 μl , gift from Grant Harzog ) , or Rpb3 ( 2 . 5 μl Neoclone W0012 ) . Following incubation with the primary antibodies , chromatin was incubated for 2 hours at 4°C with Protein A-conjugated sepharose for all ChIPs , with the exception of Rpb3 ChIPs , for which chromatin was incubated with Protein G-conjugated sepharose ( 30 μl , GE Healthcare ) . For ChIP of HSV-Bre1 and HSV-Set1 , chromatin was incubated overnight at 4°C with an antibody specific to the HSV epitope ( 2 . 5 μl , Sigma-Aldrich H6030 ) , followed by incubation as described above . For ChIP of Rtf1 , chromatin was incubated overnight at 4°C with polyclonal antisera that recognizes Rtf1 [77] . DNA was purified ( Qiagen ) and analyzed by qPCR using Maxima SYBR ( Thermo ) and primers for the 5’ coding region of PYK1 ( amplicon: +253 to +346 relative to ATG ) , the 3’ coding region of PYK1 ( amplicon: +1127 to +1270 ) , the 5’ coding region of PMA1 ( amplicon: +214 to +319 relative to ATG ) , the 3’ coding region of PMA1 ( amplicon: +2107 to +2194 ) , or a telomeric region of chromosome VI ( chromosomal coordinates , 269495 to 269598 ) . Occupancy levels were calculated using the primer efficiency raised to the difference between input and immunoprecipitated Ct values . Presented data are an average of two technical replicates for each of three biological replicates . The error bars indicate the standard error of the mean ( SEM ) . Total RNA was isolated from log-phase yeast cultures ( 1–2 X 107 ) , and 20 μg of RNA were subjected to northern blot analysis as described previously [79] . Radiolabeled DNA probes were generated through random-prime labeling reactions of PCR templates . Membranes ( Gene Screen Plus , Perkin Elmer ) were incubated with radiolabeled DNA probes from PCR fragments of SCR1 ( amplicon: -163 to +284 relative to the TSS ) , SRG1 ( amplicon: -454 to -123 relative to SER3 ATG ) , SER3 ( amplicon: +111 to +1342 relative to ATG ) , and FLO8 ( amplicon: +1515 to + 2326 relative to ATG ) . See S2 Table for primer sequences . Signals were quantified using ImageJ software relative to the SCR1 loading control , with wild type set to one . For quantification of all northern blot analyses , signals were averaged for three independent biological replicates . Error bars represent standard error of the mean ( SEM ) . Total RNA was isolated as described above and then DNase treated using the Turbo DNA-free kit ( Ambion , AM1907 ) and RNase inhibitor ( Ambion , AM2682 ) . cDNA was generated using the RETROscript kit ( Ambion , AM1710 ) with random hexamers and oligo ( dT ) primers . Quantitative PCRs were performed as described above using primers specific for the regions downstream of snoRNAs ( S2 Table ) . Signals were analyzed using the ∆∆CT method with ACT1 used as the target gene [80] . For controls , reactions lacking reverse transcriptase or template were performed . The graphs show the results of three independent biological replicates . | Chromatin , a complex of DNA wrapped around histone proteins , impacts all DNA-templated processes , including gene expression . Cells employ various strategies to alter chromatin structure and control access to the genetic material . Nucleosomes , the building blocks of chromatin , are subject to a myriad of modifications on their constituent histone proteins . One highly conserved modification with important connections to human health is the addition of ubiquitin to histone H2B . H2B ubiquitylation modulates chromatin structure during gene transcription and acts as a master regulator for downstream histone modifications . The proteins that promote H2B ubiquitylation have been identified; however , little is known about how these proteins interface with the nucleosome . Here , we exploited the genetic tools of budding yeast to reveal features of the nucleosome that are required for H2B ubiquitylation . Our genetic screen identified amino acids on the nucleosome acidic patch , a negatively charged region on the nucleosome surface , as being important for this process . The acidic patch is critical for regulating chromatin transactions , and , in our study , we identified roles for the acidic patch throughout transcription . Our data reveal that the acidic patch recruits histone modifiers , regulates histone modifications within the H2B ubiquitylation cascade , and maintains transcriptional fidelity . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [] | 2015 | The Nucleosome Acidic Patch Regulates the H2B K123 Monoubiquitylation Cascade and Transcription Elongation in Saccharomyces cerevisiae |
Tuberculosis is the deadliest infectious disease worldwide . Although the BCG vaccine is widely used , it does not efficiently protect against pulmonary tuberculosis and an improved tuberculosis vaccine is therefore urgently needed . Mycobacterium tuberculosis uses different ESX/Type VII secretion ( T7S ) systems to transport proteins important for virulence and host immune responses . We recently reported that secretion of T7S substrates belonging to the mycobacteria-specific Pro-Glu ( PE ) and Pro-Pro-Glu ( PPE ) proteins of the PGRS ( polymorphic GC-rich sequences ) and MPTR ( major polymorphic tandem repeat ) subfamilies required both a functional ESX-5 system and a functional PPE38/71 protein for secretion . Inactivation of ppe38/71 and the resulting loss of PE_PGRS/PPE-MPTR secretion were linked to increased virulence of M . tuberculosis strains . Here , we show that a predicted total of 89 PE_PGRS/PPE-MPTR surface proteins are not exported by certain animal-adapted strains of the M . tuberculosis complex including M . bovis . This Δppe38/71-associated secretion defect therefore also occurs in the M . bovis-derived tuberculosis vaccine BCG and could be partially restored by introduction of the M . tuberculosis ppe38-locus . Epitope mapping of the PPE-MPTR protein PPE10 , further allowed us to monitor T-cell responses in splenocytes from BCG/M . tuberculosis immunized mice , confirming the dependence of PPE10-specific immune-induction on ESX-5/PPE38-mediated secretion . Restoration of PE_PGRS/PPE-MPTR secretion in recombinant BCG neither altered global antigenic presentation or activation of innate immune cells , nor protective efficacy in two different mouse vaccination-infection models . This unexpected finding stimulates a reassessment of the immunomodulatory properties of PE_PGRS/PPE-MPTR proteins , some of which are contained in vaccine formulations currently in clinical evaluation .
Tuberculosis is the deadliest infectious disease worldwide and is responsible for more than 1 . 7 million deaths per year [1] . Its causative agent , Mycobacterium tuberculosis , is a slow growing bacterium inherently resistant to many antibiotics . This problem is further exacerbated by rising levels of acquired drug resistance , resulting in multi-drug-resistant ( MDR ) and extensively-drug-resistant ( XDR ) strains of M . tuberculosis , which require treatment regimens of two years with low treatment success rates and severe side effects [1–3] . These worrying developments highlight the need for a successful vaccine , halting the transmission of tuberculosis [4] . The currently used vaccine is based on Mycobacterium bovis , attenuated through serial culture by Calmette and Guérin and therefore known as Bacille Calmette-Guérin ( BCG ) [5–7] . BCG is generally believed to protect relatively well against severe forms of disseminated tuberculosis in children , but is unable to induce full protection or halt transmission of M . tuberculosis in adolescents and adults [4 , 8 , 9] . Furthermore , even these protective traits are subject to controversy , which may be caused by the plethora of genomic mutations and recombination events that have accrued during the worldwide sub-culturing of the original BCG strain [5 , 6 , 10 , 11] . One possible reason for sub-optimal protection by BCG and other candidate vaccines is the absence or secretion defect of certain immunogenic proteins . M . tuberculosis secretes many proteins through its different secretion systems , including Sec-translocation ( Sec ) , Twin-arginine-translocation ( Tat ) , or Type VII secretion ( T7S ) systems [12 , 13] . M . tuberculosis possesses five different T7S systems called ESX-1 to ESX-5 [14] . The first T7S system to be discovered was ESX-1 , identified by the Region of Difference ( RD ) 1 deletion in BCG [15] , responsible for the loss of ESX-1-mediated secretion in this vaccine strain [16 , 17] . Substrates of the ESX-1 system are responsible for the rupture of mycobacterium-containing phagosomes and represent a major virulence factor of pathogenic mycobacteria [18–21] . Corresponding to this information , the expression of the ESX-1 secretion system in BCG increased protective activity , but was also associated with increased pathogenesis [22] . Interestingly , a recently developed recombinant BCG strain expressing ESX-1 of Mycobacterium marinum was able to induce cytosolic pattern recognition and better protective responses , without a significant increase in virulence [23] . Similarly , the vaccine candidate MTBVAC was recently shown to induce immune responses to selected ESX-1 substrates and this ability was found to be the major determinant of improved protective efficacy as compared to BCG [24] . While the ESX-1 system is the best studied T7S system in mycobacteria , the ESX-5 system has the largest repertoire of substrates [25–27] . The ESX-5 system is essential for slow-growing mycobacteria , because of its role in outer membrane permeability [26 , 28] . Therefore , this system is present and considered functional in BCG . The coding sequences of the potential substrates of the ESX-5 system together form almost 8% of the coding potential of the M . tuberculosis genome [29] . Most notable amongst the ESX-5 substrates are the PE and PPE proteins , named for the proline and glutamic acid residues in their N-terminal domains . Defined functions have been described for some PE-PPE proteins , such as the lipase LipY [30 , 31] and PPE10 , the latter of which is important for capsular integrity of M . marinum [32] . Furthermore , many studies have ascribed immunomodulatory functions to PE-PPE proteins , such as altering host cytokine responses by interaction with Toll-like receptors or inhibition of antigenic presentation [33–36] . However , most PE and PPE proteins have no known functions and their high degree of homology makes them difficult to study . The latter is especially true for the two most-recently evolved subgroups of ESX-5 substrates , i . e . the PE_PGRS and PPE-MPTR proteins . Both these sub-groups are characterized by their GC-rich DNA sequences , repetitive glycine-rich amino acid motifs and high molecular weight ranging up to ~365 kDa [27 , 29] . We recently identified the PPE protein PPE38 and its highly similar , duplicated variant PPE71 , as essential factors in the secretion of both the PE_PGRS and PPE-MPTR proteins , in both M . marinum and M . tuberculosis [37] . The genes encoding PPE38 and PPE71 are organized in a 4-gene locus that also includes the esxX and esxY genes ( Fig 1A ) , which however are not required for PE_PGRS secretion in M . tuberculosis strain CDC1551 [37] . Strains with naturally occurring , or engineered , loss-of-function mutations of the ppe38-locus were unable to secrete both PE_PGRS and PPE-MPTR proteins and were more virulent in a mouse infection model [37] . Indeed , deletion of the ppe38-locus occurred at the branching point of modern Beijing ( Lineage 2 ) strains and may have aided in their global dispersal [37] . Moreover , the ppe38-locus was previously shown to be a hypervariable genetic region and many strains within the M . tuberculosis complex ( MTBC ) have polymorphisms in this locus [38] . Such polymorphisms are often caused by recombination events involving IS6110 elements [38 , 39] . Insertion , homologous recombination , deletion of IS6110 copies and/or deletion of intervening sequences between two IS6110 elements can lead to overexpression or gene deletion/truncation events with possible effects on transmission or virulence [37 , 40–43] . The most well-known of the polymorphisms affecting the ppe38-locus is the deletion of the RD5 region from BCG and several other animal-adapted strains of the MTBC [38 , 44] . The biological impact of the RD5 deletion has been a controversial subject of research and has focused solely on the phospholipase C encoding genes plcABC . Deletion of plcABC was reported to either attenuate [45] or increase virulence of M . tuberculosis [46] . However , a more recent study of the plc-genes in different mouse and cellular models showed no relevant contribution of these genes to the virulence of M . tuberculosis [47] . Here , we investigated the effect of RD5-like polymorphisms of the ppe38-locus in a number of MTBC-branches and discovered that the RD5 deletions in animal-adapted strains and the BCG vaccine strains have profound effects on the repertoire of secreted substrates in these strains . Restoration of PPE38-dependent secretion results in a wider antigenic repertoire of BCG , whereby the identification of two immunogenic epitopes in one of the substrates , i . e . the PPE-MPTR protein PPE10 , has allowed us to monitor the immunological impact of the corresponding secretion characteristics on host immune responses .
The genetically most-distant tubercle bacilli are represented by the Mycobacterium canettii clade . This outgroup mirrors the genomic diversity likely present within the ancestor of M . tuberculosis before branching and clonal expansion of the MTBC [48] . Recent studies of M . canettii have improved our understanding of adaptations that have shaped the transition from an M . canettii-like ancestor into extant M . tuberculosis , such as the gain of surface hydrophobicity through loss of lipooligosaccharide production [49] and the apparent loss of the capacity to exchange chromosomal DNA in the MTBC [50] . Interestingly , the available genome sequence information of five M . canettii isolates revealed potential polymorphisms in the ppe38-locus [48] . While strains D , K and L all possessed copies of the ppe38 and ppe71 genes , the sequence of strain J in the database indicated the potential absence of ppe38 and ppe71 from the strain . Such a deletion would be expected to affect PE_PGRS secretion [37] . However , secretion analysis revealed that all 5 isolates secreted PE_PGRS proteins ( S1A Fig ) . Subsequent PCR analysis confirmed the presence of a complete ppe38-71 locus , similar to M . tuberculosis H37Rv , for all tested M . canettii strains , including strain J ( S1B Fig ) . It is likely that the sequence polymorphisms in the previously deposited dataset may have arisen due to automated sequence assembly-associated bio-informatic artefact , which is a known problem for this region [37 , 38] . Another interesting group of strains , which were reported to have major polymorphisms in the RD5/ppe38-locus , was recently described by Lee et al . [51] . The Inuit population of the Nunavik region in Canada is affected by high levels of tuberculosis incidence . The majority of all cases in this cohort were shown to have resulted from the introduction of a single , particular M . tuberculosis strain , about one century ago . This sublineage was defined by genomic deletions , two of which affect the RD5/ppe38 locus . A 5 , 759 bp RD5-like deletion ( CDC1551-D17 ) removed the three phospholipase C genes plcABC and truncated ppe38 ( Fig 1A ) . The other ppe gene in this locus , ppe71 ( mt2422 ) , was reported to be affected by 22 bp frameshift deletion ( Fig 1A ) [51] . Reinvestigation of the sequence of ppe71 by inspection of the whole genome sequence data , and by PCR and Sanger sequence analysis revealed that this deletion was in fact a 21 bp deletion causing a 7 amino-acid deletion ( Amino acids 354-MGGAGAG-361 ) relative to PPE71 of M . tuberculosis H37Rv , but not a frameshift . This deletion has been previously described to occur also in other strains of M . tuberculosis , including CDC1551 ( MT2422 - http://www . genome . jp/dbget-bin/www_bget ? mtc:MT2422 ) [38] . To test whether the RD5-like polymorphism negatively affects PE_PGRS secretion , five strains with and one strain without this deletion were subjected to secretion analysis by immunoblotting . All strains exhibited similar secretion levels of both PE_PGRS proteins and the ESX-1 substrate EsxA , as compared to reference strain CDC1551 ( S1B Fig ) . These data show that the PPE71 variant carrying the MGGAGAG-deletion is able to sustain PE_PGRS secretion levels in M . tuberculosis , independently of truncation of PPE38 . Furthermore , there is no apparent phenotypic difference when M . tuberculosis has one or two functional copies of PPE38/71 . A striking amount of different RD5-like polymorphisms are present in the animal-adapted lineages/ecotypes of M . tuberculosis complex . These strains share their most recent common ancestor with M . africanum Lineage 6 [52 , 53] , which is reported to have two copies of ppe38/ppe71 [54] . Mycobacterium pinnipedi , a pathogen for seals and sea lions , has one intact copy of the ppe38 gene , but no esxXY-genes ( Fig 1A ) . M . bovis and Mycobacterium caprae share an identical RD5 deletion , while Mycobacterium orygis possesses a unique RD5 deletion ( Fig 1A ) [38 , 55 , 56] . To investigate the effect of RD5 deletions on PE_PGRS secretion in animal-adapted strains , we performed secretion analysis of M . bovis , M . caprae , M . orygis and M . pinnipedi ( Fig 1B ) . As expected , M . pinnipedi was the only tested species able to secrete PE_PGRS proteins in concordance with the presence of one functional copy of ppe38 ( Fig 1B ) . In contrast , M . bovis , M . caprae and M . orygis were deficient in PE_PGRS secretion , while EsxA secretion was not affected and no marked cell lysis occurred ( Fig 1B ) . Intracellular PE_PGRS expression was detected in strains with a secretion defect and was strikingly different between isolates ( Fig 1B ) . Since M . bovis and M . caprae share the same RD5 deletion with M . bovis BCG , we hypothesized that this vaccine strain is also deficient in PE_PGRS secretion ( Fig 1A ) . Indeed , BCG did not secrete PE_PGRS proteins comparable to M . tuberculosis-Δppe38-71 and the ESX-5 deficient strain eccC5::tn ( Fig 1C ) [37 , 57] . This PE_PGRS secretion defect was at least partially restored in the recombinant BCG strain complemented with the M . tuberculosis ppe38-71-locus , which we have called BCG38 . As expected , BCG and BCG38 were deficient in secretion of the ESX-1 substrate EsxA ( ESAT-6 ) and exhibited only low levels of PPE41 and EsxN . The increase of PPE41 secretion in BCG38 compared to the parental strains ( Fig 1C ) was consistent in this experiment and other replicates ( S1C and S1D Fig ) . Furthermore , five different M . bovis BCG isolates , which were selected for their relative genetic distance [6 , 10] , were all deficient in PE_PGRS secretion ( S1D Fig ) , emphasizing that all BCG strains are likely unable to secrete PE_PGRS proteins . It is of interest to note that M . bovis BCG Tice secretes higher levels of the ESX-5 substrates PPE41 and EsxN ( S1D Fig ) , likely because of its genetic duplication of the esx-5 genetic locus [10] . However , despite this ESX-5 duplication , BCG Tice is unable to secrete PE_PGRS proteins . The PE_PGRS secretion defect of BCG was not restored in a previously constructed BCG strain with a cosmid containing the complete RD5 region of M . tuberculosis H37Rv ( S1C Fig ) [16] . In contrast , introduction of the ppe38-71 locus from M . tuberculosis on an integrative plasmid constitutively expressing these genes under control of the hsp60 promoter [37] , partially restored PE_PGRS secretion of recombinant M . bovis BCG ( Fig 1C , S1C and S1D Fig ) . This finding was especially surprising since emergence of RD5-deleted M . bovis/M . caprae progenitor strains likely dates back thousands of years [52] . Taken together , our data show that the different BCG vaccines are all deficient for the secretion of PE_PGRS proteins and that this is at least partially revertible by complementation with the ppe38-71 locus of M . tuberculosis . Based on our previous work , this secretion defect is expected to affect up to 89 proteins classified as PE_PGRS or PPE-MPTR [27 , 37] . The ability to restore PPE38-dependent secretion in M . bovis BCG allowed us to investigate to what extent this secretion defect affects properties of the BCG vaccine . Many of the 89 members of the PE_PGRS and PPE-MPTR proteins have been suggested to perform biological roles in virulence and immune modulation , although the molecular mechanisms and biological relevance remain unestablished for most of these [14 , 27 , 33 , 34] . Increasing the repertoire of immunogenic proteins secreted by BCG could lead to increased protection , since protein secretion by mycobacteria is essential for the efficient induction of protective CD4+ T-cell responses [22 , 58–61] . However , restoring secretion of proteins that have been proposed to exhibit immunomodulatory functions could also decrease efficacy of the vaccine strain . In particular , recent reports suggest that PPE38 itself downregulates Major Histocompatibility Complex class-I ( MHC-I ) expression in murine macrophages [62] and that PE_PGRS47 inhibits autophagy and is responsible for reducing MHC-II-restricted antigen presentation during in vivo infection of mice [35] . We set out to establish whether presence of PPE38 and the ability to secrete PE_PGRS and PPE-MPTR proteins , affected phenotypic and functional maturation of infected murine innate immune cells . Bone marrow-derived dendritic cells ( BM-DCs ) of C57BL/6 mice were infected ( MOI = 0 . 5 ) with isogenic M . tuberculosis [37] or BCG strains , with or without the ppe38-locus . All the infected BM-DCs exhibited a clear upregulation of co-stimulatory markers CD40 , CD80 and CD86 , as well as modulation of MHC-I ( H-2Kb ) and MHC-II ( I-Ab ) expression , compared to uninfected controls . However , no differences in the induction of any such phenotypic maturation markers could be observed for the different isogenic WT and recombinant strains ( Fig 2 , S1 Table ) . Quantification of several inflammatory cytokines in the culture supernatants of the infected BM-DCs showed highly similar levels of TNFα , IL-12p40/70 and IL-6 production induced by the isogenic strains of BCG and M . tuberculosis ( Fig 2B ) . These results indicate that PPE38-dependent secretion defects are unlikely to have a major effect on the phenotypic or functional maturation of DCs , even though many PE_PGRS and PPE-MPTR proteins have previously been suggested to perform such biological roles [33 , 35] . In addition , we assessed whether PPE38-dependent protein secretion influences MHC-II-restricted presentation of other mycobacterial antigens . Such a phenotype might possibly be caused by a direct effect on the host phagocytes due to restored PE_PGRS secretion [35 , 36] , or by competition in the hosts antigen presentation machinery upon secretion of the large number of PPE38-dependent substrates . To test this hypothesis , BM-DCs were infected with serial two-fold dilutions of M . tuberculosis or BCG strains with and without the ppe38-locus . IL-2 secretion in culture medium by MHC-II restricted T-cell hybridomas specific to FbpA ( Ag85A101-120 –Fig 2C , upper panel ) or EsxH ( TB10 . 474−88 –lower panel ) was quantified by ELISA as a measure of antigen presentation and hybridoma T-cell activation . Higher levels of IL-2 were detected in response to M . tuberculosis strains compared to BCG strains , but no differences were observed between isogenic strains with , or without , functional PPE38-dependent PE_PGRS/PPE-MPTR secretion . These data show that PPE38-dependent PE_PGRS/PPE-MPTR secretion does not reduce MHC-II-restricted antigen presentation of other mycobacterial antigens by the host DCs . Together , these results suggest that introduction of PPE38 and restoration of PE_PGRS secretion do not negatively affect phenotypic and functional maturation of innate immune cells , or their capacity to present antigen to CD4+ T cells . Since we found no evidence suggesting that antigen presentation of mycobacterial antigens by DCs is negatively affected by restoration of PPE38-dependent secretion , we hypothesized that the enlarged repertoire of secreted proteins in BCG38 could increase its vaccine potential compared to the parental BCG . In parallel , we hypothesized that the capsule of BCG could be altered upon restoration of PPE38-dependent secretion . We recently reported that transposon insertions in the gene encoding an ESX-5 associated chaperone ( espG5 ) , or in the PPE-MPTR encoding gene ppe10 ( mmar_0761 ) , reduce capsule integrity of M . marinum [32] . Similarly , an eccC5::tn mutant in the M . tuberculosis strain CDC1551 , completely deficient in ESX-5 secretion , also exhibited reduced capsule integrity [32 , 57] . Since PPE10 is dependent on PPE38 for its secretion [37] , we hypothesized that restoration of PPE10 secretion might positively affect capsule integrity . The presence of an intact capsule on BCG , achieved by culturing in detergent-free growth medium , has recently been shown to be important for a more potent immune response and could therefore be relevant for the protective efficacy of BCG38 [63] . To test both hypotheses , C57BL/6 mice were subcutaneously ( s . c . ) immunized with 1 million CFU of either BCG , or BCG38 , cultured either in shaking condition in the presence of 0 . 025% Tween-80 , or in unperturbed conditions without detergent . Four weeks post-immunization , mice were challenged by an aerosol infection of M . tuberculosis H37Rv ( bacterial load: 680 CFU/lung at Day 1 , prepared without detergent ) . Mice were killed four weeks post infection , at which time lungs and spleens were harvested and assessed for bacterial burdens by CFU counting . An approximate 100-fold reduction in bacterial lung burdens was achieved by all conditions of vaccination irrespective of the presence of detergent , or the BCG vs BCG38 vaccine strains ( Fig 3A ) . This reduction of bacterial lung burden coincided with improved macroscopic state of the lungs ( S2A Fig ) . Similarly , an approximately 10-fold reduction in spleen CFUs and reduction in splenomegaly was detected in the vaccinated mice irrespective of the method of vaccine preparation ( Fig 3A , S2B Fig ) . No significant ( p<0 . 05 ) differences in bacterial burdens were observed between any of the four tested conditions in either the spleens or lungs . Together , these results show that restoration of PPE38-dependent PE_PGRS/PPE-MPTR secretion in BCG does not significantly improve protection against M . tuberculosis in the murine model used . Moreover , we did not find a significant difference in protective efficacy between conventional and detergent-free preparation of either BCG or BCG38 , suggesting that capsular integrity is not altered or does not affect protection in this model . Secretion of T7S-mediated mycobacterial proteins is essential to induce host CD4+ T-cell responses and the great majority of immunogenic and protective antigens of M . tuberculosis are secreted proteins [64] . Many of the known immunodominant antigens are PE and PPE proteins and these form an integral part in a number of subunit or recombinant vaccines [58 , 65–68] . Therefore , the finding that restoration of PPE38-dependent PE_PGRS/PPE-MPTR secretion in BCG did not significantly affect protective efficacy was surprising , particularly as up to 89 individual proteins are predicted to be concerned . In order to explain these unexpected data , we reflected on our hypotheses and found additional variables that could affect the assumptions on which they are based . In particular , while PPE-MPTR secretion was shown to be strictly dependent on PPE38 in both M . marinum and M . tuberculosis , we had no direct evidence of PPE-MPTR secretion in BCG38 . In contrast to PPE-MPTR proteins , PE_PGRS proteins may not contain immunodominant epitopes or be protective antigens [69–72] . Furthermore , although previous studies have found a strict correlation between in vitro secretion and the capability to induce CD4+ T-cell responses [23 , 58 , 68] , it is conceivable that the PPE38-dependant substrates are still membrane , or surface , associated in ppe38-71-deficient strains and thereby remain able to induce T-cell responses . Since tools to study PPE-MPTR proteins are scarce and currently insufficient to answer the questions above , we set out to develop an immunological approach to study PPE-MPTR secretion and their immunogenicity in more detail . We selected PPE10 as a model MPTR-protein , because PPE10 is predicted to be the most ancestral MPTR protein in mycobacteria [27] . The PPE domain covers the N-terminal 181 residues of PPE10 and is highly similar to other PPE proteins . The middle of the protein contains a typical MPTR repeat domain , which is very similar to other MPTR proteins . The C-terminus contains a domain unique to PPE10 , which is secreted in vitro [25 , 32 , 37] . PPE10 is also of biological interest , since it is detected in vivo in guinea pig lungs and this protein is required for capsular integrity of M . marinum [32 , 73] . We set out to assess whether PPE10 has the potential to induce CD4+ T-cell mediated immune responses in mice . To increase the likelihood of identifying immunogenic epitopes , we immunized not only C57BL/6 mice , but also C57BL/6 x CBA ( H-2b/k ) F1 mice , which express a more diverse repertoire of MHC restricting elements ( S2 Table ) . Mice were s . c . immunized with wild-type M . tuberculosis H37Rv and were killed three weeks later . Splenocytes were isolated and stimulated in vitro with a peptide library consisting of sixty 15-mers with a 5-amino acid shifting frame spanning PPE10181-487 of M . tuberculosis H37Rv [29 , 74] . None of the sixty peptides were able to induce specific T-cell mediated IFN-γ responses by splenocytes from unimmunized mice or immunized C57BL/6 mice ( S3 Fig ) . However , two peptides were immunogenic in the C57BL/6 x CBA ( H-2b/k ) F1 mice and induced high levels of IFN-γ , similar to the positive control peptide ESAT-61−20 ( Fig 4 , S3 Fig ) . Interestingly , one of these immunogenic peptides ( PPE10221-235: GSGNTGSGNLGLGNL ) was situated in the MPTR domain of PPE10 , while the other ( PPE10381-395: NVLNSGLTNTPVAAP ) was derived from the PPE10-specific C-terminal domain . The MPTR peptide PPE10221-235 has 17 close homologues within the M . tuberculosis genome ( identity > 65% , but no 100% homologues ) , while this was not the case for PPE10381-395 ( S3 Table ) . These results show that immunization with M . tuberculosis induces immune responses against PPE10 and that this response can be elicited both against the PPE10-specific C-terminal domain or the MPTR domain . The newly identified immunogenic peptides derived from PPE10 are a tool that allowed us to answer different questions regarding the PPE-MPTR proteins . First , to determine the specificity and cross-reactivity of the epitopes , we constructed a deletion mutant of ppe10 ( Rv0442c ) in the M . tuberculosis CDC1551 background by homologous recombination and phage transduction ( S4 Fig ) [75] . In contrast to M . marinum-ppe10::tn [32] , no altered colony morphology or other growth phenotype was observed in M . tuberculosis-Δppe10 . This finding is in concordance with the absence of such a phenotype in ESX-5 mutants of M . tuberculosis and highlights this as a species-specific difference between M . marinum and M . tuberculosis [32 , 57 , 76] . We performed biochemical secretion analysis on the Δppe10 strain in parallel with the strains that were examined ( see below ) for their immunogenic potential ( Fig 1C ) . In contrast to a previous report , we found that M . tuberculosis Δppe25-pe19 did secrete PPE41 and EsxN , which may be due to differences in bacterial growth conditions and/or methods in protein extraction and detection [76] . This strain harbors intact genes coding for the ESX-5-membrane complex [57 , 77] and is able to induce in vivo CD4+ T-cell responses against PE and PPE proteins , in contrast to the general ESX-5 deficient strain ΔeccD5 in the same background [58 , 68] . The Δppe10 strain showed no difference in PE_PGRS secretion . Similarly , secretion of EsxA and EsxN was not affected by deletion of ppe10 . Although slightly elevated levels of PPE41 secretion were observed , we concluded from these combined data that M . tuberculosis-Δppe10 does not have a general supersecretion phenotype as was previously reported for M . marinum-ppe10::tn [32] . To assess the specificity of the newly identified PPE10 epitopes and to better understand the effect of the ppe38-dependent secretion on immunogenicity , we immunized C57BL/6 x CBA F1 mice with the different M . tuberculosis and BCG strains for which the secretion phenotype was characterized ( Fig 1C ) . Three weeks post-immunization , splenocytes were collected and stimulated with the PPE10221-235 and PPE10381-395 peptides , as well as purified protein derivate ( PPD—a positive control for immunization by Mycobacteria ) and a number of known antigenic peptides derived from proteins secreted via ESX-1 ( EsxA1-20 [78] and CFP-1011−25 [79] ) , ESX-5 ( PE191-18 and PPE251-20 [58] ) or the twin-arginine-translocation ( TAT ) pathway ( Ag85A241-260 ) [80 , 81] . As expected , splenocytes of mice immunized with M . tuberculosis CDC1551 produced high levels of IFN-γ after stimulation with PPE10221-235 , PPE10381-395 or all other immunogenic peptides , but not when incubated with a negative control peptide ( E . coli MalE100-114 ) , or the medium control ( Fig 5 ) . The Δppe10 deletion strain did not induce IFN-γ production in response to either PPE10221-235 , or PPE10381-395 , whereas responses against the other peptides were unaffected ( Fig 5 ) . Unexpectedly , this result shows that both of the newly identified PPE10 peptides are highly specific , even though we hypothesized cross-reactivity to occur for PPE10221-235 , because of the high similarity to other MPTR domains ( S3 Table ) . As expected , the ESX-5 secretion mutant eccC5::tn did not induce T-cell responses against the ESX-5 substrates PE19 , PPE25 and PPE10 , further confirming that the export of these antigens by the ESX-5 secretion system is indispensable for the induction of T-cell immune responses [58 , 68 , 82] . Importantly , Δppe38-71 was not able to induce immunogenicity against either of the PPE10 epitopes , a phenotype that was fully reverted in the complemented strain ppe38-71-C . This confirms that secretion and in vivo immunogenicity of PPE10 as a model PPE-MPTR protein are dependent on PPE38 in the M . tuberculosis CDC1551 background , which we were previously unable to assess . Similar to Δppe38-71 , also BCG was completely unable to induce immune responses against either of the PPE10 epitopes . In contrast , BCG38 induced immunogenicity against both PPE10 epitopes at similar levels to the M . tuberculosis isolates . Together , these results clearly confirm that the secretion and in vivo immunogenicity of the ancestral PPE-MPTR protein PPE10 is strictly dependent on PPE38 . These data also provide evidence that the in vitro observed PPE38-dependence of PE_PGRS and PPE-MPTR proteins is a phenotype that can be directly translated to the in vivo situation . Here , we show that the vaccine strain BCG is unable to induce T-cell responses against the ancestral PPE-MPTR protein PPE10 , because of the deletion of its ppe38-71-locus as part of RD5 . Finally , we compared the results obtained for the different WT and recombinant BCG strains with a recently developed attenuated M . tuberculosis strain , deleted for 5 pe/ppe genes in the esx-5 locus , named MtbΔppe25-pe19 [76] . Genes encoding the ESX-5 secretion core machinery [57 , 77] are intact in this strain , as is the ppe38 gene , a finding which is confirmed by the fact that this strain induced T-cell responses against both PPE10 epitopes . This result highlights that attenuated M . tuberculosis vaccine strains may avoid certain M . bovis related secretion differences that result in immunogenic properties . The results of our epitope mapping analysis showed that C57BL/6 mice were unable to develop T-cell responses against PPE10 , which could provide an explanation for the lack of improved protection conferred by BCG38 compared to BCG . Therefore , we performed a similar experiment in these C57BL/6 x CBA F1 mice , designed to maximize any potential increase in PPE-MPTR-specific immune responses , by boosting vaccination of BCG or BCG38 with the immunogenic PPE10- peptides ( Fig 6A ) . Sixty days after s . c . vaccination with BCG strains , a CpG ( DOTAP ) -formulated peptide booster , or the adjuvant alone , was administered s . c . , followed by an intranasal booster twenty-nine days later . Mice were challenged by an aerosol challenge of M . tuberculosis H37Rv nine days after the final booster and were killed 28 days later to assess lung and spleen bacterial burdens ( Fig 6B and 6C ) . No significant differences were observed among the groups of vaccinated animals . Only a modest decrease in spleen CFUs was achieved by any of the vaccination regimens . This reduction was not significant ( p<0 . 05 ) for BCG-vaccinated mice and injected with the adjuvant alone , but was significant for the three other groups . However , no significant differences between any of the vaccinated groups was detected . Vaccination with all regimens reduced lung CFU values at least 10-fold . In fact , vaccination with BCG38 , boosted with PPE10-derived immunogenic peptides , had the highest average bacterial lung burden of the four different vaccination regimens . These data clearly oppose our hypothesis , that restoring the lack of PPE-MPTR immune responses in BCG increases its protective efficacy . Together , we could find no evidence of an immunomodulatory effect of PPE38-dependent proteins . Inversely , restoration of BCG’s capacity to secrete PE_PGRS and PPE-MPTR proteins and thereby enlarging the PE_PGRS/PPE-MPTR antigenic repertoire of BCG , did not result in improved vaccine protection in two mouse models .
We previously demonstrated that loss-of-function mutations in the ppe38-locus of M . tuberculosis block PE_PGRS and PPE-MPTR secretion and increase virulence in a mouse model [37] . In this work , we examined the correlation of known ppe38 deletions in other lineages of the MTBC with a PE_PGRS/PPE-MPTR secretion defect . We hypothesized that the success of certain clinical isolates of Lineage 4 could perhaps be explained by their RD5-like deletion , which includes ppe38 [51 , 83] . However , secretion analysis of these Lineage 4 strains revealed that a single copy of PPE71 carrying a 21 bp deletion ( corresponding to the loss of amino acids MGGAGAG ) , seems to be functional and sufficient to support PE_PGRS secretion . Similarly , although intriguing differences in protein secretion levels were observed between M . canettii strains , we found that all analyzed strains secreted PE_PGRS proteins . The anticipated polymorphisms in the ppe38-locus of selected M . canettii strains [48] were likely caused by a sequence assembly problem of repetitive sequences . These results highlight the difficulties of bio-informatic analyses of this locus , which is hampered by the high sequence similarity between ppe38 and ppe71 , that seem to cause already some discrepancies between the reference genomes of M . tuberculosis H37Rv and CDC1551 [29 , 37 , 38 , 84] . In contrast , our investigation of RD5-like polymorphisms did reveal that multiple members of the animal adapted lineage of the MTBC are completely devoid of PE_PGRS secretion because of their RD5 deletion . It should be emphasized that the RD5-like deletion of M . orygis occurred independently of that of M . bovis and M . caprae . Furthermore , even more members of the animal adapted lineage , such as M . microti , M . suricattae and the Dassie Bacillus , are reported to have independent RD5-like deletions , which we hypothesize to also block PE_PGRS and PPE-MPTR secretion [38 , 85–87] . Together , these findings suggest a specific selective advantage associated to loss of the ppe38-locus and its associated secretion phenotype in certain animal adapted strains . The modern Beijing strains , also defective in PPE38-dependent secretion , have expanded concurrently with increased human population densities and mobility [88] . These changes in the host-population alter the optimal balance between virulence/infectivity and lower the advantage to stay dormant or subclinical in the host [89] . It is tempting to speculate that the loss of PPE38 and its associated secretion and virulence phenotype has helped ancestral M . tuberculosis strains derived from human hosts , to adapt towards survival and transmission in a new host niches . We were surprised that we were able to restore the secretion defect of BCG by introducing the ppe38-locus from M . tuberculosis . Since the RD5 deletion of BCG already occurred in the most-recent common ancestor of M . bovis and M . caprae , this deletion likely dates back millennia [52] . Furthermore , the 13 years of in vitro culturing by Calmette and Guérin to create BCG and the ensuing decades of culturing while it was distributed worldwide has caused accumulation of even more mutations [6 , 10 , 11] . Still , introduction of the integrative vector constitutively expressing the ppe38-locus was clearly able to at least partially restore both PE_PGRS and PPE-MPTR secretion in BCG . Our newly identified immunogenic epitopes in the PPE-MPTR protein PPE10 , provide a tool to gain more understanding about this group of proteins . Firstly , although previous work only definitively detected the C-terminal domain of PPE10 to be secreted [25 , 32 , 37] , immunization with M . tuberculosis also clearly induced immune responses against the MPTR-associated epitope . This provides evidence that the MPTR domain is accessible to the host antigen presentation machinery and that these repetitive domains have the potential to contain functional T-cell epitopes . Furthermore , wild-type BCG and M . tuberculosis with impaired PPE38-dependent secretion were completely unable to induce immune responses against PPE10 , similar to a general ESX-5 secretion mutant . We recently reported that processing , presentation on MHC-II and recognition of immunogenic proteins by CD4+ T-cells occurs only when the protein is transported over the inner membrane by its cognate TypeVII secretion system [82] . Secreted proteins , but also the cell-wall associated protein PE19 induced strong CD4+ T-cell responses , whereas this was not the case in an ESX-5 deficient isolate , where PE19 is only present in the cytoplasm or associated to the plasma membrane [82] . Together with the current work , this is important evidence that PPE38 is essential for the translocation of PPE-MPTR proteins through the ESX-5 secretion machinery in vivo and that without PPE38 , these proteins are not surface associated or otherwise accessible to the immune system . It is perhaps striking that the PE_PGRS and PPE-MPTR secretion defect of BCG has not been previously reported , considering the amount of research done on this vaccine . Based on the available literature on PE_PGRS and PPE-MPTR proteins , it is logical to hypothesize that a vaccine strain that does not secrete these proteins might in fact be a relatively effective vaccine . Many immunomodulatory properties have been attributed to PE_PGRS and PPE-MPTR proteins [14 , 33 , 34 , 67] . Perhaps the most relevant of these , is the reported function of certain PE_PGRS proteins to inhibit antigen presentation [35 , 36] . If PE_PGRS proteins indeed inhibit antigen presentation , it would be highly detrimental to introduce a vaccine that secretes these proteins . Notably , this is an urgent question since a number of novel tuberculosis vaccine candidates based on attenuated M . tuberculosis are currently in clinical or pre-clinical development . We showed for one of these candidate vaccines ( i . e . M . tuberculosis-Δppe25-pe19 ) , that PE_PGRS and PPE-MPTR secretion is indeed fully functional [58 , 68 , 76] . Our isogenic Δppe38-71 strains of M . tuberculosis and the BCG38 strain form an ideal tool to answer such questions and to understand more about these proteins as a group . In this work , we did not find any evidence of inhibition of antigen presentation in strains secreting PPE38-dependent substrates , or lack thereof in strains without PPE38 . Similarly , and in contrast to many reports of immunomodulatory effects of PE_PGRS proteins , we did not find any evidence of differential immune modulation by strains with , or without , functional PPE38-dependent secretion . More specifically , no differences were observed in DC maturation [90] , MHC-I or -II expression [62] or cytokine production [91–93] . Finally , PE_PGRS and PPE-MPTR proteins have often been implicated as mycobacterial virulence factors [14 , 34 , 35 , 94 , 95] . The previously described increased virulence in strains lacking PPE38-dependent secretion , including the hypervirulent Beijing isolates , put this work in perspective [37] . Here , we bolster our previously published evidence that strains without PPE38 , including a number of animal adapted species and the BCG vaccine , are truly unable to translocate these proteins . Although many of these animal adapted strains have reduced virulence in humans compared to M . tuberculosis , they are clearly pathogenic for their natural host and should not be seen as attenuated [96] . This is in line with a role for PPE38-dependent substrates as virulence attenuating factors [37] . The biological roles of the PE_PGRS and PPE-MPTR proteins that are reported to be required for virulence , may not require secretion of these effector proteins or might in certain cases be due to indirect effects on other proteins . This hypothesis is further supported by the fact that many of the studies that attribute virulence traits to PE_PGRS and PPE-MPTR proteins , are performed in M . smegmatis , which lacks an ESX-5 secretion system and is unable to secrete these proteins [27 , 33 , 77] . Further work on the biological function of PE_PGRS and PPE-MPTR proteins , either on an individual basis or grouped , will have to take into account these findings and critically assess the impact of localization on effector function . Whether complementation of the ppe38-locus in BCG might further attenuate BCG strains or otherwise affect persistence in the host was not studied in further detail , in part because BCG38 did not confer superior protection to M . tuberculosis infection . However , we do not expect significant virulence differences between BCG and BCG38 , since PPE38-dependent virulence effects generally occur in the chronic infection stage [37] , when BCG is already expected to be cleared . Perhaps the most relevant finding of this work is that BCG is unable to secrete PE_PGRS and PPE-MPTR proteins and therefore does not raise T-cell responses against these proteins . Previous studies have shown that antibodies can be raised against PE_PGRS proteins , suggesting that it could be a beneficial property of a vaccine to secrete these proteins [25 , 97 , 98] . Here we provide evidence that PPE-MPTR proteins can be immunogenic in mice , which is further supported by a recent publication investigating immunogenicity of the PPE-MPTR protein PPE39 [72] . Kim et al . identified two immunogenic epitopes of which one ( MTBK_2482085−102 ) is located in the PPE-domain and has high homology to non-MPTR PPE proteins , while the other ( MTBK_24820217−234 ) was located in the MPTR domain of this protein . Interestingly , the authors reported that vaccination with the recombinant PPE39 protein induced a higher level of protection against M . tuberculosis Erdman , compared to a hypervirulent Beijing isolate [72] . This difference might be explained by our data , which suggest that immune responses against the MPTR epitope would not be helpful against a PPE38-deficient Beijing isolate . A related issue that requires further work is whether the PPE38-dependent secretion effect in modern Beijing isolates is somehow related to that of the BCG vaccine and whether their respective secretion defects affect vaccine efficacy . There is strong evidence for the importance of PPE-MPTR proteins in human immune responses , because the PPE-MPTR protein PPE42 ( Rv2608 ) is an integral part of the subunit fusion-protein vaccine candidate ID93 [65 , 71] . The fusion protein ID93 consists of four different proteins and has been tested as a vaccine candidate in both a Phase 1 and Phase 2A clinical trial [99 , 100] . Bertholet et al . 2008 demonstrated that PBMCs isolated from PPD+ healthy subjects produced IFN-γ in response to PPE42 and that almost 70% of subjects showed a reaction against the recombinant protein in a recall experiment [71] . Interestingly , 100% of PPD+ subjects exhibited recall responses against the other ( non-MPTR ) PPE proteins that were tested , which could possibly be explained by exposure to modern Beijing , or other PPE38-deficient strains , in the subject cohort . PPE42 was selected as part of the ID93 vaccine due to its excellent ability to induce both humoral and cellular immune responses and immunization with PPE42 provided protection in mice almost comparable to BCG [65 , 71] . In Guinea pigs , ID93 significantly boosted the protection induced by BCG , which was interpreted as an ability to boost immune responses elicited by BCG [65] . However , based on our work it should be assumed that BCG does not induce immune responses against the PPE-MPTR protein PPE42 and that boosting with ID93 may in fact broaden antigenic repertoire of the combined vaccination . Similarly , ID93 is able to induce protective immune responses to the M . tuberculosis Beijing isolate HN878 , but it is unclear what the role of PPE42 is in this response . The analyses performed in Bertholet et al . 2010 and Baldwin et al . 2015 were performed with the four-gene fusion protein ID93 and not with the individual PPE42 subunit , which makes it impossible to assess these questions more thoroughly . What remains clear however , is that the PPE-MPTR protein PPE42 is an important part of a vaccine currently in clinical trials . The finding that ID93 includes a protein to which parental BCG is likely not able to induce immune responses , may actually put the proven booster qualities of this vaccine candidate in a different light and lead to optimal strategies to employ it . The question whether immune responses against PPE38-dependent proteins are important for a vaccine to be protective against tuberculosis , needs an urgent answer , especially since it concerns a total of 89 proteins . There are multiple vaccine candidates in clinical , or pre-clinical , development that are based on attenuated M . tuberculosis strains and which likely secrete PE_PGRS and PPE-MPTR proteins [24 , 58 , 67 , 68 , 76 , 101] . Should we knock-out ppe38-71 in these vaccine candidates to avoid immune modulation by the secreted substrates , or should we prioritize these vaccine candidates , because they have a broader potential repertoire of epitopes ? Should BCG vaccination be boosted by vaccine candidates including PPE-MPTR proteins such as ID93 , or should this be avoided ? Are there differences between designing vaccine candidates against strains secreting PE_PGRS/PPE-MPTR proteins and those with a PPE38-dependent secretion defect , such as the modern Beijing isolates ? Are murine or other small animal infection models appropriate to predict PE_PGRS and PPE-MPTR-mediated impact on vaccine efficacy ? These are questions that we are not yet able to answer in this work , but they reveal the need to increase our understanding of PE_PGRS and PPE-MPTR proteins . Better knowledge on PE_PGRS/PPE-MPTR proteins is not just an intellectual goal , but may also help to make more informed decisions in the design of novel vaccines against tuberculosis .
All strains used in the study and the sources they are derived from can be found in S5 Table . Unless otherwise specified , all mycobacterial strains were grown on Middlebrook 7H11 solid medium ( Difco ) supplemented with OADC ( BD Biosciences ) , or liquid 7H9 medium supplemented with ADC supplement and 0 . 05% Tween-80 . Antibiotics were added where opportune at a concentration of 50μg/ml for Hygromycin ( Euromedex ) , or 25μg/ml for Kanamycin ( Sigma ) . Strains were incubated at 37ºC . Liquid cultures were grown in shaking conditions at 80 rotations per minute . For animal-adapted strains M . bovis , M . caprae , M . orygis and M . pinnipedii , 0 . 2% w/v of Pyruvate ( Sigma ) was added to the growth medium [102] . Infection stocks of M . tuberculosis H37Rv used for aerosol infection experiments and BCG or BCG38 vaccination stocks without Tween-80 were prepared by inoculating 0 . 1 OD/ml bacteria in 100ml liquid culture without Tween-80 . This culture was incubated for 7 days , after which it was washed with phosphate buffered saline ( PBS ) and sonicated ( 5x ( 100 pulses of 0 . 1s ) ) and left to rest for at least one hour before collecting the cell suspension considered to obtain a single-cell solution of encapsulated mycobacteria . Standard vaccination stocks were prepared in Dubos medium containing 0 . 025% Tween-80 in standing conditions and were harvested at an optical density between 0 . 4 and 0 . 7 OD600/ml . RD5 deletions were PCR verified by previously published primers specific for plcA ( rv2351c , S4 Table ) , which produce a product of approximately 500bp when this gene is present [52] . Primers amplifying the ppe38-71-locus ( S4 Table ) produce a 3378bp product when the complete ppe38-71 locus is present [38] . This includes two copies of ppe38/71 ( mt2419/mt2422 ) flanking the esxX ( mt2420 ) and esxY ( mt2421 ) in between in CDC1551 . When only one copy of ppe38 and no esxX/esxY are present this PCR produces a product of approximately 1500 bp [38] . The complementation plasmid containing the ppe38-locus from CDC1551 ( mt2419-22 ) under expression of hsp60 promoter was previously described [37] . The cosmid containing the RD5 region ( pYUB::RD5 ) was part of the library described by Bange et al . 1999 and contains the genetic region spanning 2 , 611 kb– 2 , 645 kb of the M . tuberculosis H37Rv reference genome [29 , 103] . M . tuberculosis-Δppe10 was constructed as described by Bardarov et al . [75] . The homologous recombination construct was created by a PCR combining primers PPE10 KO LF & LR to amplify the 3’ end of rv0442c and another PCR with primers PPE10 KO RF & PPE10 KO RR to amplify the 5’ end of rv0442c ( See S4 Table for primer sequences ) . After phage packaging and infection , seven transformed colonies were tested by PCR with either primer PPE10 ( mtb ) flank F & p0004s-HR , or PPE10 ( mtb ) flank R & p0004s-HL ( S4A and S4B Fig ) . All colonies were found to have the correct deletion spanning from 152bp to 1133bp after the 5’ of rv0442c . We attempted to complement the Δppe10 mutant with a previously published plasmid ( p19kPro::rv0442c-HA ) overexpressing HA-tagged PPE10 under control of the lpqH promotor [25] . Although clones expressing the HA-tag on this plasmid were obtained , these had a considerable in vitro growth defect , which would conflict with in vivo and in vitro studies and therefore this complemented strain was not analyzed further . Strains were pre-cultured until mid-logarithmic phase under normal growth conditions ( described above ) . Cultures were washed two times in 7H9 medium without ADC , supplemented with 0 . 2% Dextrose and 0 . 05% Tween and were incubated in this medium for 48 hours . Cultures were centrifuged to separate cells and the supernatant was filtered through a 0 . 02μm filter , after which it was TCA-precipitated to concentrate . Cellular material was washed with PBS , resuspended in solubilisation/denaturation buffer and boiled for 10 min at 95°C . After sterilisation by heating for 2 hours at 80°C , samples were sonicated to disrupt cells and boiled at 95°C during 10 minutes . Samples were loaded on 12% or 4–12% SDS-Page gels ( NuPage , Novex , Life technologies ) and transferred to nitrocellulose filters by dry western blotting ( iBlot , Invitrogen ) . Proteins were stained by primary antibodies: Anti-PGRS 7C4 . 1F7 [25] ( Clone 7C4 . 1F7 was a kind gift from Michael J . Brennan , USA ) , polyclonal anti-SigA ( Kind gift from I . Rosenkrands , Denmark ) , Rabbit polyclonal anti-EsxN ( rMTb9 . 9A ) [104] , monoclonal ESAT-6 ( hyb76-8 ) , or anti PPE41 [105] . BM-DCs derived from C57BL/6 ( H-2b ) female mice were generated directly in 6-well plates and infected at day 6 of culture with different mycobacterial strains at M . O . I of 0 . 5 in RPMI 1640-GlutaMax medium ( Invitrogen ) containing 10% FBS ( 4 x 106 cells/well in 4 ml volume ) . After over-night of infection at 37°C and 5% CO2 , IL-6 ( clone MP5-20F3 for coating and clone MP5-32C11 for detection , BD Pharmingen ) , IL12p40/70 ( clone C17 . 8 RUO , BD Pharmingen ) and TNF-α ( clone 1F3F3D4 for coating and clone clone XT3/XT22 for detection , eBioscience ) cytokine production was quantified in the culture supernatants by ELISA . For viability and phenotypic maturation evaluation , infected DCs were washed with PBS and incubated first with Live/Dead-Pacific Blue reagent ( Invitrogen ) during 35 minutes at 10°C in the dark . Cells were then washed twice and incubated with appropriate dilution of anti-CD16/CD32 ( 2 . 4G2 mAb , BD Pharmingen ) during 20 minutes followed by surface staining by 30 minutes of incubation with appropriate dilutions of APC-anti-CD11b ( BD Pharmingen ) , PE-Cy7-anti-CD11c ( BD Pharmingen ) , FITC-anti-CD40 ( clone HM40-3 , SONY ) , FITC-anti-CD80 ( B7-1 ) ( clone 16-10A1 Biolegend ) , FITC-anti-CD86 ( B7-2 ) ( clone PO3 , SONY ) , FITC-anti-MHC-II ( I-A/I-E ) ( clone MS/114 . 15 . 2 , eBioscience ) , FITC-anti-MHC-I ( H-2kb ) ( clone AF6-88-5-5-3 , eBioscience ) or FITC-anti-IgG1k isotype control . The stained cells were washed twice with FACS buffer ( PBS containing 3% fetal bovine serum ( FBS ) and 0 . 1% NaN3 ) and then fixed with 4% paraformaldehyde during 18h at 10°C prior to sample acquisition by a LSR Fortessa flow cytometer system ( BD Bioscience ) and BD FACSDiva software . The obtained data were analyzed using FlowJo software ( Treestar , OR , USA ) . BM-DCs derived from BALB/c ( H-2d ) female mice were used at day 6 of culture as antigen presenting cells . Cells were seeded in 96-well plates at 5 x 104 cells/well and loaded with 1 μg/ml of homologous or negative control synthetic peptides , or infected with different mycobacterial strains with serial two-fold dilutions of M . O . I . , starting at M . O . I . = 10 , in RPMI 1640-GlutaMax medium ( Invitrogen ) containing 10% FBS . After 18h of infection at 37°C and 5% CO2 , cells were washed twice with RPMI medium to eliminate the IL-2 possibly produced by the infected DCs and then co-cultured with 1 x 105 cells/well of T-cell hybridoma specific to EsxH/TB10 . 474−88 ( 1G1 ) or Ag85A101-120 ( 2A1 ) , respectively restricted by I-Ad or I-Ed . After over-night of co-culture at 37°C and 5% CO2 , the IL-2 secretion was quantified in the culture supernatants by ELISA ( clone JES6-1A12 for coating and clone JES6-5H4 for detection , BD Pharmingen ) . A peptide library of sixty 15-mers with a 5-amino acid shifting frame , spanning amino acids 181–487 of PPE10 ( Rv0442c ) , was constructed commercially ( Mimotopes Europe , United Kingdom ) . Epitope screening of PPE10 and immunogenicity assays were performed as previously described [58] , with some modifications . Briefly , 6-8-week-old female C57BL/6 ( H-2b ) or C57BL/6 x CBA F1 ( H-2b/k ) mice were immunized s . c . with 1 x 106 CFU/mouse of different mycobacterial strains obtained from exponential culture in Dubos medium . Epitope mapping was performed with mice immunized with M . tuberculosis H37Rv . Three to four weeks post-immunization , mice were sacrificed and pool of total splenocytes ( n = 2 mice per group ) were restimulated in 96-well flat-bottom plates ( TPP , Den- mark ) at 5 x 105cells per well in HL-1 medium ( Biowhittaker , Lonza , France ) , complemented with 2 mM GlutaMax ( Invitrogen , Life Technologies , France ) , 5 x 10−5 M β-mercaptoethanol , 100 U/ml penicillin and 100 μg/ml streptomycin ( Sigma-Aldrich , France ) in the presence of 10–20 μg/ml of individual peptides . IFN-γ production in the supernatant was quantified by ELISA after 72h of culture at 37°C and 5% CO2 ( clone AN-18 for coating and clone R46A2 for detection ) , BD Pharmingen . BCG and BCG38 were grown in 10ml Dubos medium or in 100ml 7H9-medium with ADC-supplement without Tween-80 . M . tuberculosis H37Rv and BCG-strains cultured without Tween-80 were sonicated ( 5 X 100 pulses; 0 . 1 seconds/pulse; 0 . 9 seconds’ rest; amplitude 30% ) to disrupt clumps and were frozen at -80°C . Frozen stocks were counted for CFU’s before immunization to assess dose while the dose of Dubos-grown strains was estimated based on optical density . Eight-week-old C57BL/6 mice ( n = 5 mice/group ) , were immunized with 1 x 106 CFU/mouse of BCG Danish ( cultured—or + Tween-80 ) , or BCG38 ( cultured—or + Tween-80 ) in 200 μl PBS . Eight mice were concurrently injected with sterile PBS . Thirty days after vaccination , mice were challenged with aerosolized WT M . tuberculosis H37Rv strain . Three mice were sacrificed to assess bacterial lung burdens 1 day post challenge ( assessed at 680 CFU/lung ) . All other mice were killed four weeks post-challenge due to human end-point criteria of unvaccinated mice . Lungs and spleens were homogenized by beadbeating , serially diluted in PBS and plated on 7H11 plates with ( Lungs ) or without ( spleens ) BBL MGIT PANTA ( Beckton Dickinson , Ireland ) . The Prime-boost vaccination and challenge experiment was performed similar as above , with the following modifications . BCG or BCG38 were precultured in Dubos medium and five first generation C57BL/6 x CBA crossover mice were left unvaccinated or s . c . immunized ( n = 5 mice/group ) . Eight weeks post-immunization , a subcutaneous boost was administered . This boost consisted of 200 μl/mouse of formulation containing 50 μl of each PPE10-derived peptide ( PPE10221-235 and PPE10381-395 ) ProteoGenix , France , 30 μg of CpG 1826 oligodeoxynucleotides as adjuvant ( Sigma-Aldrich , France ) at 1 μl/mL concentration , 60 μl of liposomal transfection reagent DOTAP ( N-[1- ( 2 , 3-DioleOyloxy ) ]-N , N , N-Trimethyl Ammonium Propane methylsulfate , Roche , France ) and 10 μl Opti-MEM ( Life Technologies , France ) as described in Sayes et al . , 2016 [68] ) . Four weeks later , an intranasal boost was given to mice via intra-nasal route , under anesthesia as described in Sayes et al . , 2016 , 25 μl/mouse contained 10 μg of PPE10 peptides , 2 μg of CpG at 10 μl/mL concentration , 10 μl of DOTAP and 3 μl Opti-MEM contained in 20 μl/mouse [68] . Ten days after the intranasal boost , mice were aerosol challenged with WT M . tuberculosis H37Rv strain . Three non-immunized mice were killed one day post challenge to assess infectious dose administered , which was calculated at 220 CFU/lung . Four weeks later all other mice were killed and one lung and the spleen were homogenized with a MillMixer organ homogenizer ( Qiagen , Courtaboeuf , France ) and plated to assess bacterial burdens on 7H11 Agar medium supplemented with ADC ( Difco , Becton Dickinson ) . The CFU were counted after 3–4 weeks of incubation at 37°C . All immunized and infected mice for immunogenicity and protection experiments were placed and manipulated in isolator in BSL-III protection-level animal facilities at the Institut Pasteur . To determine the statistical significance of the data , analyses were performed by use of GraphPad Prism software ( GraphPad Software , La Jolla , CA , USA ) , using ordinary one-way ANOVA followed by Tukey’s test for multiple comparisons . All animal experiments were performed in animal facilities that meet all legal requirements in France and by qualified personnel in such a way to minimize discomfort for the animals . All procedures including animal studies were conducted in strict accordance with European and French regulations ( Directive 86/609/CEE and Decree 87–848 of 19 October 1987 ) . All protocols were reviewed and approved by the Institut Pasteur Safety and Animal Care and Use Committee ( Protocol 11 . 245 ) and the local ethical committee CETEA “Comité d'Ethique en Expérimentation Animale” ( approved protocols CETEA 2012–0005 and CETEA 2013–0036 ) . | One of the major findings of the pioneering Mycobacterium tuberculosis H37Rv genome sequencing project was the identification of the highly abundant PE and PPE proteins , named after their N-terminal motifs Pro–Glu ( PE ) or Pro–Pro–Glu ( PPE ) . Within the 20 years of research since then , many claims were made that PE/PPE proteins , including the two large subgroups encoded by repetitive sequences with very high GC content ( PE_PGRS and PPE-MPTR families ) , are exported to the bacterial surface or beyond , and show broad immunomodulatory impact on host-pathogen interaction . We thus screened strains from different branches of the M . tuberculosis complex , including the attenuated Mycobacterium bovis BCG vaccine strains , for their capacity to export PE_PGRS/PPE-MPTR proteins . Strikingly , we found that BCG strains were unable to export the plethora of PE_PGRS/PPE-MPTR proteins due to the absence of the region of difference RD5 , which in M . tuberculosis encodes PPE38 , required for PE_PGRS/PPE-MPTR export . Surprisingly , the restoration of PE_PGRS/PPE-MPTR export by genetic complementation in recombinant BCG did not result in immunomodulatory changes or altered protection in mouse models . Our results thus put into perspective the numerous reports on virulence-associated immunomodulatory impact of individual PE_PGRS and PPE-MPTR proteins and open novel questions on their biological function ( s ) . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [
"protein",
"transport",
"blood",
"cells",
"bacteriology",
"medicine",
"and",
"health",
"sciences",
"immune",
"cells",
"immune",
"physiology",
"pathology",
"and",
"laboratory",
"medicine",
"pathogens",
"cell",
"processes",
"immunology",
"microbiology",
"vaccines",
"physi... | 2018 | RD5-mediated lack of PE_PGRS and PPE-MPTR export in BCG vaccine strains results in strong reduction of antigenic repertoire but little impact on protection |
Metabolic flux is frequently rerouted through cellular metabolism in response to dynamic changes in the intra- and extra-cellular environment . Capturing the mechanisms underlying these metabolic transitions in quantitative and predictive models is a prominent challenge in systems biology . Progress in this regard has been made by integrating high-throughput gene expression data into genome-scale stoichiometric models of metabolism . Here , we extend previous approaches to perform a Temporal Expression-based Analysis of Metabolism ( TEAM ) . We apply TEAM to understanding the complex metabolic dynamics of the respiratorily versatile bacterium Shewanella oneidensis grown under aerobic , lactate-limited conditions . TEAM predicts temporal metabolic flux distributions using time-series gene expression data . Increased predictive power is achieved by supplementing these data with a large reference compendium of gene expression , which allows us to take into account the unique character of the distribution of expression of each individual gene . We further propose a straightforward method for studying the sensitivity of TEAM to changes in its fundamental free threshold parameter θ , and reveal that discrete zones of distinct metabolic behavior arise as this parameter is changed . By comparing the qualitative characteristics of these zones to additional experimental data , we are able to constrain the range of θ to a small , well-defined interval . In parallel , the sensitivity analysis reveals the inherently difficult nature of dynamic metabolic flux modeling: small errors early in the simulation propagate to relatively large changes later in the simulation . We expect that handling such “history-dependent” sensitivities will be a major challenge in the future development of dynamic metabolic-modeling techniques .
In response to environmental changes , microbes modulate their metabolic activity through a complex interplay of biochemical and regulatory networks . The dynamics of these changes is a poorly understood process , relevant for many applications ranging from infectious diseases to environmental remediation . With the rise of genome-scale stoichiometric models of metabolism [1] , these challenges have been addressed through the development of algorithms that overlay gene expression data onto these models to quantitatively study the effects of genetic regulation on cellular metabolism . One of the most widely used approaches for genome-scale predictions of metabolic fluxes is Flux Balance Analysis ( FBA ) [2]–[5] . FBA uses a steady state approximation and linear programming to determine optimal solutions to the problem of allocation of metabolic resources through a metabolic network . Several FBA-based methods have been proposed to integrate measurements of mRNA abundance , often with the goal of improving the prediction of fluxes in a metabolic network . One general approach is to constrain the maximum flux through reactions whose catalyzing enzyme genes have low expression levels . Some examples of this strategy include regulatory FBA ( rFBA ) [6] , steady-state regulatory FBA ( SR-FBA ) [7] , integrative FBA ( iFBA ) [8] , E-Flux [9] and Probabilistic Regulation Of Metabolism ( PROM ) [10] . Another way to integrate context-specific data is to match changes in flux with statistically significant changes in mRNA levels over time , a strategy employed by Metabolic Adjustment by Differential Expression ( MADE ) [11] . Yet another strategy is not to constrain fluxes directly , but instead to penalize reactions whose fluxes deviate from their coding genes' expression by introducing a cost function to be minimized . Two examples of this strategy include that of Shlomi et al . [12] and Gene Inactivity Moderated by Metabolism and Expression ( GIMME ) [13] . GIMME , the method upon which we will build in this work , is a particular extension of FBA that maximizes metabolic consistency with gene expression data , producing a set of fluxes that both satisfy the stoichiometric constraints of the metabolic model and provide a context-specific prediction that is informed by experimental data . These methods vary widely in the details of their implementation , but they all ultimately have to grapple with a number of common obstacles and limitations [14] . Most notably , irrespective of whether regulatory information is used as a constraint or as part of the objective , these approaches require making some assumptions on how mRNA expression levels end up affecting fluxes . Part of the problem is the complex relationship between mRNA expression and protein levels [15] . In this respect , methods that use expression levels as part of the objective ( e . g . , through maximization of consistency ) rather than as hard constraints , have the advantage of allowing a certain flexibility in the final choice of flux values . An additional issue associated with the use of mRNA levels to inform fluxes is the necessity ( in many but not all approaches ) to choose a universal threshold below which expression can be effectively deemed unlikely to support flux . It is also important to note that most of the literature on integration of expression with flux balance modeling focuses on static cases , without exploring the feasibility and potential issues associated with the application to time course data . Here , in an attempt to advance our understanding of the interplay between metabolism and regulation in time-dependent processes , we present a new algorithm named TEAM ( Temporal Expression-based Analysis of Metabolism ) . TEAM integrates dynamic flux balance analysis ( dFBA ) [16] with time-dependent gene expression data , using a cost minimization scheme similar to GIMME [13] . In addition to representing a unique example of integration of time-dependent gene expression with dFBA , the TEAM approach introduces some important innovations relative to the GIMME method . In particular TEAM takes advantage of an additional large compendium of gene expression data [17] to estimate gene-specific expression penalties , effectively taking into account the individuality of expression patterns identifiable in different genes . Furthermore , through TEAM , we introduce a new , simple sensitivity analysis that helps estimate the predictive power of the approach under different choices of parameters . For a succinct overview of the TEAM algorithm , refer to Figure 1 . To test TEAM's ability to predict bacterial behavior in the face of changing environmental conditions , we apply it to data collected during batch growth of Shewanella oneidensis MR-1 under minimal lactate aerobic conditions [18] . S . oneidensis is a dissimilatory metal-reducing gammaproteobacterium that was discovered in Lake Oneida , NY in 1988 and has since been shown to be able to utilize over 20 different electron acceptors [19]–[21] . This unusual ability allows Shewanella species to adapt to many different habitats that often contain oxic/anoxic transition zones [19] and an abundance of various fermentation products such as lactate , formate and hydrogen [22] . In the experiment we use for our analysis , extracellular metabolites ( high performance liquid chromatography , HPLC ) , gene expression levels ( Affymetrix microarrays ) , and population size ( optical density , OD ) were measured over the course of 50 hours ( see also Materials and Methods ) . Similar to previous S . oneidensis growth experiments [23] , [24] , these data displayed excretion and re-uptake of acetate and pyruvate during growth , a pattern that could not be explained by regular dFBA simulations [18] . Eiteman et al . attribute this phenomenon , called overflow metabolism , to the imbalance between the enzymatic capacity of the TCA cycle to fully oxidize acetyl-CoA and the rate of carbon consumption [25] . The excess production of NADH by the TCA cycle is thought to repress the TCA cycle genes themselves , forcing the usage of anaerobic pathways that do not produce NADH , such as the acetate generation pathway . One of the goals we set in developing TEAM is precisely to be able to reconcile gene expression data with metabolic constraints , to help understand otherwise indecipherable metabolic behavior , such as the metabolic overflow observed in the HPLC data . More generally , we propose TEAM as a strategy for marrying two often detached views of bacterial physiology: environmental resource utilization and internal enzymatic functional states . The complexity that arises when these two views are integrated allows one to draw important conclusions about the behavior of bacteria that may not have been previously possible .
Our interest in exploring novel avenues for integrating dynamic models of metabolism with measurements of gene expression was partially motivated by the desire to account for metabolic behaviors that could not be predicted through regular dFBA , such as the overflow metabolism in S . oneidensis described above . Specifically , we wanted to recapitulate three striking features of the experiment as observed from the HPLC data: ( 1 ) the nearly simultaneous exhaustion of all carbon sources and ammonium in the media , ( 2 ) the excretion and subsequent re-uptake of acetate from the media , and ( 3 ) the excretion and subsequent re-uptake of pyruvate . These experimental measurements are shown in Figure 2A . The results of a standard dFBA simulation for this system are shown in Figure 2B ( see Materials and Methods for details ) . Using conventional dFBA we were able to qualitatively match the predicted lactate and ammonium dynamics to those in the collected data . However , the method failed to predict the correct depletion times for both lactate and ammonium , and also failed to account for the presence of acetate and pyruvate . The failure of dFBA to capture some unique features of our experiment led us to try to incorporate gene expression into our simulation . We did so by merging GIMME with dFBA into a preliminary version of TEAM . Each iteration of TEAM completes a GIMME optimization , mathematically formulated , in analogy to other FBA algorithms ( [2] ) ( see Materials and Methods for details ) as:where S is the stoichiometric matrix , lbi and ubi are , respectively , the lower and upper bounds of flux Vi , VBM is the biomass production rate ( or growth flux ) and ci is a penalty assigned to reaction i based on the expression of its constituent genes . TEAM's implementation of GIMME diverges from the original implementation in [13] in two ways . First , TEAM uses experimental measurements of biologically necessary fluxes , most notably the growth rate ( or , potentially , of any exchange flux ) to impose specific magnitudes to the corresponding fluxes in the model . This is in contrast to [13] , where the minimal flux passing through each required metabolic functionality ( RMF ) was calculated as some percentage ( a free parameter in the system ) of the maximal flux which could pass through that RMF ( as calculated using flux balance analysis , see Materials and Methods ) . Second , and most importantly , TEAM and GIMME differ in how the coefficients ci of the penalty function are calculated . This penalty was calculated in [13] by first propagating expression measurements from each annotated gene in the model to its corresponding reaction using the Boolean gene-to-reaction mapping rules . Then , if the expression associated with a reaction exceeded a user-defined threshold p , that reaction was assigned a penalty ci of zero . Otherwise , the reaction was assigned a penalty equal to the difference between the threshold and its expression . In TEAM , we modified this protocol by first calculating the penalty of each gene , and then propagating this penalty up to each reaction . In the following section , we will show how this small change enabled us to incorporate more data on the expression characteristics of each gene into TEAM to generate markedly better predictions . As a first trial , we assigned a common , global penalty threshold ( herein referred to as a Type 1 threshold ) to all genes in the model . We tested many thresholds , and found that the threshold falling in the θ = 70th percentile of all expression measurements from the microarrays in our experiment appeared to give the most accurate predictions . We will defer from commenting on the quantitative accuracy of this version of TEAM until later on , when we will do so not only across all possible penalty thresholds , but also across different methods of assigning such thresholds . As shown in Figure 2C , the TEAM simulation with Type 1 penalty threshold was able to reasonably capture the qualitative dynamics of acetate in addition to lactate and ammonium . Although the magnitude of TEAM's predicted acetate dynamics greatly overestimated the experimental data , the results were nevertheless promising . The predicted acetate dynamics showed significant improvement over the dFBA simulation , which exhibited no acetate dynamics whatsoever . However , despite testing of many penalty thresholds , we were not able to find any TEAM simulations with Type 1 thresholding which displayed any pyruvate secretion/uptake . This , combined with a great deal of variability in the timing and magnitude of acetate dynamics , prompted us to search for ways to refine TEAM . Despite our success in using TEAM to recover acetate dynamics , we were still unable to capture the dynamics of pyruvate in the media . We began to consider the possibility that assigning an identical penalty threshold to each gene in the model was causing us to lose valuable information regarding the likelihood that each gene was active . This motivated us to inspect the distribution of gene expression values for each gene in the S . oneidensis model . We assembled a compendium of gene expression data for S . oneidensis using the M3D database [17] . For each gene in the database , we generated a histogram of gene expression values built from all the available microarrays in the database and supplemented with our own microarrays from the current experiment . A representative sampling of these gene expression histograms is shown in Figures 3A and 3C . It became quite clear that two genes in the model could have significantly different expression characteristics . This is illustrated in Figures 3A–3D , which show the distribution of expression measurements for two enzymes essential to lactate metabolism in S . oneidensis . The distribution of measured expression for D-lactate dehydrogenase was found to be tightly centered around its mean with a very small standard deviation . In contrast , expression for acetate kinase exhibited a much broader multi-modal distribution with a significantly higher standard deviation . This variability in the distribution of expression values became even more striking when plotting the distribution of the means and standard deviations of expression measurements across all genes , shown in Figures 3E and 3F , respectively . A gap of over two orders of magnitude was observed over all genes in the model . Biologically , the disparities in expression signatures among the genes in S . oneidensis may have arisen from a variety of biological sources . One possibility is that some genes may code for mRNAs with relatively high translational efficiency , or for enzymes with relatively high catalytic rates , thus requiring fewer mRNAs in order to achieve an identical metabolic flux . Another possibility is that the products of some genes may be constantly required for the operation of the cell ( such as the enzymes of central carbon metabolism ) , while others are only needed in particular situations ( such as transporters for specific carbon sources ) . Prompted by the observation that individual genes showed unique expression characteristics , we developed two new methods for calculating penalty thresholds customized to each gene in the model . In the first ( referred to herein as Type 2 ) , we used our compendium of gene expression data to calculate a cumulative distribution function ( CDF ) for each gene in the metabolic model . Then , a common percentile θ was chosen for all genes in the model . Next , we used each gene's CDF to assign the expression level corresponding to this percentile as the penalty for that particular gene . We then calculated a penalty for each and propagated this penalty to each reaction in the model as described earlier and in the Materials and Methods . The second new thresholding method ( Type 3 ) proceeds exactly as Type 2 thresholding , except that each gene's penalty is now normalized by that gene's standard deviation , as calculated from our compendium of expression data . Upon completing TEAM simulations with these two new thresholding methods for the same θ as the most accurate Type 1 simulation , we immediately observed the appearance of excretion and subsequent re-uptake of pyruvate in the media ( Figures 2D , 2E ) . Next , we sought to systematically assess whether our refined penalty methods show significantly improved predictive capabilities when compared to dFBA and TEAM with Type 1 thresholding . We pursued this in two ways . First , we studied the predicted secretion patterns of all of the TEAM methods across the entire range of potential penalty thresholds θ . To do so , we calculated the total amount of each different carbon source found in the media over the entire course of the simulation for each penalty threshold . We did this for each penalty threshold and for all three TEAM methods . The results are shown in Figures 4 , S1A , and S2A . These figures highlight that only the Type 2 and 3 TEAM methods with unique penalties for each gene were able to predict the excretion and re-uptake of pyruvate . The Type 1 method failed to predict any pyruvate dynamics in the external media for the entire range of possible penalties . Despite predicting pyruvate , in a small range of penalty thresholds the Type 2 and 3 methods also spuriously predicted the excretion and re-uptake of formate and glycolate , two intermediary metabolites which we confirmed were not present in the experiment . As a second step towards assessing the effect of different thresholding methods on secretion patterns , we developed a quantitative assessment of their relative predictive accuracy . We decided that because we were most concerned with recapitulating the excretion and re-uptake of pyruvate and acetate , we would focus on each method's ability to accurately predict the dynamics of these metabolites . For a given simulation , we calculated the residual squared error between the predicted concentration of acetate and pyruvate in the media and summed over all time points . The total error was plotted against penalty threshold for all three TEAM methods and shown in Figure 5 . The results illustrate that by accounting for the individuality of genes , the two refined TEAM methods performed at least as well or better than the original method for all penalty thresholds . For penalty thresholds in the range of 30% to 70% , the refined methods perform significantly better , while at either extreme of the thresholds , the difference between methods is smaller . For all three TEAM methods , we found that changing the penalty threshold had a large impact on the quantitative accuracy of our model , and enabled us to make an informed choice of a penalty threshold which seemed to best match our experimental observations . Our promising results using custom thresholds with S . oneidensis prompted us to test whether accounting for the heterogeneity of gene expression would also facilitate the integration of gene expression in flux balance models across other datasets . Specifically , we tested TEAM on a yeast growth transition dataset [26] , previously used to evaluate the performance of the MADE approach [11] , as well as on experimental data on the behavior of a synchronized yeast population undergoing metabolic oscillations [27] . In both cases we found that gene-specific thresholds ( Type 2 ) improve the consistency of flux predictions with gene expression data ( Figures S4 and S5B ) , as well as the capacity to predict metabolite secretion ( Figure S5A ) . Given that our refined penalization methods ( Types 2 and 3 ) produced quantitatively more accurate results than the original ( Type 1 ) method , we next inspected how varying the penalty threshold for these refined methods influenced the predicted dynamics of pyruvate , acetate , glycolate , and formate secretion . As shown in Figure 4 , as the Type 2 and Type 3 penalty thresholds increase , zones of qualitatively different behavior emerge . Acetate is always excreted regardless of the penalty threshold ( Figure 4A ) . Glycolate , formate and pyruvate , however , are only excreted in the intermediate zone between penalty thresholds θ = 45% and θ = 72% . Furthermore , as shown in Figure 4B , in this intermediate zone , lactate is completely consumed within 28 to 30 hours , while in the peripheral zones it is consumed between 30 and 34 hours . In this intermediate zone , we find that the early exhaustion of lactate is strongly correlated to high concentrations of intermediate carbon sources ( pyruvate , acetate , formate , and glycolate ) in the media . In a very narrow range of thresholds , from θ = 65% to θ = 72% , we observe the secretion of acetate and pyruvate , but not glycolate and formate . This qualitative agreement led us to identify this range of thresholds as the “optimal range” within which we expected TEAM's predictions of metabolic activity to be most accurate . However , because we did not obtain any measurements of internal fluxes from the experiment , we were unable to further explore how these predictions correlated with in vivo fluxes . Instead , we turned to studying TEAM's novel predictions of formate and glycolate secretion . Although these two metabolites were not observed in the HPLC measurements , their appearance in TEAM's predictions suggests that S . oneidensis may be capable of secreting the two metabolites under some as-yet unidentified conditions . We decided to investigate in fine detail the mechanisms linking the secretion of formate and glycolate to lactate exhaustion . This scenario is analyzed in Figure 6 , which highlights , at the individual flux level , several of the dramatic differences in TEAM predictions as the penalty threshold is increased . In one of the three qualitatively different behaviors observed ( at a threshold of 65% ) , lactate is imported significantly faster than the measured rate in HPLC . This excess of imported carbon is then funneled through several pathways including the TCA cycle , the glyoxylate shunt , a formate-producing cycle , and acetyl-CoA synthetase . Each of these pathways results in the production of carbon compounds ( CO2 , glycolate , and acetate , respectively ) which are excreted into the media . In contrast , the simulation at a threshold of 85% displays a more tempered rate of lactate usage , leading , through the TCA cycle , to secretion of CO2 and acetate . Given that our algorithm minimizes the sum of the absolute values of fluxes , it is somehow surprising that , in the 65% threshold regime , TEAM would predict overflow metabolism . Can this be explained in terms of actual energetic requirements for the cell ? We found that the increase in NADH produced as a result of importing excess lactate and metabolizing it via lactate dehydrogenase ( which produces pyruvate and NADH from lactate and NAD+ ) provided adequate reducing power to the cell . The resulting pyruvate is then converted into whichever intermediate carbon sources ( acetate , formate , or glycolate ) minimize the inconsistency between gene expression and flux . Previous studies using isotope tracing to infer flux have reported similar increased activity of both the glyxoylate shunt and a proposed serine oxidation cycle producing formate in S . oneidensis in aerobic , carbon-limited conditions [23] . Here , our simulations suggest that the transcriptional response of S . oneidensis to changing environmental conditions dictates the routing of flux into these pathways . Our investigation of pyruvate dynamics led us to another curious but intuitive observation: we found that the availability of a large repertoire of intermediate metabolites early in the time course led to a high diversity of metabolic activity later on in the simulation . Because these metabolites can be funneled through a larger variety of pathways than lactate , the model is able to select from among all these pathways to find the minimally penalized reaction path . For example , for several hours in the top panel of Figure 6 , TEAM predicts that both glycolate and formate are secreted into the media . This means that later on , TEAM has the option of importing either one of these carbon sources , but actually imports glycolate first and then formate . This is a direct result of a high penalty associated with pyruvate formate lyase required to utilize formate and no penalty associated with the reactions required to import glycolate . Thus , the model chooses the sequence of carbon source usage in best agreement with the gene expression . In contrast , for a higher penalty threshold in the bottom panel of Figure 6 , TEAM has no access to formate and glycolate in the media . This means that while the gene expression is identical to the intermediate zone , a different set of environmental conditions results in starkly different behavior .
The growing abundance of high throughput gene expression datasets has led to a call for methods integrating these experimental data with stoichiometrically based genome-scale models of metabolism . Our implementation of TEAM explored some of the challenges associated with developing such methods . In particular , we found useful ways of incorporating assorted data types ( OD , microarray data ) to constrain some of the otherwise free parameters of TEAM . We discovered that accounting for the heterogeneity of expression across different genes leads to an increase in predictive accuracy . Most importantly , it was simple to identify those penalty thresholds expected to be the most accurate , simply by matching qualitative predictions ( i . e . acetate and pyruvate secretion ) to experimental observations . Despite these successes , we still observed qualitatively broad shifts in TEAM's predictions as certain parameters varied , and we introduced a simple technique for sensitivity analysis which teased out precisely where these shifts took place . This sensitivity analysis enabled us to identify a narrow range of penalty thresholds , within which we were confident of TEAM's predictions . This suggests that in future analyses , it may be more appropriate to report a summary of results across the whole spectrum of thresholds , using a metric of agreement with experimental data as a criterion for choosing the “optimal threshold” . We suggest that such sensitivity analyses should become a central component of future efforts to integrate gene expression with flux balance models . A common thread that ran through each of our successive improvements to the original GIMME algorithm was the use of experimental measurements to improve the predictive accuracy of TEAM . Rather than use all of our data to evaluate the performance of TEAM , we found that some types of data were better suited to generating more informed models , while others seemed to be more useful in validation . In particular , one user-defined parameter from the original GIMME algorithm ( the minimal RMF flux ) was completely eliminated simply by linking its value to the observed experimental biomass flux . Another parameter , the penalty threshold of each gene , morphed from a common value for all genes to a quantity unique to each gene and directly determined by prior measurements of that gene's typical expression behavior . The elimination of these otherwise relatively unconstrained parameters enabled us to systematically evaluate the performance of TEAM . Furthermore , these improvements came at very little cost in terms of experimental effort . The collection of OD data is standard in metabolic engineering , and our supplementary microarray data was freely available in the M3D database . Building on prior work on the GIMME algorithm , we assessed TEAM's sensitivity to penalty thresholds and concluded that broad , qualitative changes in TEAM's predictions ( such as the appearance of glycolate and formate in the media ) were not due to changes in the penalization of a single or small group of genes . Instead , it was the total consistency of fluxes over the entire network that led to these shifts in TEAM's behavior . In many cases , we found two genes in the same pathway in S . oneidensis exhibited opposing expression behavior ( i . e . one gene's expression would be rising , while the other's would simultaneously fall ) . By integrating these expression measurements with a model that enforces mass-balance constraints , TEAM was able to reconcile otherwise conflicting signals and output a coherent pattern of metabolic fluxes that best fit the available data . This highlighted the value that methods integrating expression data with metabolic models have over more classical techniques for analyzing expression data in isolation , like simple pathway enrichment . Incompatible trends in the expression of the enzymes of one metabolic pathway were made much more coherent by connecting them to the operation of the metabolic network as a whole . Looking carefully at our predictions , we found that even the best TEAM predictions did not precisely match the timing and magnitude of acetate and pyruvate dynamics from the experimental data . While there may be many sources for the discrepancies between TEAM's predictions and the data , one prominent and unresolved question regards the error associated with using mRNA abundance as a proxy for the activity of a metabolic reaction ( typically related to the total concentration of enzyme in the cell ) . Recently , a number of experimental studies [15] , [28] have addressed the question of correlation between mRNA and protein abundance . While there is some correlation between mRNA and protein levels , it now appears that a more relevant question is the relationship between the half-lives and production rates of both mRNA and protein . In particular , Schwanhäusser et al . [15] showed that different genes displayed characteristically different combinations of mRNA and protein half-lives . These combinations were linked to a model of energetic resources in the cell , based on the argument that different blends of mRNA and protein stability may be associated with the functional role a particular protein plays within the cell [15] . Although difficult to obtain , information about protein half-lives could be directly integrated into TEAM by calculating a gene's penalty based on its expression integrated over a time interval . This may lead to delays in the onset of a penalty , as well as penalties that remain active for long periods of time . It is noteworthy that such time-dependent improvements would heavily rely on TEAM's dynamic nature; static simulations of GIMME would be unable to capture the diversity of dynamic behaviors in mRNA and protein . Finally , even the integration of precise proteomics data needs to be treated with care . Fendt et al . [29] report cases in which changes in metabolite concentration correlate both positively and negatively with enzyme concentration , suggesting that one should not necessarily expect strong correlations between metabolic flux and enzyme abundance . Our study of the predicted appearance of glycolate and formate in the media led us to another major conclusion: a spurious prediction about the excretion of metabolites early in a simulation can lead to very significant qualitative errors from TEAM later on . The difficulty was that the gene expression TEAM used was intimately tied to very specific environmental conditions . If TEAM predicted media conditions that included nutrients not found in the true experimental conditions , then the simulation had access to certain metabolic pathways ( for example , C1 metabolism of formate ) which could not have been active in the experiment . We reasoned that by imposing adequately high conformity with gene expression with high penalties , we would be able to prevent this spurious behavior . In fact , this is precisely what we observed: at high levels of penalty threshold , we no longer found glycolate or formate present in the media . The disappearance of these two metabolites was directly linked to a reduction in the import of lactate early in the simulation . In general , along the time course , there is a tight mutual dependence between the rates of metabolite uptake/secretion , and the transcriptional regulation of the pathways for producing or utilizing those metabolites . For S . oneidensis , this amounted to the rapid intake of lactate ( faster than required if oxidative phosphorylation were used , but slower than required if glycolate and formate were secreted ) , resulting in the overflow metabolism associated with the secretion of acetate and pyruvate . We expect that future efforts to develop dynamic genome-scale metabolic models will encounter similar temporal sensitivity issues . Improvements in accuracy will depend on the ability to prevent predictions of qualitatively spurious media conditions . Finally , the history-dependent sensitivity of TEAM underscores the underappreciated interplay between gene expression and environmental conditions . The upregulation of genes associated with a particular pathway is frequently used as a proxy for inferring increased metabolic activity in the pathway itself , e . g . in the analysis of large expression datasets associated with human disease , such as cancer . Our work with TEAM suggests that the inference of metabolic activity directly from gene expression data can be quite misleading . In addition to effects associated with the delay between transcriptional and metabolic response , distinct extracellular environments , coupled with identical gene expression profiles , can re-organize the activity of metabolic pathways in substantially different ways . Therefore , we would argue that future studies of metabolism must carefully account for the environmental context within which gene expression is measured .
TEAM uses four sets of data as inputs , in addition to the stoichiometric model: time-dependent , OD-based , biomass measurements ( OD ) , time-dependent gene expression microarray measurements ( EXP ) , a reference compendium of gene-expression data unrelated to the current experiment ( in our case , obtained from the M3D [17] database and labeled M3D ) and initial concentration of nutrients in the growth medium ( MEDIA ) . To make the data sets OD , EXP , M3D , and MEDIA congruent with each other and with the algorithm architecture , the data is interpolated for the appropriate time interval across the entire experimental period . We used a time interval Δt of 1 hour , and performed the interpolation using the Matlab interp1 function . The reactions in the stoichiometric model can be characterized as either exchange or biological fluxes . Biological fluxes are associated with enzyme-catalyzed metabolic reactions and transport reactions . Exchange fluxes act as source and sink reactions that balance the biological fluxes . Formally we define these two sets as follows: ( 1 ) ( 2 ) By convention , a positive flux through an exchange reaction means that a metabolite is secreted , and conversely a negative exchange flux corresponds to uptake of a metabolite . Therefore , a lower bound on an exchange flux is equivalent to the maximal uptake rate of the corresponding transportable metabolite in a time interval Δt . TEAM is based on the previously described dFBA [16] and GIMME [13] methods . For a comprehensive overview of the TEAM method , see Figure 1 . TEAM produces a time-series of metabolic flux predictions V ( t ) by identifying , at each ( discrete ) time t , the metabolic flux distribution that is most consistent with measured gene expression data at that time . The resulting flux distribution is assumed to be valid for a time interval Δt , and is used to update the concentrations of nutrients in the media for the subsequent time interval . In this way , a series of static optimizations are linked to each other by the repeated updating of external media nutrient availability . A TEAM simulation is initialized by setting the initial concentration of nutrients in the growth medium . Let ei ( t ) represent the concentration ( in mM , considering a working volume of 1liter ) of the ith component of the medium at time t . We initialize e to reproduce the experimentally known initial medium composition at time t = 0: ( 3 ) Next , we initialize the problem so that a dFBA iteration can be completed . We use the current metabolite concentrations to infer the lower bounds ( meaning maximal possible inflow ) on all exchange fluxes: ( 4 ) Furthermore , we set the initial biomass concentration in TEAM equal to the appropriately scaled experimentally measured optical density: BM ( 0 ) = OD ( 0 ) . In order to consistently solve the problem for the total biomass available , at each time point we convert the constraints on the biological fluxes from specific ( lb ( 0 ) , ub ( 0 ) ) , defined per unit of biomass , mmol/gDW·hr , as in standard FBA , to total ( lb , ub , in mmol/hr ) : ( 5 ) ( 6 ) Next , in analogy with [13] , we determine the minimal flux through a required metabolic functionality ( RMF ) . Imposing RMF fluxes in TEAM is necessary in order to prevent the output of the trivial flux distribution V = 0 . Because TEAM attempts to minimize the inconsistency between a flux distribution and gene expression data , the trivial solution is always optimal unless the user explicitly makes it infeasible . The only RMF used in this work is biomass production , although the TEAM protocol will work equally well for any other choice of RMF . In this manuscript , measurements of the biomass flux were collected through growth data ( OD ( t ) ) . When this is the case , TEAM can explicitly calculate the lower bound on the biomass flux to be ( 7 ) where d is the death rate . In this work , we used d = 0 . 06 . The heart of the TEAM algorithm performs two optimization steps . In the first step , following [13] , we minimize the inconsistency between metabolic flux and gene expression data . To do so , we associate a penalty , ci ( t ) , with every reaction i in the model ( see Gene Penalty Calculation section of Materials and Methods ) . This penalty reflects our expectation , based on gene expression , that a reaction is “inactive , ” i . e . that it is unlikely to carry flux . The total penalty is minimized by solving the linear programming problem: ( 8 ) Because there may be many alternative optimal solutions , we complete a secondary optimization to select the one that minimizes the sum of absolute value of all fluxes , while keeping the inconsistency constant . The minimization of the sum of absolute values of fluxes had been described before [30] , and can be formulated as: ( 9 ) To complete the dFBA iteration , the media nutrient concentrations and the biomass are updated using the newly calculated exchange fluxes: ( 10 ) ( 11 ) The vector e is then used to assign lower bounds to all exchange fluxes in the next time step of TEAM using Equation 4 . Gene penalties are used in the main optimization step of TEAM to identify flux states that minimize the flux through reactions with relatively low expression . As described in detail below , gene penalties are determined by comparing the expression value of a gene with a predefined threshold . Gene penalty calculation is done in three steps: threshold determination , expression comparison , and gene-to-reaction conversion . We applied TEAM to data obtained from a growth experiment of S . oneidensis MR-1 in carbon-limited conditions , as described by [18] . The data set contained three types of data: a time-series gene expression data set , a time-series media metabolite concentration data set , and a time-series measurement of biomass . The gene expression data was measured using an S . oneidensis MR-1 microarray Affymetrix chip platform and included 19 measurements at various time points between 0 and 50 hours . To derive the background data set used to calculate gene-specific penalty thresholds , we combined this set of 19 microarrays with 262 compatible microarray data sets obtained from the M3D database [17] . All gene expression data was normalized using the “affyrma” function in the Bioinformatics Toolbox in MATLAB [31] . The external metabolite concentration data set was measured using high performance liquid chromatography ( HPLC ) , and provided an abundance profile for various metabolites in the media over time . This data set was not integrated directly using TEAM , but was used to measure the performance of our method against experimental data , as well as to define our starting media condition . Finally , the biomass growth dataset was measured using optical density ( OD ) . This data was used in conjunction with a genome-scale metabolic model iSO783 of S . oneidensis MR-1 described in [21] . This model contains 774 reactions encoded by 783 genes , and 634 unique metabolites . The model includes the gene-to-reaction mapping used to associate gene information with the reactions in the metabolic model , as described above . | Understanding the dynamic response of microorganisms to environmental changes is a major challenge in systems biology . In many cases , these responses manifest themselves through changes in gene transcription , which then propagate to adjust flow through metabolism . Here , we implement a Temporal Expression-based Analysis of Metabolism ( TEAM ) by dynamically integrating a genome-scale model of the metabolism of S . oneidensis with high-throughput measurements of gene expression and growth data . TEAM recapitulates the complex cascade of secretion and re-uptake of intermediary carbon sources that S . oneidensis exhibits in the experimental data . We show that these complicated metabolic behaviors are best captured when TEAM explicitly accounts for each gene's unique transcriptional signature . Furthermore , by way of a newly proposed sensitivity analysis , we reveal and study the inherent difficulty of dynamic metabolic flux modeling: small changes early in a simulation can easily spread and lead to significant changes towards the end of it . We expect that further development of robust dynamic flux balance methods will need to overcome such “history-dependent” sensitivities in order to achieve increased predictive accuracy . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"microarrays",
"biochemical",
"simulations",
"genome",
"expression",
"analysis",
"microbial",
"metabolism",
"systems",
"biology",
"genomics",
"biology",
"computational",
"biology",
"metabolic",
"networks",
"microbiology",
"genetics",
"and",
"genomics"
] | 2012 | Temporal Expression-based Analysis of Metabolism |
ArtinM , a D-mannose binding lectin from Artocarpus heterophyllus , has immunomodulatory activities through its interaction with N-glycans of immune cells , culminating with the establishment of T helper type 1 ( Th1 ) immunity . This interaction protects mice against intracellular pathogens , including Leishmania major and Leishmania amazonensis . ArtinM induces neutrophils activation , which is known to account for both resistance to pathogens and host tissue injury . Although exacerbated inflammation was not observed in ArtinM-treated animals , assessment of neutrophil responses to ArtinM is required to envisage its possible application to design a novel immunomodulatory agent based on carbohydrate recognition . Herein , we focus on the mechanisms through which neutrophils contribute to ArtinM-induced protection against Leishmania , without exacerbating inflammation . For this purpose , human neutrophils treated with ArtinM and infected with Leishmania major were analyzed together with untreated and uninfected controls , based on their ability to eliminate the parasite , release cytokines , degranulate , produce reactive oxygen species ( ROS ) , form neutrophil extracellular traps ( NETs ) and change life span . We demonstrate that ArtinM-stimulated neutrophils enhanced L . major clearance and at least duplicated tumor necrosis factor ( TNF ) and interleukin-1beta ( IL-1β ) release; otherwise , transforming growth factor-beta ( TGF-β ) production was reduced by half . Furthermore , ROS production and cell degranulation were augmented . The life span of ArtinM-stimulated neutrophils decreased and they did not form NETs when infected with L . major . We postulate that the enhanced leishmanicidal ability of ArtinM-stimulated neutrophils is due to augmented release of inflammatory cytokines , ROS production , and cell degranulation , whereas host tissue integrity is favored by their shortened life span and the absence of NET formation . Our results reinforce the idea that ArtinM may be considered an appropriate molecular template for the construction of an efficient anti-infective agent .
Global immunization regimes have eradicated smallpox and controlled a large number of other infections [1 , 2] . Indeed , vaccines have been successful against infections caused by extracellular pathogens or those whose pathogenesis is mediated by toxins . Under these circumstances , vaccines confer protection by inducing antibodies that neutralize the inoculum and prevent the establishment of infections [3] . This effective formula is not applied to prevent infections with intracellular pathogens [4] because they require T-cell mediated immunity for elimination . Therefore , a research field has emerged that concerns the development of alternative prophylactic or therapeutic agents to enhance host cellular response . In this context , agonists of innate immunity receptors , especially Toll-like receptors ( TLRs ) , provide promising approaches [5] . The interaction of agonists with TLRs triggers cell signaling and production of inflammatory and anti-inflammatory mediators [6] . This process , beyond inducing early mechanisms of host defense , primes and orchestrates antigen-specific adaptive responses [7] . The ability of activated TLRs to modulate adaptive immunity motivates ongoing trials of new drugs based on natural or synthetic TLR ligands for infectious diseases in humans [8] . ArtinM , a D-mannose-binding lectin obtained from the seeds of Artocarpus heterophyllus , binds to TLR2 N-glycans and functions as a TLR2 agonist that exerts immunomodulatory properties [9] . The ArtinM interaction with TLR2 on macrophages and dendritic cells results in high levels of IL-12 production , driving immunity towards the T helper ( Th ) 1 axis [10] . This ability accounts for the protection conferred by ArtinM administration against Leishmania major [11] , Leishmania amazonensis [12] , Paracoccidioides brasiliensis [10 , 13] , Neospora caninum [14] , and Candida albicans [15] infections in mice . Beyond acting on antigen presenting cells , ArtinM exerts activities on lymphocytes [16] , mast cells [17 , 18] , and neutrophils [19 , 20] . This pleiotropic activity on immune cells is considered to account for the ArtinM property of conferring resistance against intracellular pathogens [21] . Concerning neutrophils , the cell type focused in this work , they are known to participate in the protection against intracellular pathogens , through mechanisms that involve phagocytosis , cell degranulation , ROS production , release of lipid mediators , and formation of neutrophil extracellular traps ( NETs ) [22] . Further mechanisms known to favor host defense are the release of cytokines combined with changes in cell life span [23] . Our previous work showed that ArtinM induces neutrophil migration by haptotaxis [24] , due to the concomitant interactions of ArtinM CRDs with N-glycans on neutrophil surface receptors , such as those linked to C-X-C chemokine receptor 2 ( CXCR2 ) , and glycoproteins of the extracellular matrix , such as laminin [25 , 26] . Also , ArtinM activates neutrophils , causing tyrosine phosphorylation , L-selectin shedding , and interleukin-8 ( IL-8 ) and leukotriene B4 ( LTB4 ) secretion . These responses result in the enhancement of phagocytic and microbicidal abilities of neutrophils [19 , 20] and indicate that ArtinM activates neutrophils hugely . Although effective against pathogens , neutrophils also account for exacerbated inflammation and tissue injury [27] , a fact that caused concerns regarding the possibility of using ArtinM to design a novel class of immunomodulatory agents acting through carbohydrate recognition . Although exacerbated inflammation was never observed in the ArtinM-treated animals , we always had concerns regarding the occurrence of inflammatory tissue injury . Indeed , it is unclear how ArtinM may take advantage of neutrophil activation to eliminate pathogens , without promoting tissue damage . Therefore , in this study , we focused on understanding the mechanisms through which neutrophils contribute to the protective effect of ArtinM against intracellular pathogens and how this process occurs without exacerbating inflammation .
The Ethics Committee of the Clinical Hospital of the Faculty of Medicine of Ribeirão Preto , University of São Paulo , approved this study ( Doc . Number: 10012/2009 ) and all the adult volunteers signed an informed consent form prior to blood and/or urine donation . ArtinM ( ID: Q7M1T4_ARTIN , available on UniProtKB database ) was extracted from Artocarpus heterophyllus seeds and purified by sugar affinity chromatography as previously described by Santos-de-Oliveira et al . ( 1994 ) [24] . The concentration of ArtinM lectin used to treat the neutrophils in this study was 2 . 5 μg/mL , as previously utilized [20] , except in some assays , as specified . Heparinized human blood was layered on a density gradient of a neutrophil isolation medium ( Monopoly Resolving Medium; ICN Pharmaceuticals , USA ) , as previously described by Toledo et al . ( 2009 ) [20] . The collected polymorphonuclear ( PMN ) cells were washed twice; a fraction was labeled with anti-CD16b antibody ( BD—Biosciences , USA ) , and the purity was analyzed by flow cytometry ( <95% , S1A Fig ) , and additionally by cytology from cytocentrifuged preparations Diff-Quick-stained ( Laborclin , Brazil ) ( S1B Fig ) . L . major ( LV39 strain ) promastigotes were grown in Schneider’s medium supplemented with 1% antibiotic-antimycotic , 10% heat-inactivated fetal calf serum ( FCS ) , 2 mM L-glutamine ( all from Invitrogen , USA ) , and 2% human urine , pH 7 . 2 . Neutrophils were pre-treated with ArtinM , and after 30 min , they were infected with stationary-phase promastigotes ( 5–7 day ) at a parasite-PMN ratio of 3:1 . Control neutrophils were not treated with ArtinM and not infected . Plates/tubes were centrifuged ( 300 × g/ 10 min ) and incubated at 37°C in RPMI 1640 or Hank’s Balanced Salt Solution ( HBSS ) in a humidified atmosphere containing 5% CO2 . Some assays were performed by infecting neutrophils with green-fluorescent L . major forms ( mβT3-LV39 ) , which were kindly provided by Prof . Angela Kaysel Cruz and Dr . Mônica Cristina Terrão . Human neutrophils ( 2 × 106 cells/mL ) were previously treated with ArtinM for 30 min , and infected with L . major . Infected/untreated neutrophils were used as controls . After 3 and 20 h post-infection , the cultures were cytocentrifuged ( 50 × g , 5 min ) and Diff-Quik-stained ( Laborclin , Brazil ) . The number of neutrophils with intracellular parasites was determined by count among 200 cells/condition/time , using an Olympus BX50 microscope coupled to a Nikon DXM-1200 photographic system . To analyze the intracellular leishmanicidal activity , we assessed the parasite viability , as described by Tavares et al . ( 2014 ) [28] , with slight modifications . Briefly , neutrophils that were treated with ArtinM and untreated controls were infected and cultured for 3 h with the parasites . Uninternalized parasites were discarded from cultures by washing twice for 5 min at 100 × g , and the culture was immediately fed with supplemented Schneider’s medium . The cells were then cultured at 26°C for an additional 48 h . The intracellular leishmanicidal activity was determined by assessing the number of extracellular motile promastigotes produced . To analyze if some degranulated content had leishmanicidal activity , we performed a free cell assay based on methodology previously demonstrated by Mikus and Steverding ( 2000 ) [29] . Neutrophils treated with ArtinM for 1 h and untreated controls were pelleted , and the supernatants of the cells were collected . The supernatants were incubated with antibodies to myeloperoxidase ( anti-MPO; ab62141; Abcam; 1/100 ) or neutrophil elastase ( anti-NE; ab21595; Abcam; 1/100 ) for 30 min . Control supernatants were not incubated with the antibodies . After the incubation , 1 × 105 parasites were added to the supernatants for 1 h , and then assayed for viability in Schneider’s medium containing Alamar Blue ( Life Technologies , USA; 1/10 dilution ) for an additional incubation of 24 h . Fluorescence measurement was performed on a FLx800 Fluorescence Microplate Reader ( BioTek Instruments , USA; excitation , 590 nm; emission , 635 nm ) . Fluorescence count data from unchallenged , serially diluted parasites were used to obtain a standard curve of viable parasites . Freshly isolated human neutrophils ( 2 × 106 cells/mL ) were previously treated with ArtinM for 30 min and infected with L . major . Neutrophils without ArtinM treatment and uninfected with L . major were used as negative controls . PMA stimulated ( 50 nM ) neutrophils were used as the positive control . Cytokine levels were quantified in the neutrophil supernatants 20 h after infection and at the same time in the supernatants of controls . TNF , IL-1β ( BD—Biosciences , USA ) , and TGFβ-1 ( R&D Systems , USA ) were measured by sandwich ELISA , according to the manufacturer’s instructions . The neutrophil chemiluminescence assay was performed in 96-well microplates using the procedure described by Lucisano-Valim et al . ( 2002 ) [30] , with slight modifications . Freshly isolated human PMNs ( 2 × 105 cells/well in HBSS ) were mixed with the chemiluminescent probes luminol ( 0 . 1 mM ) or lucigenin ( 0 . 1 mM ) . The mixture was incubated at 37°C for 3 min , and the reaction was initiated after adding ArtinM , phorbol myristate acetate ( PMA , 0 . 1 μM ) , formyl-Met-Leu-Phe ( fMLP , 1 μM ) , or HBSS . Some assays were performed by first incubating the cells with ArtinM during 30 minutes , and then by adding fMLP , PMA or L . major ( MOI 3:1 ) . The luminol- and lucigenin-enhanced chemiluminescence was measured in a microplate luminometer ( LB 960 Centro , Berthold Technologies , Germany ) , and light emission was recorded in photon counts per second ( CPS ) for 30–60 min , at 37°C . AUC represents the area under the time–course curve , which was used to determine the total amount of measured ROS . Neutrophil elastase activity was evaluated as previously described [31] , with some modifications . Briefly , 1 × 105 neutrophils were treated for 30 min with different concentrations of ArtinM ( 2 . 5 μg/mL to 312 ng/mL ) , fMLP ( 1 μM ) , or medium only , and NE activity was detected by using the substrate N-succinyl-alanine-alanine-valine-p-nitroanilide ( SAAVNA ) ( 1 mM ) , which is cleaved by the enzyme released in the supernatant , forming p-nitroaniline as one of the products , which was spectrophotometrically quantified , using the microplate reader PowerWaveX ( BioTek Instruments , USA; 405 nm ) . Myeloperoxidase , as well as elastase , are azurophilic granule contents of neutrophils . Their intracellular levels were measured in neutrophils after 20 h of incubation with or without ArtinM , and challenged or not with L . major . To detach ArtinM-treated neutrophils , it was necessary to use EDTA-glucose ( 10 μM/5 μM ) containing PBS ( pH 7 . 2 ) . Then , the cells were washed twice with PBS , and immediately fixed and permeabilized using the Cytofix/Cytoperm kit ( BD—Biosciences , USA ) , following the manufacturer’s instructions . Next , the cells were incubated with anti-MPOPE or anti-Hamster IgG1PE isotype control antibodies ( BD—Biosciences , USA ) , for 20 min . Immunofluorescent staining was analyzed by flow cytometry using Guava EasyCyte Mini ( Millipore , USA ) . Neutrophils ( 2 × 105 cells/well in HBSS ) were incubated with or without ArtinM , PMA ( 50 nM ) , or both together ( ArtinM+PMA ) in a 96-well plate for 2 and 4 h at 37°C , in 5% CO2 . NET-DNA was quantified using a modified version of a previously published method [32] . Bacterial endonuclease EcoRI ( 200 unit/mL ) was added for 10 min to partially digest any released NETs . Next , the cells and debris were pelleted by centrifugation at 600 × g for 10 min . A sample of the supernatant was added to SYTOX green nucleic acid stain ( Invitrogen; 1 μM ) in a black 96-well plate and incubated at 37°C for 10 min . NET-DNA fragments were quantified using a FLx800 Fluorescence Microplate Reader ( BioTek Instruments , USA; excitation 485 nm , emission 528 nm ) and the results have been expressed as Δ fluorescence values . Neutrophils ( 2 × 106 cells/mL in HBSS ) were incubated on poly-L-lysine-treated glass coverslips in a 24-well plate and treated or not with ArtinM or PMA , followed by incubation at 37°C for 6 h . Samples were collected , by gently removing coverslips , fixed with 3% paraformaldehyde ( 20 min at room temperature ) , and blocked in PBS supplemented with 3% FCS for 1 h at room temperature . Coverslips incubated overnight at 4°C with anti-neutrophil elastase antibody ( ab21595; Abcam; 1:200 ) were washed and incubated for 1 h at room temperature with Alexa Fluor488 goat anti-rabbit IgG secondary antibody ( Molecular Probes; 1:1000 ) , and then washed and mounted with ProLong Antifade containing DAPI ( Molecular Probes ) . Images were obtained using a Leica CTR 6000 fluorescence microscope , with Leica Application Suite software ( Wetzlar , Germany ) . To assess cellular morphology , freshly isolated human neutrophils ( 2 × 106 cells/mL in RPMI-1640 ) were treated with ArtinM , IL-8 ( 25 nM ) , Lysis Buffer ( ammonium chloride 70 mM , Tris 10mM ) , or non-treated , and incubated during 3 and 20 h at 37°C in 5% CO2 . After incubation , the cells were detached , as described above for MPO detection , cytocentrifuged onto a microscope slide using a Cytospin 3 cytocentrifuge ( Thermo Shandon , USA ) , Diff-Quick-stained ( Laborclin , Brazil ) and examined using light microscopy ( Olympus BX50—Olympus América INC , USA ) . Photomicrographs were taken with a Nikon DXM-1200 camera ( Nikon instruments , USA ) . DNA electrophoresis was performed to determine the effects of ArtinM treatment on neutrophils DNA degradation . The analysis was performed using a kit , the Wizard SV genomic DNA purification system ( Promega Corporation , USA ) . Briefly , after 24 and 48 h of incubation , 4 × 106 cells were washed twice with PBS and lysed , and genomic DNA was isolated . The extracted DNA was quantified using the NanoVue Plus spectrophotometer ( GE Healthcare , USA ) . A sample of 1 μg of DNA was analyzed using 1 . 5% agarose gel electrophoresis and stained with ethidium bromide . The DNA was then visualized under UV light on ChemDoc MP Imaging System and photographed using ImageLab software v . 4 . 0 ( both from Bio-Rad Laboratories , USA ) . Regarding the analysis of DNA fragmentation , the band area was quantified using ImageJ Software , and represented graphically in pixels2 . Neutrophil apoptosis was assessed by Annexin V staining [33] . Human neutrophils ( 2 × 106 cells/mL in RPMI-1640 ) , were treated with ArtinM , IL-8 , Lysis Buffer , or non-treated , and co-incubated or not with L . major during 3 and 20 h at 37°C in 5% CO2 . After incubation , the cells were detached , as described above ( MPO detection ) , washed once and suspended in 100 μL of annexin V binding buffer ( 140 mM NaCl , 2 . 5 mM CaCl2 , 1 . 5 mM MgCl2 , and 10 mM HEPES , pH 7 . 4 ) , containing annexin V-FITC or PE ( 1 μg/mL ) for 15 min . Some assays were performed on neutrophils infected with green-fluorescent L . major forms ( mβT3-LV39 strain ) , in order to detect if the dying cells are the ones infected . Immunofluorescence staining was analyzed by flow cytometry , using a Guava EasyCyte Mini instrument ( Millipore , USA ) . Human neutrophils ( 2 × 106 cells/mL in RPMI-1640 ) were treated with ArtinM , IL-8 , Lysis Buffer , or non-treated , and incubated during 3 and 20 h at 37°C in 5% CO2 . JC-1 staining solution ( 10 μM ) was then added to the wells and incubated at 37°C for 15 min . The fluorescent intensity for monomeric forms of JC-1 was measured using the FLx800 Fluorescence Microplate Reader ( BioTek Instruments , USA; excitation 485 nm , emission 528 nm ) . Statistical analyses were performed by Student t test , one way ANOVA followed by Bonferroni's post-test , and two way ANOVA followed by Bonferroni's post-test ( all using GraphPad Prism software version 6; GraphPad ) , as indicated at the graph . The p values <0 . 05 were deemed statistically significant .
To evaluate whether neutrophil activation by ArtinM contributes to its protective effect against intracellular pathogens [21] , we compared the internalization and elimination of L . major by neutrophils that were pre-treated or were not pre-treated with ArtinM . We assessed the effect of ArtinM treatment on the frequency of neutrophils with internalized parasites . At 3 h after in vitro infection , the number ( 15±0% ) of ArtinM-treated neutrophils with internalized parasites was 88% higher than the number ( 7 . 5±0 . 7% ) verified in untreated neutrophils ( Fig 1A ) . The difference increased 140% ( 14 . 5±0 . 7% vs 34±2 . 8% ) when the same assay was performed 20 h after infection ( Fig 1A ) . We also evaluated the leishmanicidal activity of the ArtinM-treated neutrophils . Cells were washed at 3 h post-infection and cultured in Schneider`s medium for additional 48 h . At this point , the number of mobile parasites ( 2 . 9x105±0 . 7x105 ) recovered from the culture of ArtinM-treated neutrophils was , on average , 50% lower than the number of viable parasites ( 5 . 8x105±1 . 3x105 ) recovered from untreated neutrophils ( Fig 1B ) . In conclusion , our data show that ArtinM treatment increased the ability of neutrophils to eliminate L . major , a fact that favors the idea that neutrophils can contribute to the protective effect of ArtinM against infection . The prominent production of inflammatory or anti-inflammatory cytokines by neutrophils contributes to an appropriate microenvironment for L . major elimination or survival , respectively [34 , 35] . We quantified the TNF , TGF-β , and IL-1β cytokines in the supernatant of non-treated or ArtinM-treated neutrophils that were infected or not with L . major . Twenty h after ArtinM-treatment , the supernatant contained 7-fold higher concentration of TNF ( 241 . 7±5 . 59 ρg/mL ) than the supernatant of untreated neutrophils ( 65 . 65±5 . 47 ρg/mL ) . A similar level was detected when the assay was performed with L . major infected neutrophils ( 245 . 77±3 . 93 ρg/mL and 93 . 77±0 . 68 ρg/mL—Fig 2A ) . On the other hand , TGF-β production was 3-fold augmented in L . major-infected neutrophils ( 169 . 6±32 . 01 ρg/mL ) , while ArtinM-treated cells , infected ( 56 . 99±16 . 73 ρg/mL ) or not ( 27 . 51±3 . 14 ρg/mL ) , released levels that were similar to the ones verified for untreated cells ( 54 . 94±15 . 27 ρg/mL—Fig 2B ) . IL-1β was not detected in the supernatant of untreated or L . major infected neutrophils . In contrast , we detected IL-1β production by ArtinM-treated neutrophils ( 14 . 37±2 . 19 ρg/mL ) , which increased by 6-fold when these ArtinM-treated cells were infected with L . major ( 94 . 67±7 . 01 ρg/mL—Fig 2C ) . Taken together , our results show that ArtinM treatment enhances the production of TNF and IL-1β , but not of TGF-β by human neutrophils . Instead of the high levels of TGF-β produced by L . major-infected neutrophils , the ArtinM pre-treated neutrophils released higher levels of TNF and IL-1β . Releasing of the contents of neutrophil granules contributes to elimination of pathogens [36] . In the case of L . major infection , elastase released by azurophilic granules of neutrophils plays an outstanding protective role in host responses [37] . Here , we examined the release of neutrophil elastase ( NE ) and myeloperoxidase ( MPO ) by cells that were treated or not with ArtinM . The occurrence of neutrophil degranulation was verified by detection of ( 1 ) decreased intracellular content of MPO , using flow cytometry analysis , and ( 2 ) by augmented levels of NE in the cell supernatant , assessed by enzymatic activity quantification . The intracellular MPO content in ArtinM-treated neutrophils was 4-fold lower than that verified in untreated cells ( Fig 3A ) . Consistent with this , 2-fold higher levels of NE activity were detected in the supernatant of ArtinM-treated neutrophils in comparison to that observed in the supernatant of untreated cells ( Fig 3B ) . This last assay was actually performed to verify the period necessary for occurrence of neutrophil degranulation in response to ArtinM , and the highest NE activity was detected shortly ( 5–60 min ) following treatment , even when the concentrations used were as low as 312 ng/mL of lectin , as demonstrated by a time-course assay ( S2 Fig ) . We also examined the neutrophil degranulation after L . major infection . Regarding MPO intracellular content , we verified that the infection per se was able to inhibit the degranulation of untreated cells ( 70%—Fig 3A ) . In contrast , the intracellular levels of MPO decreased drastically in the ArtinM-treated neutrophils ( 3-fold ) , showing that the lectin promotes degranulation even in L . major infected neutrophils ( Fig 3A ) . Therefore , ArtinM induces degranulation of both uninfected and infected neutrophils . To assess the leishmanicidal activity of the neutrophil secreted products , we incubated parasites with the supernatant of cells that were pre-stimulated with ArtinM . The number of viable L . major was 66% lower following incubation with the supernatant of ArtinM-treated neutrophils ( 2 . 0x104±0 . 6x104 ) than with the supernatant of untreated cells ( 6 . 0x104±6 . 4x104 ) . In order to evaluate the specific contribution of MPO and NE , released by azurophilic granules , to the observed leishmanicidal activity , the supernatant of ArtinM-treated neutrophils was pre-incubated with antibodies specific to MPO or NE , and then added to the parasite suspensions . Our results show that anti-MPO had no effect on the number of viable L . major ( 1 . 5x104±0 . 07x104 ) , whereas anti-NE antibodies inhibited the leishmanicidal activity provided by the supernatant of ArtinM-treated neutrophils once the number of parasites recover in this condition ( 5 . 5x104±0 . 2x104 ) was similar to that found on the supernatant coming from untreated neutrophils ( Fig 3C ) . These data suggest that the leishmanicidal activity of ArtinM-treated neutrophils is due , at least partially , to an NE-mediated mechanism . Besides killing microbes through its direct enzymatic activity [38–40] , NE also contributes to the formation of neutrophil extracellular traps ( NET ) [41] , which constitutes a mechanism for capture and kill microorganisms [42] and for host tissue damage [43] . To investigate whether neutrophil treatment with ArtinM could result in NET formation , the DNA concentration was measured in the supernatant of neutrophils , 2 and 4 h after ArtinM-treatment . The DNA levels detected in the supernatant of ArtinM-treated cells were as low as those found in untreated neutrophils . In contrast , neutrophil stimulation with PMA ( positive control ) resulted in the detection of high DNA levels ( > 5- fold , Fig 4A ) , at both periods analyzed . When we assayed L . major-infected neutrophils , we observed that untreated cells formed NET , in concordance with previous reports [44] , but , in contrast , ArtinM-treated cells do not form NET ( Fig 4A ) . The microscopic observation of the assayed cells was consistent with the DNA measurement in the neutrophil supernatant ( Fig 4B ) . Therefore , we concluded that the ArtinM treatment does not induce NET formation and inhibits the formation of NET that follows L . major infection . Once we had verified that ArtinM triggers intense neutrophil activation not associated with NET formation , we evaluated whether lectin induces the production of ROS , which is directly implicated in NET formation [45] . We monitored the levels of ROS following exposure to the ArtinM stimulus , through reactions with luminol or lucigenin , which are oxygenated by H2O2 ( and its derived species ) , or by superoxide anion , respectively [46] . As shown in Fig 5 , ArtinM-treated neutrophils , as well as untreated control cells , did not produce ROS since low and stable levels were demonstrated by both luminol- and lucigenin-chemiluminescence detection . In contrast , neutrophils stimulated with PMA or fMLP ( positive controls ) triggered high ROS production ( Fig 5A and 5B ) . Although ArtinM did not induce ROS production , it did not inhibit the production induced by other agents , a fact that was demonstrated by the observation that following the incubation with ArtinM , neutrophils responded to PMA- or fMLP-stimulus , with ROS levels similar to the ones measured in cells that were not pre-incubated with ArtinM ( Fig 5C and S3 Fig ) . The luminol-monitored ROS detection revealed that a higher amount ( 26-fold augmented ) was produced by infected ( 3 . 93x107±1 . 0x106 AUCxCPS ) compared to uninfected neutrophils ( 1 . 49x106±0 . 2x106 AUCxCPS ) . When the infected cells were pre-treated with ArtinM , the ROS production increased by 25% ( 5 . 3x107±1 . 1x106 AUCxCPS—Fig 5C ) . Taken together , in spite of not inducing uninfected cells to produce ROS , ArtinM enhances ROS production by L . major infected neutrophils , providing a mechanism that rapidly eliminates the parasite . The sustained neutrophil activation induced by ArtinM , which was detected even 20 h after treatment ( S4 Fig ) , as well as the absence of NET formation and ROS production , motivated us to evaluate the survival rate of ArtinM-treated neutrophils . Several methods were used to assess the occurrence of cell death , such as the analysis of neutrophil morphological changes , fragmentation of genomic DNA , phosphatidylserine ( PS ) exposure , and disruption of the mitochondrial transmembrane potential . The morphology of neutrophils was evaluated by optical microscopy . At 3h post-treatment , the microscopic features of ArtinM- or IL-8-treated cells , as well as of untreated cells , were typical of live neutrophils . At the same time point , cells treated with lysis buffer , as expected , showed remarkable nuclear condensation , as usually observed in apoptotic neutrophils [5 , 6] . After 20 h , nuclear condensation was verified in untreated cells , while the ArtinM- or IL-8-treated ones preserved their original morphology ( Fig 6A ) . Flow cytometry analysis of neutrophils that were incubated for 3 h with ArtinM ( 8 . 43±1 . 0% ) , or IL-8 ( 7 . 13±0 . 3% ) showed levels of PS exposure that were similar to the ones detected in untreated cells ( 6 . 98±0 . 8% ) and were significantly lower ( 4-fold ) than those detected in lysis buffer-treated cells ( 26 . 64±0 . 3%—Fig 6B ) . After 20 h , high exposure of PS was detected on untreated cells ( 58 . 66±2 . 7% ) whereas 20% and 34% lower levels were measured in ArtinM ( 45 . 89±3 . 9% ) and IL-8 ( 38 . 8±1 . 4% ) -treated cells , respectively . The analysis of neutrophils 20 h after treatment with lysis buffer was barred due to the scarce number of remaining cells ( Fig 6B ) . ROS production , mostly mitochondrial , is implicated in the initiation of cell apoptosis [47] . Since disruption of the mitochondrial trans-membrane potential ( ΔΨ ) is one of the earliest intracellular events occurring in apoptotic cells , we examined mitochondrial instability through the detection of JC-1 monomer . Fig 6C shows that the treatment of neutrophils with ArtinM ( 5091±276 . 5 MFI ) or IL-8 ( 4446±800 . 8 MFI ) favored mitochondrial stability , since at 3 h post-treatment , the levels of JC1 monomer were at least half of those detected in untreated ( 10020±105 . 4 MFI ) or lysis buffer-treated neutrophils ( 12186±1037 MFI ) . At 20 h , although JC1 monomer levels increased in ArtinM ( 9331±211 . 2 MFI ) or IL-8 ( 8066±635 MFI ) treated neutrophils , they remained significantly lower ( -15% ) than those detected in untreated neutrophils ( 11035±891 MFI ) . JC1 monomers were not detected in the rare neutrophils remaining 20 h after treatment with lysis buffer . The electrophoresis analysis of DNA showed that ArtinM-treated neutrophils displayed at least two-fold less fragmented DNA than untreated cells at 24 h of incubation . At 48 h of incubation , treated and untreated neutrophils displayed similar DNA fragmentation . ( Fig 6D and 6E ) . To summarize , these data demonstrate that ArtinM treatment of human neutrophils postponed apoptosis , as shown by delayed disruption of the mitochondrial trans-membrane potential , DNA fragmentation , nuclear condensation and PS exposure . Altogether , these results revealed that ArtinM treatment prolongs neutrophil survival . We found that ArtinM prolongs neutrophil survival , a fact that is considered to facilitate L . major infection [48] . Then , we analyzed the PS exposure on ArtinM-treated neutrophils that were infected with L . major . The ArtinM effect , exerted 3 hours after stimulation on the neutrophils PS exposure , at least doubled when the cells were infected with L . major , compared to uninfected cells ( 34 . 8±2 . 07% vs 11 . 97±0 . 74% , Fig 7 ) . In addition , these infected/ArtinM-treated neutrophils exhibited 40% higher PS exposure than the untreated/infected neutrophils ( 34 . 8±2 . 07% vs 21 . 41±3 . 92% , Fig 7 ) . Uninfected cells , treated or not with ArtinM , were used as control , at the same time point . The obtained results ( 11 . 97±0 . 74% and 11 . 55±0 . 61% ) were close to those shown in Fig 6B ( 3h ) . Therefore , ArtinM stimulus delays the death of non-infected neutrophils ( at 20 hours’ time point , Fig 6B ) , and accelerates the death of L . major-infected neutrophils ( Fig 7 ) , favoring the parasite elimination . Some assays performed by infecting neutrophils with fluorescent L . major forms allowed verifying that 91% of the AnnexinV labeled cells were infected . A close proportion ( 89% ) was verified among ArtinM-treated cells as well ( S5 Fig ) . We therefore concluded that apoptotic neutrophils were most frequently L . major infected cells , and that ArtinM did not change this distribution .
The immunomodulatory lectin ArtinM induces Th1 immune response and confers resistance to intracellular pathogens [21] , without causing apparent tissue damage . The response to ArtinM is induced by its interaction with glycoconjugates on several immune cells , namely , macrophages , dendritic cells , neutrophils , mast cells and lymphocytes [10 , 11 , 13 , 16 , 17 , 24–26 , 49] . Considering that neutrophils constitute a two-edged sword , accounting for resistance to pathogens and also for tissue injury , a full exploration of the neutrophil responses is required to envisage a possible application of ArtinM , or its analogues , as immunomodulatory agent . In this study , we verified that ArtinM enhances the neutrophils ability of eliminating the intracellular pathogen L . major . The parasite killing was associated with strong neutrophil activation , not accompanied by NET formation , which is known to cause tissue damage . Although macrophages are the definitive refuge for Leishmania species in the host , neutrophils are considered by many as transitional shelters for the few invader parasites that survive the toxic extracellular milieu [50] . This idea is based on the observation that in the early stages of L . major infection , caused by the bite of an infected sandfly , large amounts of neutrophils are attracted to the site . Since a low number of macrophages reside in the invaded tissue and the recruited neutrophils failed to kill L . major , the authors concluded that the invading parasites depended on the rapidly recruited neutrophils to survive [51] . Once inside the neutrophils , L . major promastigotes postpone neutrophil apoptosis until 2 days [48] and are silently transferred to macrophages , without activating the immune response [51] . In opposition to this “Trojan horse” mechanism of parasite evasion , ArtinM accelerates the death of L . major-infected neutrophils , favoring the process of parasite elimination . The production of the anti-inflammatory cytokine TGF-β favors the silent uptake of apoptotic cells by macrophages , whereas the production of pro-inflammatory mediators like TNF and IL-1β decreases after the uptake of apoptotic cells [52] . Consequently , the prominence of TGF-β correlates with permissibility to L . major infection , whereas high production of TNF is associated with resistance to infection [35] . Our observation that L . major-infected neutrophils largely augmented TGF-β production and diminished the secretion of pro-inflammatory mediators is consistent with previous demonstrations that neutrophils are important contributors toward providing the required microenvironment for parasite survival [34 , 53 , 54] . Notably , the treatment of neutrophils with ArtinM inverted this pattern by significantly diminishing TGF-β production and augmenting the secretion of pro-inflammatory cytokines . We found that ArtinM promoted IL-1β secretion by human neutrophils , which was maximum when the lectin was added to L . major infected neutrophils . This is a relevant finding because IL-1β maturation , as recently demonstrated , results from activation of the NLRP3 inflammasome , which is an innate platform that crucially restricts parasite replication . The NLRP3 inflammasome triggers inducible nitric oxide synthase ( NSO2 ) -mediated production of NO , a potent leishmanicidal factor [55] . In addition , NLRP3 inflammasome activation is associated with ROS production [56] , which is a major factor for Leishmania killing by human cells [57] . Indeed , macrophages derived from human monocytes do not produce NO after classical activation [58] or upon infection with Leishmania [59] . Thus , neutrophils , monocytes , and macrophages can control the parasites via ROS that are produced during the respiratory burst process [60] . By promoting ROS production and activating the NLRP3 inflammasome , ArtinM facilitates Leishmania killing , as shown in this work . Considering that the importance of IL-1β , derived from inflammasome , for conferring resistance to Leishmania infection was demonstrated for other species than L . major [55] , we plan performing further experiments with L . braziliensis-infected neutrophils , to better assess the IL-1β relevance for the ArtinM-induced protection against the parasite . Another trump against Leishmania infection provided by ArtinM treatment was the increased degranulation of human neutrophils . The intracellular levels of MPO were diminished in ArtinM-treated neutrophils , whereas the NE activity was augmented extracellularly . Both MPO and NE are stored in azurophilic granules , and their enzymatic activities are known to be implicated in the degradation of microorganisms in phagolysosomes . MPO is a key component of the oxidative burst , producing hypochlorous acid and other reactive oxidants [61] , whereas NE is a serine protease that degrades the outer membrane of microorganisms , inducing their elimination [39] . NE and MPO have low capacity to be exocytosed from the neutrophils azurophilic granules during infectious and inflammatory processes [36 , 62–64] . We found that treatment with ArtinM increased the exocytosis of these granules , even when the cells were infected with L . major . Many authors attribute an important role to neutrophil degranulation in mounting a response that culminates in pathogen elimination or control [36 , 65 , 66] . Regarding NE , recent data revealed that its release into the extracellular environment induces macrophage activation via TLR4 , which culminates in augmented internalization and elimination of L . major [67] . In the current work , we did not investigate the signaling through TLR4 , nor the relationship with macrophages , but we demonstrated that the ArtinM-induced NE release was followed by elimination of L . major by human neutrophils ( Figs 1 and 2 ) . The released NE accounts for the ArtinM leishmanicidal effect , once the NE presence in the extracellular space correlates with L . major elimination , as indicated by experiments using anti-NE neutralizing antibodies . Although NE is a serine protease that promotes microbe killing , our study apparently constitutes the first demonstration of its leishmanicidal activity , a finding that surely deserves further investigation . ArtinM-treated neutrophils , in sterile conditions , did not produce ROS , while other neutrophil activators such as fMLP and PMA induced rapid and intense responses . The L . major infection of neutrophils increased ROS production , which was even higher in ArtinM-treated neutrophils . It is well established that ROS provides an important mechanism to combat L . major [68] , but is also able to cause tissue injury . The ROS production induced by ArtinM was restricted to infected cells whose life span was shortened by ArtinM treatment , allowing us to postulate that the ability of ROS to cause tissue damage could be reduced . The effect of ROS on parasite elimination could be preserved , since it is effective in the early stages of infection [69] . ROS production accounts for NET formation , as demonstrated by the observation that neutrophils treated with NADPH oxidase , a pharmacological inhibitor of ROS production , do not form NET in response to conventional stimuli [70] . We showed that ArtinM-stimulated neutrophils , either uninfected or infected with L . major , did not form NET . This means that ArtinM inhibits the NET formation induced by the infection itself [44] and additionaly by PMA The inhibition occurred in spite of NE release and ROS production , which favor NET formation and were verified to occur in ArtinM-treated neutrophils . NET formation constitutes a mechanism whereby neutrophils eliminate pathogens , including L . amazonensis promastigotes [71] . Other Leishmania spp , namely , L . donovani , L . major , L . infantum , and L . mexicana , escape from their toxicity despite NET formation and trapping by these web-like structures [44 , 72 , 73] . The fact that ArtinM , although inhibiting NET formation , facilitates L . major clearance by human neutrophils reinforces the idea that NET formation is not required for L . major elimination . On the other hand , the absence of NETs can favor host tissue integrity , since NET-associated proteases and granular proteins have been shown to damage host tissue [74] . We conclude that ArtinM treatment of human neutrophils enhances the clearance of the intracellular pathogen L . major , through mechanisms that include: ( 1 ) production of inflammatory cytokines , i . e . , high TNF and IL-1β , and virtual absence of TGF-β; ( 2 ) increased neutrophil degranulation; the released elastase promotes L . major elimination; ( 3 ) increased ROS production by infected neutrophils ( 4 ) shortened neutrophil survival . On the other hand , the host tissue integrity is favored by ( 1 ) the short period of ROS production and ( 2 ) the absence of NET formation . Fig 8 delineates our model of ArtinM effects on neutrophils during L . major infection , based on our results and assumptions . | Vaccination is a successful way to eliminate infectious diseases . The generated antibodies neutralize the invading microbe and avoid the establishment of infection . Vaccines are efficient to prevent infections by pathogens living outside rather than inside the host`s cells . This occurs because protection against intracellular pathogens requires the engagement of T lymphocytes . The discovery of receptors on innate immunity cells opened new perspectives in trying manners to stimulate effective response against intracellular pathogens . Frequently the microbial sensing by Toll-like receptors ( TLRs ) besides triggering immediate defense also orchestrates adaptative immunity towards T-cell response . Therefore , TLR ligands started to be assayed in new anti-infective approaches . Our laboratory has been investigating the immunomodulation induced by lectins , which are ubiquitous sugar-binding proteins . Our primary model is ArtinM , from the seeds of jackfruit , a lectin that binds to TLR2 sugar chains on macrophages and dendritic cells and promotes production of cytokines that engages T lymphocytes in a process that culminate with elimination of intracellular pathogens . Concomitantly , ArtinM activates other immune cells , including neutrophils , which contributes to the pathogen elimination , but may also account for tissue damage . This last possibility led us to investigate the lectin effects on neutrophils deeply . We analyzed neutrophils treated with ArtinM and infected with Leishmania major . We concluded that the leishmanicidal ability of ArtinM-stimulated neutrophils was due to augmented release of inflammatory cytokines , ROS production , and cell degranulation . Otherwise , host tissue integrity is favored by shortened cells lifespan and absence of NET formation . This work reinforces the idea that ArtinM can be an appropriate molecular template for the construction of an efficient anti-infective agent . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [
"blood",
"cells",
"cell",
"physiology",
"cell",
"death",
"innate",
"immune",
"system",
"medicine",
"and",
"health",
"sciences",
"immune",
"cells",
"immune",
"physiology",
"cytokines",
"pathology",
"and",
"laboratory",
"medicine",
"intracellular",
"pathogens",
"pathoge... | 2016 | Neutrophils Contribute to the Protection Conferred by ArtinM against Intracellular Pathogens: A Study on Leishmania major |
Antibodies binding to the surface of virions can lead to virus neutralisation . Different theories have been proposed to determine the number of antibodies that must bind to a virion for neutralisation . Early models are based on chemical binding kinetics . Applying these models lead to very low estimates of the number of antibodies needed for neutralisation . In contrast , according to the more conceptual approach of stoichiometries in virology a much higher number of antibodies is required for virus neutralisation by antibodies . Here , we combine chemical binding kinetics with ( virological ) stoichiometries to better explain virus neutralisation by antibody binding . This framework is in agreement with published data on the neutralisation of the human immunodeficiency virus . Knowing antibody reaction constants , our model allows us to estimate stoichiometrical parameters from kinetic neutralisation curves . In addition , we can identify important parameters that will make further analysis of kinetic neutralisation curves more valuable in the context of estimating stoichiometries . Our model gives a more subtle explanation of kinetic neutralisation curves in terms of single-hit and multi-hit kinetics .
Antibodies are the most efficient way the immune system fights viruses before they infect host cells . Most of the available vaccines against viral pathogens stimulate the immune system to produce antibodies against a variety of molecular patterns on the viral surface , the epitopes . Each antibody response consists of many different antibodies directed against different epitopes or directed against the same epitope but varying in their binding strengths . During the first three quarters of the last century , these antibody mixtures were tested for their neutralising potential . However , rational vaccine design requires knowledge of which specific antibodies have the highest neutralising potential . A vaccine should then stimulate the immune system to produce these antibodies . To study antibody binding and to characterise antibodies , different methods have been proposed . In the early days of virology , the number of antibodies required to neutralise a virion was studied for many different viruses , using the concepts of chemical binding kinetics . Antibodies were added to virion populations and at intervals of one or more minutes , samples were taken and immediately diluted . This stopped antibody binding and the surviving virions were counted in plaque assays [1] . The theory employed for interpreting the kinetic neutralisation curves is based on early work on western equine encephalitis virus and poliomyelitis virus [2] . The basic assumption in these models is that there is at least one critical binding site on the virion surface . The virion is neutralised as soon as one of these binding sites is bound to an antibody . The shape of the time-neutralisation curves was thought to carry information on the number of antibodies needed for virion neutralisation . A sudden decline in the time-neutralisation curve ( no lag-phase ) , was interpreted as a single-hit mechanism , i . e . that the binding of one antibody is sufficient to neutralise the virion [2] . In contrast , a lag phase at the start of the time-neutralisation curve was interprereted as a multi-hit mechanism . Experimentally observed time-neutralisation curves decline to a certain level , further neutralisation does not seem possible . In the early framework , this leveling-out was interpreted as a persistent virus fraction that cannot be neutralised . According to this framework , the experimentally obtained plots were often interpreted as a proof for a single-hit mechanism [2] . By varying the experimental conditions , however , some evidence for multi-hit mechanisms arose ( reviewed for example in [3] , [4] ) . McLain and Dimmock [5] used these methods to study the neutralisation of HIV and suggested that three antibodies can neutralise a single HIV-virion . Klasse and Sattentau [6] reviewed these low numbers critically and introduced the differentiation between binding kinetics and occupancy , i . e . the number of antibodies attached to a virion . They show that the minimum occupancy required for viral neutralisation only influences the slope of the binding kinetic curves and not necessarily the shape . Other evidence against a single-hit mechanism of neutralisation comes from imaging HIV-virions while they infect a cell [7] . It seems as if more than one HIV-spike interact with target cell receptors . An HIV-spike consists of three heterodimers ( envelope protein , Env ) , each comprising one gp120 that is loosely attached to one surface embedded gp41 [8]–[10] . The spikes are also referred to as trimers . These trimers establish contact with target cell receptors and mediate infection of the cell [11]–[13] . This makes the trimers the perfect targets for neutralising antibodies [14] , [15] . However , the contact region between the target cell and the virion seems to be relatively small in comparison to the virion's surface . One antibody could easily bind to a trimer that is not engaged in the contact region and this binding would not prevent the attachment process . Therefore , the concept of stoichiometries was introduced into virology . The question of how many antibodies must bind to a whole virion for neutralisation was broken down into studying the number of interactions of spikes and cellular receptors required for viral entry ( stoichiometry of entry ) and how many antibodies must bind to a single spike such that it loses functionality ( stoichiometry of ( trimer ) neutralisation ) [16]–[21] . One can then calculate the number of antibodies that have to bind to a single virion for neutralisation , including random binding effects [22] . Note that the term stoichiometry is not as strictly used in virology as in chemistry . In chemistry the term describes the quantitative relationship between reactants and products . By contrast , in virology , the term stoichiometry describes how many molecules are involved in certain processes . The interpretation of neutralisation kinetic curves as single-hit neutralisation and the concepts of stoichiometry , in which many more antibodies have to bind for neutralisation , seem to contradict each other . Binding kinetics describe the change of compounds during a chemical reaction . This concept was transferred to describe antibody binding mechanisms for neutralisation kinetics . For chemical binding curves the concentration of the reactants are measured over time . In neutralisation kinetics one does not measure the concentration of antibody-virion complexes but the percentage of neutralised virions . This means that a second reaction is needed to predict neutralisation kinetics . Direct observation of the products is therefore not possible , making it error-prone to conclude single-hit kinetics out of neutralisation kinetics curves . The concept of occupancy and stoichiometry assumes that there is a minimal number of viral spikes that have to engage with cellular receptors to mediate entry . Further it is assumed that if antibodies knock out a sufficiently high number of spikes the virion is unable to infect any cell . This framework does not incorporate the antibody binding process and only informs about the numbers of antibodies that have to bind to a virion population for neutralisation . It cannot inform about the antibody concentrations that are needed to reduce the infectivity of a virion population . Here we derive a mathematical framework combining the concepts of chemical binding kinetics and stoichiometries to study neutralisation kinetics . This framework makes it possible to predict neutralisation curves as obtained in earlier studies . Important parameters have not been measured so far , but we show how these parameters influence the predicted neutralisation curves .
Whilst the formulation in the model section is generic in the sense that it is applicable to all enveloped viruses , the model is mainly inspired by the neutralisation of HIV by monoclonal IgG antibodies . In this system , it is unlikely ( i ) that an IgG antibody binds to two epitopes of the same spike , due to geometrical reasons , and ( ii ) that an antibody binds to two epitopes of neighbouring spikes , due to the low spike density [23] , [24] . We therefore assume that one antibody binds with one of its two Fab-regions . Hence , we can describe binding of one antibody , , to one epitope of a spike , , as a chemical reactionin which is the antibody binding constant and its dissociation constant of the first antibody binding . A spike with epitopes specific for one antibody type can bind to antibodies according to the chemical binding reaction ( 1 ) with the binding constant and dissociation constant for forming and dissociation of the complex , respectively . Employing the concepts of chemical kinetics , this reaction equation can be translated into the following set of differential equations [25] . Each equation describes the time evolution of the concentration of one of the components over time . Note that concentrations are indicated by square brackets . ( 2 ) In these equations , the exponent is the order of the dissociation reaction of the complex , the exponent is the order of the th reaction step with respect to the complex and the exponent is the order of the th reaction step with respect to the antibody , for . Note that we use the notation for and for , respectively . Each virion expresses a certain number of spikes on its surface . Equation 2 describes the binding to spikes as if they were in solution . However , the spikes are attached to viral surfaces . Zhu et al . [26] visualised 40 HIV-1 virions with cryo-electron microscopy and found that the number of spikes per surfaces varies from virion to virion with a mean expression of spikes per virion . Variation in spike numbers in other viruses might also occur . We therefore define the spike number distribution with as the fraction of virions with spikes , where ranges from 0 spikes to the maximal spike number . For each time step we calculate the concentration of spikes bound to antibodies according to equation 2 . In the second step , the spikes are re-distributed to the virions . The fraction of spikes bound to antibodies at time , arises from the concentrations of spikes bound to antibodies: ( 3 ) for . The concentrations are determined by ( numerically ) solving the system of ODEs in equation 2 . For the sake of simplicity , we write instead of wherever possible . For given fractions of spikes bound to antibodies , , the probability that a virion with spikes has exactly spikes bound to antibodies for follows a multinomial distribution and is ( 4 ) A virion with fewer than spikes is never infectious according to the definition of the stoichiometry of entry . If the spike number is at least , the virion is infectious if it has spikes with fewer than bound antibodies . Therefore , the total number of spikes with bound antibodies must sum up to with for the virion to be infectious . As an example , let us consider the virion sketched in Figure 1 ( A ) . It has spikes each consisting of subunits ( which again is inspired by the structure of an HIV virion ) . This virion has spikes bound to 0 antibodies , spikes bound to 1 antibody , and spikes bound to 2 and 3 antibodies , respectively . Let us assume that at time point the concentration of spikes bound to one antibody , , equals , and the concentrations , and the concentration of unbound spikes is . The fraction of spikes bound to antibodies is then and . The probability that a virion with spikes has unbound spikes , and five , one , three spikes bound to one , two , three antibodies , respectively , given these concentration is 0 . 0117 = 1 . 17% . To calculate the probability that a virion with spikes is infectious we have to calculate the probability that a virion with spikes has or more spikes with fewer than antibodies . Thus the probability that a virion with spikes is infectious is ( 5 ) In the second sum we sum over all possibilities that the number of spikes with fewer than bound antibodies equals . This sum can be re-written by going through all combinations of these numbers and reads then . The percent infectivity of a viral stock at time , is the number of infectious virions at time , , divided by the number of infectious virions at time , or without any bound antibody , . This quantity can be experimentally measured by plaques assays [2] , [3] or in infectivity assays with pseudotyped virions [16] . To calculate , we weigh this probability ( equation 5 ) with the probability that a virion has spikes , . In addition , we have to divide by the probability that a virion has at least spikes , because the infectivity of a viral stock obtained with infectivity assays is always normalised with the infectivity of a viral stock without any antibodies . Thus we obtain: ( 6 ) where and as defined in equation 3 . A remark about the units of the reaction constants: As concentration is measured in , the product of reaction constants and product concentrations must have the unit for every summand on the right hand side of the equations in Equation 2 . The reaction kinetic equations are generic in the sense that they allow for any possible reaction order in any step with respect to any product . Thus the units of the reaction constants are where are the reaction orders in respect to the product and , respectively . For simplicity , we omit the units in the following . A summary of the parameters used in the models can be found in Table 1 . All calculations are implemented in the R language for statistical computing [27] and are available in Dataset S1 .
HIV virions express trimers of the heterodimeric envelope proteins ( Envs ) gp120 and gp41 embedded on their surface [8]–[10] . As a monoclonal antibody binds to a well-defined region only present once per envelope subunit [14] , up to three antibodies can bind to a whole spike , thus . According to [24] the average distance between two spikes is bigger than the distance between two Fab-regions of one antibody , which lies in the range of 15 nm . In addition , the average distance between two epitopes must be smaller than the diameter of a trimer , which is 10 . 5 nm . Therefore intra-spike and inter-spike binding of two Fab regions of the same antibody is unlikely in the case of HIV . Hence we assume that each antibody binds to one epitope . The additional binding of an antibody to a trimer that is already bound to one or two antibodies might be hindered , e . g . by sterical hindrance . This is reflected in the model by differences in the binding and dissociation constants . Zhu et al . [26] counted the spike numbers of 40 HIV-1 virions and found a mean of 14 spikes with a variance of 49 . However , the sample size is too small to take the fraction of virions with spikes as a measure for the real fractions . In an earlier publication [20] , we therefore defined a discretised Beta-distribution with mean 14 and variance 49 and we will use this distribution for the HIV-specific model . This distribution was chosen because the Beta-distribution is defined on a closed set and has a high flexibility depending on two parameters which can be expressed in terms of mean and variance . The form of the distribution ranges from peaks at the edges to one peak in the centre . As we do not have an experimentally determined trimer distribution but only mean and variance , a discretised version of the beta distribution might come as close to the real distribution as possible . McLain and Dimmock [5] studied the kinetics of three monoclonal rat antibodies against HIV-1 IIIB . We extracted the kinetic neutralisation data to which we fitted our model of the percent infectivity ( equation 6 where the fraction of spikes bound to antibodies at time , , is calculated according to equations 2 and 3 ) . To this end , we allowed the stoichiometry of entry to be or 19 as found for different model assumptions in [20] . We further assumed the distribution of trimer numbers to follow the discretised B-distribution defined previously [20] . McLain and Dimmock [5] did not measure the virus concentration directly . The antibody concentrations are only shown in . As we do not know the exact molar weight of these antibodies we cannot calculate the antibody concentration . In addition , it is not possible to reconstruct the exact virus ( and hence the exact spike ) concentration out of the syncytium-formation assay . Therefore , we used the starting concentrations and as above . The binding and dissociation rates are also not known . Therefore , we performed a non-linear regression on the data by simultaneously estimating the binding and dissociation constants and the stoichiometry of entry and neutralisation . Finally , we applied this fitting routine to the elementary reaction model and the stoichiometric reaction model . The stoichiometric reaction model fits McLain and Dimmock's data [5] significantly better than the elementary reaction model . This is in accordance with the finding that the elementary reaction model does not reflect experimentally observed kinetic neutralisation curves and supports the conclusion that antibody binding reactions are not elementary reactions . Figure 5 shows the best fits for the three different antibodies employing the stoichiometric reaction model . The stoichiometry of entry , , equals 2 for all three antibodies . The stoichiometry of trimer neutralisation is 1 for ICR39 . 3b and ICR39 . 13g and for ICR41 . 1i . The reaction constants are shown in Table 2 . In parameter estimations as performed here , confidence intervals would be determined by a bootstrap routine . As there are only eight data points per antibody and eight variables to estimate , this method cannot be used to derive confidence intervals for the parameters . In addition , the statistical power of these estimates is not strong . However , our estimation procedure shows that the model framework can be applied to kinetic neutralisation data and can be used to estimate stoichiometric parameters . Applying the theoretical framework presented in [22] , the average number of antibodies having to bind to one average virion is then 23 for ICR39 . 3b and ICR39 . 13g and 41 for ICR41 . 1i . Other viruses may have other strategies including different receptors or pathways like endocytosis ( for an overview over different entry mechanisms see e . g . [29] ) . Even though the mechanisms for entry differ substantially , all viruses have to attach to cellular receptors via viral spikes as a first step in infection . These spikes are excellent targets for neutralising antibodies . For example , influenza type A virus is estimated to express spikes on its surface [30] . Hemagglutinin , the spike responsible for viral entry [31] , is also a trimeric protein , i . e . [32] but not all of the 450 spikes expressed on the virion's surface are hemagglutinin proteins . Hepatitis C Virus is a small ( diameter 40–60 nm ) , enveloped virus . The viral spike that plays a major role in viral entry consists of two envelope proteins , E1 and E2 , forming heterodimers [33] , [34] . Our model ( equations 2 and 6 ) is formulated with enough flexibility that we can account for variation in trimer number distribution and variation in binding sites within a trimer . However , we only test the effect of variation in the trimer number distribution here . In Figure 6 we show the kinetic neutralisation curves for different viral populations . Curves in red are based on virions with a mean trimer number distribution of 10 , black 14 and blue 36 . The higher the trimer number is , the slower neutralisation happens . This means the more spikes a virion expresses , the more antibodies must bind for neutralisation . The dashed red line and the dashed blue line are based on virions with exactly 10 and 36 spikes , respectively . The dotted red line is based on spike numbers varying from 2 to 18 and the dotted blue line 0–72 spikes . Comparing the dotted and the dashed lines , one sees that variation in spike numbers has an effect on the kinetic neutralisation curves . However , more variation in spike numbers does not necessarily means slower neutralisation .
In this paper we derive a model for antibody neutralisation that combines binding kinetics with stoichiometries . Antibodies bind to the viral surface spikes according to a simple chemical multi-step reaction . Whether a virion is still infective is defined via the concept of stoichiometries: at least spikes must be bound to fewer than antibodies each . With this framework it is possible to predict published observations of kinetic neutralisation curves . In the past , the interpretation of kinetic neutralisation curves was based on a theoretical framework derived in [2] . A straight decline in the time-log ( percent infectivity ) curve was interpreted as a single-hit mechanism , i . e . one antibody is sufficient to neutralise a whole virion . The main assumption of this theory was that there are critical and non-critical binding sites on the virion . As soon as an antibody binds to one critical site , the whole virion was assumed to be neutralised . However , in our model there is no need to subdivide the binding sites into critical and non-critical . The curvature is determined by the binding and dissociation constants . The former concept interpreted the levelling out as a persistent virus fraction . In our model , levelling out is due to antibody binding and dissociation kinetics . The model for the percent infectivity ( equation 6 ) depends on many parameters but we are mostly interested to estimate the stoichiometry of entry and neutralisation . Unfortunately , all other parameters can have a big influence on the predictions of the kinetic neutralisation curves . The reaction orders of antibody binding in the several binding steps are not known . We tested two simple scenarios here . As the binding of an antibody to an epitope involves huge binding sites and not only single atoms , the reactions might not follow simple elementary reaction principles and should be studied in more detail to confirm or neglect our assumed scenario . Besides the reaction orders , the most influential parameters are the binding and dissociation constants . They not only shape the curve at the beginning of the reaction but they also have a huge impact on whether it is possible to estimate the stoichiometry of trimer neutralisation from kinetic neutralisation curves . Therefore , we recommend to study these parameters in a different experimental setup . The starting concentration of antibodies and spikes can easily be measured at the beginning of the experiment . The spike concentration is equal to the number of virions times the average number of spikes per virion divided by the volume of the solution tested . The number of virions can be measured by quantitative real time polymerase chain reaction and the average number of spikes per virion by counting spikes on cryo-electron microscopical pictures of virions [26] . The estimation of the reaction constants and the stoichiometry of entry and neutralisation must be seen more as a proof of our method than a reliable estimate of stoichiometric parameters . An additional confounding factor might be non-functional spikes that are still able to bind antibodies . This would result in lower net antibody concentrations . If there are many non-functional spikes , this effect might have a non-negligible influence on the prediction of the percent infectivity with our model . We therefore recommend to quantify the amount of non-functional spikes . As soon as more data becomes available , our framework can be used to estimate stoichiometrical parameters . In our study , we reanalysed kinetic neutralisation curves for three rat monoclonal antibodies against HIV-1 IIIB [5] . The virions of the human immunodeficiency virus express low trimer numbers ( mean [7] ) and the spikes responsible for entry consist of three identical gp120/gp41 subunits . We therefore assume that each monoclonal antibody binds with only one Fab-region and up to three antibodies can bind to one spike . Under these assumptions we also checked the influence of the trimer number distribution on the kinetic neutralisation curves . The higher the mean trimer number is , the slower virions are neutralised . Other viruses can differ in their route of entry , but entry always involves attachment to cellular receptors [29] . Antibodies binding to viral spikes can therefore at least theoretically confer neutralisation . This means that our framework can also explain neutralisation of other viruses . However , if the spike density exceeds a certain threshold , antibodies can bind with their second Fab region and the concept of avidity comes into play . The binding strength that exists between one Fab region and one epitope is called affinity . If the second Fab region of an IgG antibody binds to another epitope the binding strength between the antibody and the pathogen increases more than the twofold binding strength between one Fab region and one epitope . This enhanced binding strength is called avidity . For simplicity , we did not account for avidity in the model presented here . We explained why this can be done in the case of HIV earlier . For other viruses with a higher spike density , however , avidity may play an important role . In this case , our model must include more complex structures of antibody spike complexes in the form of . The number of binding and dissociation constants in this case will be increasing tremendously . In our model , antibodies can bind and fall off any epitope . However , there are some HIV-antibodies that lead to irreversible destruction of the trimer [35] . Studying these antibodies with our model framework requires setting the dissociation rate to 0 . If the destruction of the trimers happens when fewer antibodies bind to a spike than binding sites on the spike all dissociation rates of spike-antibody complexes with more than this threshold number must be set to 0 . In this study we focus on the analysis of kinetic neutralisation curves . Virologists normally characterise antibodies according to the concentration at which 50% of the neutralising effect is reached , the IC50 . To this end , the neutralisation potential of antibody solutions of different concentrations are tested . We have shown , that the prediction of the kinetic neutralisation curves depend on the startimg concentration of antibodies . When defining a time point at which the percent infectivity should be measured , we can also adapt our model to predict titration curves . How well these predictions can be used for estimating stoichiometrical parameters is the subject of future studies . So far , our results focus on in vitro systems with monoclonal antibodies . In vivo systems are far more complicated . The immune system elicits a huge variety of different antibodies with different reaction constants and different concentrations . In the future it will be necessary to study how different antibodies interact with each other , e . g . do they synergise or antagonise ? It may also be possible that the binding of one antibody leads to conformational changes within the trimer leading to revelation of another epitope that is targeted by a more potent antibody . These mixtures of antibodies will require more elaborate models than the framework presented here . Similar to the concept of in epidemiology it might not be necessary to neutralise every single virion but reduce the amount of non-neutralised virions such that on average , each virion produces less than one offspring [36] . This might be already reachable with antibody concentrations that do not confer 100%neutralisation . However , whether a vaccine-induced antibody response or passive immunisation with antibodies lead to full neutralisation of all virus particles in vivo also depends on the concentration of virions and antibodies across different body compartments , such as blood or mucosal surfaces . The virion concentration as well as the antibody concentration could vary substantially from compartment to compartment and the antibody concentration might not be sufficient for neutralisation in some of them . To date , the presented framework still needs conformation by experimentalists . As pointed out above , the most important experiment to be done is the determination of the reaction constants . Once these are available , our framework can be used to infer stoichiometries . With the help of stoichiometries it is possible to determine the numbers of antibodies needed for neutralisation in vitro . If the antibodies behave similarly in vivo , our models make it possible to compare different antibodies on a rational basis as soon as the stoichiometrical values will have been determined for different antibodies . By extending our framework , it might be possible to also study interactions between different antibodies more rationally which will complete the picture of antibody based neutralisation . | How many antibodies have to bind to a virus particle such that it is prevented from infecting a cell ? This seemingly simple question has not been answered yet . However , this number is crucial to determine whether a vaccine can stimulate the immune system to elicit enough antibodies to neutralise virus before starting an infection . Two different approaches have been applied to answer this question , leading to contradictory results . One approach is inspired by concepts from binding kinetics , the other approach is a more conceptual one . Here , I describe the advantages and disadvantages of either approaches and condense the advantages of both into one model framework . I show under which conditions the framework can be used to identify the number of neutralising antibodies . In addition , this model can explain why viruses might not completely loose their infection potential even when there is a huge excess of antibodies . | [
"Abstract",
"Introduction",
"Models",
"Results",
"Discussion"
] | [
"medicine",
"infectious",
"diseases",
"theoretical",
"biology",
"virology",
"hiv",
"biology",
"computational",
"biology",
"microbiology",
"viral",
"diseases"
] | 2013 | Virus Neutralisation: New Insights from Kinetic Neutralisation Curves |
M . africanum West African 2 constitutes an ancient lineage of the M . tuberculosis complex that commonly causes human tuberculosis in West Africa and has an attenuated phenotype relative to M . tuberculosis . In search of candidate genes underlying these differences , the genome of M . africanum West African 2 was sequenced using classical capillary sequencing techniques . Our findings reveal a unique sequence , RD900 , that was independently lost during the evolution of two important lineages within the complex: the “modern” M . tuberculosis group and the lineage leading to M . bovis . Closely related to M . bovis and other animal strains within the M . tuberculosis complex , M . africanum West African 2 shares an abundance of pseudogenes with M . bovis but also with M . africanum West African clade 1 . Comparison with other strains of the M . tuberculosis complex revealed pseudogenes events in all the known lineages pointing toward ongoing genome erosion likely due to increased genetic drift and relaxed selection linked to serial transmission-bottlenecks and an intracellular lifestyle . The genomic differences identified between M . africanum West African 2 and the other strains of the Mycobacterium tuberculosis complex may explain its attenuated phenotype , and pave the way for targeted experiments to elucidate the phenotypic characteristic of M . africanum . Moreover , availability of the whole genome data allows for verification of conservation of targets used for the next generation of diagnostics and vaccines , in order to ensure similar efficacy in West Africa .
Mycobacterium africanum causes up to half of human TB in West Africa and displays differences in patient characteristics and immunoepidemiological features with M . tuberculosis , as reviewed earlier in this journal [1] . First described in 1968 in Dakar , Senegal [2] , M . africanum used to be classified using biochemical methods , until unambiguous classification became possible using molecular methods and two different lineages were identified , M . africanum West African type 1 , common to Eastern West Africa , and M . africanum West African type 2 , common to Western West Africa [3] . Additionally , it became clear that the former “East African M . africanum” is genetically part of M . tuberculosis sensu stricto [4] . The prevalence of M . africanum varies within West Africa , with the highest prevalence of M . africanum West African 2 identified in Guinea Bissau ( 51% , [5] ) and the highest prevalence of – West African 1 recorded in Benin ( around 28% , [6] ) . While comparisons between prior prevalence estimates based on biochemical speciation and current estimates based on molecular speciation deserve caution , the prevalence of M . africanum appears to be decreasing in Cameroon [7] and Senegal ( unpublished results ) . Comparisons between patients infected with M . africanum West African 2 and M . tuberculosis suggest that M . africanum is somewhat attenuated in its ability to cause disease in immunocompetent hosts [8] and is more common in HIV co-infected patients relative to M . tuberculosis in The Gambia [9] , yet not in Ghana [10] . Moreover , patients infected with M . africanum West African 2 , as well as their household contacts , are less likely to mount an IFNg response to ESAT-6 than those infected with M . tuberculosis [11] . These two types of M . africanum , West African 1 and West African 2 , are distinct sub-species within the M . tuberculosis complex although it has been suggested that these clades are better described as ecotypes of the M . tuberculosis complex rather than sub-species [12] . M . africanum West African type 2 is phylogenetically closer to the animal strains like M . bovis , with which it shares deletions RD7 , 8 and 10 [13] , [14] , [15] , although an animal reservoir for M . africanum West African type 2 has not been detected [16] . Subtractive hybridization of M . africanum West African type 1 and type 2 from M . tuberculosis H37Rv revealed shared and unique genomic differences [3] , yet these experiments were not designed to identify regions present in M . africanum but absent from M . tuberculosis . Here we take advantage of available genomic information for strains of the Mycobacterium tuberculosis complex from different sequencing platforms to present and analyze the first complete M . africanum West African type 2 genome , that of clinical isolate GM041182 ( here designated M . africanum GM041182 in the remainder of this manuscript ) , detailing a novel lineage-defining deletion and an array of species-specific pseudogenes .
Mycobacterium africanum GM041182 was isolated in The Gambia in 2004 from a 27 year old HIV uninfected male patient with 3+ smear positive pulmonary tuberculosis . This patient provided written informed consent for participation in the TB Case Contact cohort study , which had been approved by the joint Gambia Government/MRC ethics committee . Moreover , the same ethics committee approved genotyping of bacteria isolated from the patients enrolled in this cohort , and the data were analyzed anonymously . Primary isolation was done in an automated liquid culture system ( Bactec MGIT 960 , BD ) and drug susceptibility testing for first line drugs on solid medium identified no resistance . Genomic DNA was extracted from a single colony sub-culture using the CTAB method [17] and genotyped using spoligotype analysis [18] and PCR for Large Sequence Polymorphism RD702 [3] . The genome of Mycobacterium africanum GM041182 was sequenced to approximately 10-fold shotgun coverage , comprising 92612 end sequences , from pOTW12 ( with insert sizes 3–4 kb ) and pMAQ1Sac_BstXI ( with insert sizes of 4–5 kb and 5–6 kb ) genomic shotgun libraries using big-dye terminator chemistry on ABI3730 automated sequencers . End sequences from large insert Fosmid libraries in pCC1Fos with an average insert size of 38–42 kb provided scaffold information with approximately 0 . 2-fold coverage from 2077 end sequences . A 454 FLX sequencing run provided approximately 10-fold single-end shotgun coverage , comprising 224 , 378 end sequences from 3kb DNA fragments . In addition an Illumina GAII sequencing lane provided approximately 50-fold single-end shotgun sequence , comprising 6083237 end sequences from 200–300 bp fragments and 37 cycles of sequencing . All repeat regions and gaps were bridged by read-pairs or end-sequenced polymerase chain reaction ( PCR ) products again sequenced with big dye terminator chemistry on ABI3700 capillary sequencers . The sequence was manipulated to the ‘Finished’ standard [19] and is deposited in EMBL/Genbank/DDBJ under accession number FR878060 . Coding sequences were initially identified by using Glimmer3 [20] and then manually curated using Frameplot [21] and Artemis [22] . All genes were annotated in Artemis using standard criteria [23] . Genome comparisons were visualized in the Artemis comparison tool [24] . Sequence clustering and analysis was performed by using ClustalX 2 . 0 [25] and MEGA4 [26] . To corroborate the phylogenetic position of the GM041182 isolate within the MTBC we took advantage of the availability of Illumina GAIIx runs for different clinical strains representative of the MTBC [27] . We mapped reads for each strain to the genome of GM041182 using MAQ [28] and single nucleotide polymorphisms were called as described in Comas et al . 2010 [27] . A total of 9 , 699 positions were identified to vary in at least one strain after exclusion of positions with heterozygous calls or deletions ( no coverage positions ) . A phylogeny was inferred using the number of nucleotide differences between strains as the distance measure and Neighbour-joining as the reconstruction method , and 1 , 000 bootstrap pseudo-replicates were performed to assess the reliability of the clades . Alternative molecular evolution models and phylogenetic methods were not carried out , as a similar set of strains was extensively analyzed before and no difference in topology was observed between the different approaches [27] . All the phylogenetic analyses were carried out using MEGA5 package [29] . A two step process was carried out to identify mutations that either led to the pseudogenization of previously described genes or generated new potential CDS in lineages of the MTBC . Because the genomes of M . tuberculosis H37Rv and M . bovis AF2122/97 were completed by shotgun sequencing and their annotation manually curated we used them to infer a first list of candidate pseudogenes when compared to M . africanum GM041182 . CDS were designated as pseudogenes if they contain in the alignment of homologous positions between the three strains either a frameshift or nonsense mutation , were truncated by a deletion event , or interrupted by a large insertion event . As a second step we focus on the microevolution of the MTBC by assessing whether the events leading to truncated or novel CDS were shared among strains of the different lineages of the complex . We took advantage of the availability of draft shotgun sequences of strains belonging to the different lineages ( see Supplementary Table S1 for a list of strains and sources ) . The polymorphisms were corroborated in other strains by blast searches and manual inspection of the alignments . To assign evolutionary directionality to the changes we used as an outgroup the M . canettii genome ( accession number HE572590 ) .
General features of the M . africanum GM041182 genome are unremarkable relative to other members of the M . tuberculosis complex with a typical %G+C content ( 65 . 6% ) and a genome size ( 4 , 389 , 314 bp ) between the usual values for M . bovis ( 4 . 34–4 . 37 Mbp ) and M . tuberculosis ( 4 . 40–4 . 42 Mbp ) . The M . africanum GM041182 genome is also collinear with those of M . bovis and M . tuberculosis and shares the majority of coding sequences ( CDSs ) . Identification of CDSs present in M . bovis and M . tuberculosis but absent from strains of M . africanum has been presented in several publications to date so will not be further detailed here [3] , [30] , [31] . However , the availability of the M . africanum GM041182 genome sequence has enabled the search for M . africanum-specific sequences and the identification of M . africanum-specific pseudogenes . We took advantage of the publicly available Illumina sequencing data for 23 strains representative of the MTBC including the sequences of two lab-adapted strains , M . tuberculosis H37Rv and M . bovis Ravenel , as well as the sequence of a strain classified as M . canetti which we used as an outgroup . We mapped the Illumina short-reads to the newly generated M . africanum GM041182 and called for high-confidence polymorphisms . After exclusion of those SNP calls falling in PE/PPE genes and in phage-related regions of the genomes we used an alignment of 9 , 699 ‘core’ SNP calls ( positions in the genome of M . africanum GM041182 where at least one strain has a SNP and no strain has a putative deletion or heterozygous call ) . The resulting phylogeny ( Figure 1 ) placed M . africanum GM041182 as part of the M . africanum West-African 2 clade ( also known as Lineage 6 ) , clustered closely to another strain originally isolated in The Gambia ( GM0981 ) . The phylogeny also reflects the great diversity of human M . tuberculosis complex strains found in West African countries with circulating strains from at least three different lineages; the two M . africanum clades and different sub-lineages belonging to the Euro-American lineage ( Lineage 4 ) which are thought to be recently re-introduced in Africa , as the Euro-American lineage is supposed to have originated in the European region [14] . It has been proposed that due to historical migrations and the low-infectious dose during aerosol transmission of human tuberculosis the effective population size of the bacilli could be reduced . This phenomenon could lead to increased genetic drift , limiting the removal of detrimental mutations through natural selection . Relaxed selection can also act during adaptation to a new niche on those genes for which a selective advantage for maintenance is lost; alternatively , the gene function has become disadvantageous in the new niche . Through base-level inspection of the genome sequences we identified pseudogenes in M . africanum GM041182 and verified pseudogene annotation in M . tuberculosis H37Rv and M . bovis AF2122/97 . We identified 120 pseudogenes across the three genomes ( M . africanum GM041182 , M . tuberculosis H37Rv , M . bovis AF2122/97 ) ; 20 were in PE-PGRS/PPE family CDSs and in insertion sequence element transposase genes . Both PE-PGRS/PPE family CDSs and insertion sequence elements are known to be associated with intra-genomic recombination and are susceptible to gene disruption [32] , [33] , [34] . We compared the remaining candidate pseudogenes with available draft genomes from different strains belonging to the MTBC lineages and the genome sequence of a M . canetti strain as outgroup . By using an outgroup we could determine the genotype of the most likely common ancestor of the MTBC for the different candidate pseudogenes and determine which ones were shared by other strains apart from M . africanum GM041182 , M . tuberculosis H37Rv , M . bovis AF2122/97 ( Figure 2 , Supplementary Table S2 ) . We found that some of the pseudogenes identified were strain-specific occurring only in one of these three strains ( 20 in GM041182 , 7 in H37Rv and 9 in M . bovis ) . More importantly , some of the pseudogene mutations were shared by a large group of strains . For example , 12 were common to the M . africanum West-African clade 2 and 13 common to both M . africanum clades . In terms of function , the majority of pseudogenes are hypothetical proteins ( N = 39 ) , PE-PGRS family proteins and phage-related ( 20 ) , metabolic enzymes ( 13 ) and transcriptional regulators ( 5 ) ( Figure 3 ) . Many seem to affect systems which are likely to have functional redundancy due to the presence of paralogous or analogous genes or pathways in the genome . For example , there are three pathways for trehalose biosynthesis in mycobacteria [35]; M . africanum and M . bovis each have a trehalose biosynthesis pseudogene but are affected in different genes . M . bovis treY , encoding maltooligosyltrehalose synthase , has a frameshift due to an internal 806bp deletion while M . africanum has a nonsense mutation in a gene ( MAF20180 ) which has been shown to encode a trehalose-phosphate phosphatase . As a component of cell-wall glycolipids , trehalose has been implicated in host tissue damage [35] . Another example is the P450 family of enzymes: there are 21 in the genome of M . africanum GM041182 , two of them , MAF35300 and MAF31280 , are disrupted in all M . africanum strains while another ( MAF22860 ) , which has been shown to be essential for viability in M . tuberculosis , was identified in M . africanum GM041182 [36] . Furthermore , of the 10 loci containing at least one polyketide synthase gene , one is disrupted in GM041182 , another is disrupted in all West-African 2 clade strains , and another in West-African 2 and animal strains . The mycobacterial MmpL-family of proteins have a function in lipid transport and have been shown to contribute to M . tuberculosis intracellular survival [37] . Both clades of M . africanum , as well as animal strains , carry the same nonsense mutation in the 3′ end of the mmpL12 gene ( MAF15490 ) and in M . bovis the mmpL1 gene ( MAF04040 ) has a central frameshift . These mutations may be predicted to impair lipid transport function , although the presence of 12 mmpL paralogues per genome implies some degree of redundancy . Another redundant system affected by mutation in M . bovis is the so-called mammalian cell entry ( mce ) operons . M . bovis has two adjacent pseudogenes ( mce2D and mce2E ) in one of the four mce ( mammalian cell entry ) operons . In M . tuberculosis , deletion of the mce2 operon attenuates the ability to infect mice [38] , and deletion of more than one mce operon has a cumulative effect indicating non-redundant roles during infection [39] . The ability to metabolize nitrate to nitrite is thought to be important for M . tuberculosis to persist under anaerobic conditions during dormancy and also appears to have functional redundancy [40] . M . africanum GM041182 has a pseudogene relevant to nitrate metabolism ( narX ) and all M . africanum and animal strains harbor a narU pseudogene . Although the majority of nitrate reductase activity in vitro is due to narGHJI [41] , narX , which encodes a fusion protein equivalent to parts of NarG , NarJ and NarI , has also been shown to have a role in dormancy [40] . NarU is thought to be involved in transport of nitrate into and nitrite out of the bacterial cell though again its function is thought to be secondary to that of the more active narK2 which coincidentally is adjacent to narX . M . africanum West-African clade 2 strains have frameshift mutations in one of the 17 adenylate cyclase genes in the genome ( MAF03880 ) . In M . tuberculosis the MAF03880 orthologue ( Rv0386 ) was recently found to produce a cyclic AMP burst within macrophages that influences cell signaling . Loss of Rv0386 resulted in lower TNF-a induction , decreased immunopathology in animal tissues , and diminished bacterial survival [42] . Three genes with a role in drug efflux have been disrupted; one in M . africanum GM041182 strain ( MAF03440 ) , one in M . bovis ( orthologue of MAF18990 ) and one in all the so-called ‘modern’ MTBC strains ( MAF23460 ) . The isoniazid inducible gene , iniA ( MAF03440 ) , thought to be involved in an efflux pump for two of the 1st line TB drugs , isoniazid and ethambutol [43] has a 5′ nonsense mutation in M . africanum GM041182 . In M . tuberculosis an iniA deletion mutant showed increased susceptibility to isoniazid [43] , suggesting that M . africanum may be more susceptible to isoniazid than M . tuberculosis . This mutation is however not present in other M . africanum strains of which the genome is available , nor in clinical isolates of the same lineage originating from Burkina Faso and Cote d'Ivoire ( data not shown ) , suggesting that this polymorphism is unique to strain GM041182 . Deletion of the M . smegmatis orthologue of MAF18990 has been shown to result in reduced resistance to ethidium bromide , acriflavine and erythromycin [44] . More interesting is the evolution of the MAF23460 gene . Its homologue in H37Rv is Rv2333c . By inspecting the alignment of both genes a single base pair deletion in the H37Rv leads to a longer product than that observed in M . africanum GM041182 strain ( 524 residues in M . africanum GM041182 versus 538 residues in M . tuberculosis H37Rv ) . By comparing with the rest of strains of the complex it becomes clear that the single base deletion occurred in the common ancestor of ‘modern’ lineages representing in this case a possible gain of function rather than a pseudogenetization per se of the ancestral genes . Rv2333c has been shown to be involved in export of spectinomycin and tetracycline and thus contributes to the intrinsic resistance of M . tuberculosis to these antibiotics [45] , which may thus be more effective against M . africanum . Further pseudogenes affect non-redundant systems such as biosynthesis of vitamins B12 ( cobalamin ) and B6; three genes in a cobalamin biosynthesis operon ( MAF20880 , MAF20850 and MAF20870 ) have the same pseudogene allele in both M . africanum and M . bovis while the pdxH ( MAF26250 ) vitamin B6 biosynthesis gene has a central frameshift mutation in M . africanum GM041822 strain . Supplementation of these vitamins may support growth of M . africanum and reduce the growth delay of M . africanum relative to M . tuberculosis . A notable non-redundant pseudogene is the previously identified orthologue of Rv3879c ( MAF38940 ) , part of the RD1 region [46] , that was found to be essential for ESAT-6 secretion , but not CFP-10 , in M . marinum but not in M . tuberculosis . In a recent study using immunoblots for ESAT-6 and control antigens , we found ESAT-6 secretion to be similar between M . africanum GM041182 , M . africanum GM041182 complemented with Rv3879c , and M . tuberculosis H37Rv [47] , which does not corroborate the attenuated ESAT-6 response in M . africanum infected people [11] . Given the ESX-1 homology throughout the MTBC it is currently not clear how equal amounts of secreted ESAT-6 between M . africanum GM041182 and M . tuberculosis H37Rv can correlate with the attenuated ESAT-6 response observed in M . africanum infected people . Ongoing immunoepidemiological analyses however suggest that the attenuated ESAT-6 phenotype may cluster with sub-lineages within the M . africanum West African 2 lineage . In addition , we identified a deletion in the upstream regulatory region of Rv3616c whose expression is related with ESAT-6 secretion [48] . This polymorphism in GM041182 is shared with animal strains , in which it is responsible for decreased expression of Rv3616c ( Roger Buxton , personal communication ) . Interestingly , ESAT-6 is highly immunogenic in M . bovis infected cows [49] , suggesting that the genetic basis for the attenuated ESAT-6 response observed in M . africanum infected persons is specific to M . africanum . Finally , the M . tuberculosis orthologue of MAF29630 ( Rv2958c ) encodes a glycosyl transferase which has been shown , in a co-infection assay , to confer increased resistance to killing by human macrophages [50] . Both M . africanum clades and M . bovis have a single base pair insertion that shortens the gene product ( 367 residues in M . africanum GM041182 versus 429 residues in M . tuberculosis H37Rv ) . On comparative genomics of M . bovis and M . tuberculosis H37Rv , a region unique to M . bovis was designated TB deleted 1 ( TbD1 ) [51] . Subsequent work identified the TbD1 deletion to be shared by “modern” M . tuberculosis lineages , with an intact TbD1 region in other animal strains , M . africanum , and “ancient” M . tuberculosis [15] . Proteins from the TbD1 region were however not immunogenic in an ELISPOT assay that aimed to identify lineage specific immune responses [52] . Comparison of the M . africanum GM041182 genome with reference shotgun sequences for M . bovis ( strains BCG Pasteur 1173P2 , BCG Tokyo 172 , AF2122/97 ) and M . tuberculosis ( strains H37Rv , H37Ra , CDC1551 , F11 ) genomes revealed a single region present in M . africanum GM041182 but deleted in M . bovis and “modern” M . tuberculosis strains . This M . africanum specific locus was designated RD900 . The locus is 3 , 141 bp long and contains a single complete gene ( designated maf1 ( MAF12860 ) ) and the 3′ end of another ( MAF12870 ) ; maf1 encodes a putative ATP-binding cassette ( ABC ) transport protein that has a central ATP-binding domain and six possible membrane-spanning domains in the C-terminal portion . In addition the N-terminal region contains a putative Forkhead associated ( FHA ) domain that may confer the ability to bind DNA and thereby potentially act as a transcriptional regulator . The LpqY-SugA-SugB-SugC ABC transporter ( Rv1235-Rv1238 ) , one of four ABC transporters in M . tuberculosis , has recently been characterized as a recycling system mediating the retrograde transport of the sugar trehalose produced and released by the bacterium [53] . Other ABC transport proteins may mediate efflux of drugs and other compounds ( Rv1218c , [54] ) , with implications for immune responses ( Rv1280c-Rv1283c , [55] ) . Of the 51 proteins in the Pfam database with the same domain architecture as RD900 , only twelve are from outside the order Actinomycetales ( seven from Cyanobacteria and five from Chloroflexi ) ; none have been experimentally characterised . Assuming complete absence of recombination , comparison of the RD900 region of M . africanum GM041182 with all available genomes from the Mycobacterium tuberculosis complex suggests that intact RD900 ( M . africanum GM041182 ) represents the ancestral state of this region and the RD900 region was independently deleted in two lineages: the “modern” M . tuberculosis lineage and a sub-branch of the animal associated lineage leading to M . bovis ( Figure 1 ) . We checked this in two ways . First , we aligned complete genomes available for the MTBC with M . africanum GM041182 . Secondly , we did local BLAST searches of the region including the two flanking genes ( from MAF12850 to MAF12880 ) ( see Figure 4 ) . The very unusual occurrence of independent deletions generating the same functional gene , pknH , is explainable when the flanking regions are considered . In M . africanum GM041182 , maf1 is flanked by similar , co-directional genes , each encoding a protein kinase . The RD900 deletion appears to have been generated by recombination between these flanking genes to form pknH , a protein kinase-encoding gene found in M . bovis and M . tuberculosis ( Figure 2a ) , thus pknH is apparently a composite gene made up by intra-genomic recombination between two homologous , physically close , genes . Accordingly we have designated the flanking genes in M . africanum GM041182 as pknH1 ( downstream ) and pknH2 ( upstream ) . These flanking CDSs in M . africanum GM041182 have a high level of amino acid identity for the first two-thirds of their length ( Figure 2b ) followed by divergent sequences from codon 424 ( PknH1 ) and 373 ( PknH2 ) , onwards . However , the homologous regions of PknH1 and PknH2 in M . africanum GM041182 have two significant differences . The first is a 21 amino acid sequence insertion/deletion region , present in PknH1 from codons 194 to 214 but absent from PknH2 . The second is a substitution region of 53 amino acids in PknH1 ( 298–350 ) with low identity to a region of 23 amino acids in PknH2 ( 277–299 ) . These two differences account for the 50 codon difference in the region of high identity between PknH1 and PknH2 . More importantly , the substitution region can be used to demonstrate that the deletion of the pknH gene in M . tuberculosis ( called RD900h , Figure 1 ) was independent of the RD900 deletion found in M . bovis ( RD900a , Figure 1 ) . Deletion RD900h generates a composite pknH gene in M . tuberculosis identical to M . africanum pknH1 in the substitution region while the RD900a deletion found in M . bovis generates a composite pknH gene with a substitution region identical to that of M . africanum pknH2 . This implies that the RD900h and RD900a deletions had end points before and after the substitution region , respectively . The remaining 3′ portion of pknH for M . tuberculosis and M . bovis has a high degree of similarity to M . africanum pknH1 , consistent with the architecture of the region in M . africanum . The formation of an intact protein kinase in M . tuberculosis and M . bovis , where deletion could easily have resulted in two non-functional gene fragments , could be used to suggest a selective advantage for the reduction in the number of protein kinases caused by the RD900 deletion . However , this conclusion must be tempered by the ease with which this deletion can be generated . We assume that at least one functional pknH gene is required by strains of the M . tuberculosis complex so it is not unexpected that extant strains appear to have a functional gene at this locus . Broadening the phylogenetic scope of the analysis to include complete genomes from other members of the genus Mycobacterium reveals further patterns of mycobacterial genome evolution associated with RD900 . It seems plausible that the ancestral chromosome arrangement for the clade including the M . tuberculosis and M . avium complexes was similar to that seen in M . marinum where an extra 13 , 772 bp flanks the pknH2 side of RD900 ( Figure 5 ) . This flanking region ends with another protein kinase gene which we designate here as pknH3 and the M . tuberculosis complex ancestral genome may have undergone a deletion due to recombination between pknH2 and pknH3 . The M . ulcerans Agy99 genome appears to have undergone the deletion of MURD111 ( which removes pknH2 ) and a long-range rearrangement separating pknH1 and maf1 on the left flank from the right hand flank which actually carries pknH3 . The genomes of M . avium and M . avium subspecies paratuberculosis have simple RD900 deletions akin to those seen in the modern M . tuberculosis and M . bovis lineages; M . leprae TN genome has a similar deletion pattern but with extra DNA loss equivalent to the region from 1412285 to 1423476 in M . africanum GM041182 , which results in the loss of nine genes including pknH1 , maf1 and pknH2 . Curiously , M . smegmatis has a similar deletion pattern to M . leprae suggesting a convergent event in this distant “rapid growing” relative . Also interesting is the high degree of synteny between M . africanum GM041182 and M . kansasii ATCC 12478 with equivalent RD900 arrangement and colinearity extending for at least 6 kilobases on both flanks .
The M . africanum GM041182 genome is , as expected , highly homologous to those of other members of the M . tuberculosis complex , yet contains a unique sequence , RD900 , that was independently lost during the evolution of two important lineages within the complex; the “modern” M . tuberculosis group and the lineage leading to M . bovis . In addition , RD900 is variably present in atypical mycobacteria , with evidence for repeated independent deletion events . We can expect to learn more about the phylogenetic position of this deletion as more mycobacterial genomes are sequenced , with this complete M . africanum West African 2 sequence serving as an alternative reference for the mapping of further M . africanum genomes generated using Next Generation Sequencing techniques . Determining the function of the deleted gene , maf1 , and the phenotypic consequences of its deletion will require further study but nevertheless this occurrence may provide valuable insight into the evolution of the complex . The similarity in pseudogene repertoire suggests that M . africanum has a similar evolutionary history to M . bovis and it is tempting to speculate that this may have involved adaptation to a non-human animal host , though it must be noted that for both lineages nearly half of the pseudogenes are unique , so subsequent adaptations may have occurred since their divergence reflecting contemporary niche differences . Thus far , no candidate animal reservoir has been detected for M . africanum . Extensive searches among cattle , sheep , pigs , and goats in the Gambia and neighbouring countries have not identified mycobacterial infection nor disease [16] , [56] , [57] . Phylogenetically , the Dassie bacillus [58] and the recently identified Mycobacterium mungi [59] , are the closest relatives of M . africanum within the M . tuberculosis complex . The Dassie bacillus has been isolated from Dassies , or Rock Hyrax , in South Africa [60] , and M . mungi causes disease in troops of banded mongoose in northern Botswana [59] . However , an extensive search for mycobacteria in terrestrial small mammals in Benin , West Africa , did not identify any members of the M . tuberculosis complex [61] . The lack of spread of M . africanum from West Africa to the Americas at the time of the slave trade remains enigmatic . Today , M . africanum is rarely isolated outside of West Africa , typically in first degree immigrants [62] . In a study in Ghana , host polymorphisms were identified with differential protection against M . tuberculosis versus M . africanum in both directions [63] , [64] , although the degree of selective advantage conferred by these polymorphisms is unclear . The majority of the pseudogenes detected are only disrupted by a single base mutation , either by an insertion/deletion leading to a frameshift or by substitution leading to a nonsense mutation . As expected for a recently evolved pathogen no further disrupting mutations have been identified in the pseudogenes . Similarly , in a comparison of several MTBC genomes that included GM041182 , no mutations were identified in the promoter region of the pseudogenes , supporting the notion that “pseudogenization” in the MTBC is recent [65] . A formal statistical testing of the rate of acquisition of pseudogenes cannot be carried out because of the bias in the discovery of the pseudogenes described to those observed in the comparison of M . africanum GM041182 with the two other strains leading to a phenomenon of pseudogene discovery bias . However , the high number of pseudogenes in M . africanum and other strains of the M . tuberculosis complex ( MTBC ) suggest that genome erosion is ongoing . Most likely this reflects several different phenomena that have lead to the downsizing of the MTBC genomes as compared to other free-living or opportunistic Mycobacteria [66] , [67] . This could be partly due to its recent evolution as an intracellular pathogen , making some functions that served a free-living lifestyle redundant to the MTBC , which was therefore prone to lose the function due to relaxed selection . At the same time natural selection can act to favour the loss of some genes . These “anti-virulence genes” can be lost because they can be detrimental for the pathogenic lifestyle as has been described for other species [68] and suggested for some known deletion events in the MTBC [69] . Finally , the increased genetic drift imposed by transmission bottlenecks and changes in population size of its host , lead to a weakened effect of natural selection and increased accumulation of functional mutations , many of them detrimental [14] . Further studies , such as complementing the virulence gene Rv0386 in M . africanum and assessing the effect in the appropriate animal model , can assess to which extent its lower progression to disease is explained by these pseudogenes . Moreover , the presence in M . africanum GM041182 of the original version of MAF23460 ( Rv2333c ) without gain of function suggests that M . africanum ( and other ancestral M . tuberculosis complex lineages ) lack this functional efflux pump and may be more susceptible to antibiotics , possibly including spectinomycin and tetracycline . Differentiating these processes by comparative genomics within and outside the complex could provide clues about how the tight relationships between MTBC species and their respective hosts arose in the first place , and how the ongoing erosion described here generates different genetic backgrounds within the complex than can explain some of the differences associated with diversity in disease outcome [70] . | Mycobacterium africanum , a close relative of M . tuberculosis , is studied for the following reasons: M . africanum is commonly isolated from West African patients with tuberculosis yet has not spread beyond this region , it is more common in HIV infected patients , and it is less likely to lead to tuberculosis after one is exposed to an infectious case . Understanding this organism's unique biology gets a boost from the decoding of its genome , reported in this issue . For example , genome analysis reveals that M . africanum contains a region shared with “ancient” lineages in the M . tuberculosis complex and other mycobacterial species , which was lost independently from both M . tuberculosis and M . bovis . This region encodes a protein involved in transmembrane transport . Furthermore , M . africanum has lost genes , including a known virulence gene and genes for vitamin synthesis , in addition to an intact copy of a gene that may increase its susceptibility to antibiotics that are insufficiently active against M . tuberculosis . Finally , the genome sequence and analysis reported here will aid in the development of new diagnostics and vaccines against tuberculosis , which need to take into account the differences between M . africanum and other species in order to be effective worldwide . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [
"medicine",
"bacterial",
"diseases",
"infectious",
"diseases",
"genomics",
"evolutionary",
"biology",
"biology",
"computational",
"biology",
"microbiology",
"population",
"biology"
] | 2012 | The Genome of Mycobacterium Africanum West African 2 Reveals a Lineage-Specific Locus and Genome Erosion Common to the M. tuberculosis Complex |
Influenza A viruses ( IAVs ) inhibit host gene expression by a process known as host shutoff . Host shutoff limits host innate immune responses and may also redirect the translation apparatus to the production of viral proteins . Multiple IAV proteins regulate host shutoff , including PA-X , a ribonuclease that remains incompletely characterized . We report that PA-X selectively targets host RNA polymerase II ( Pol II ) transcribed mRNAs , while sparing products of Pol I and Pol III . Interestingly , we show that PA-X can also target Pol II-transcribed RNAs in the nucleus , including non-coding RNAs that are not destined to be translated , and reporter transcripts with RNA hairpin structures that block ribosome loading . Transcript degradation likely occurs in the nucleus , as PA-X is enriched in the nucleus and its nuclear localization correlates with reduction in target RNA levels . Complete degradation of host mRNAs following PA-X-mediated endonucleolytic cleavage is dependent on the host 5’->3’-exonuclease Xrn1 . IAV mRNAs are structurally similar to host mRNAs , but are synthesized and modified at the 3’ end by the action of the viral RNA-dependent RNA polymerase complex . Infection of cells with wild-type IAV or a recombinant PA-X-deficient virus revealed that IAV mRNAs resist PA-X-mediated degradation during infection . At the same time , loss of PA-X resulted in changes in the synthesis of select viral mRNAs and a decrease in viral protein accumulation . Collectively , these results significantly advance our understanding of IAV host shutoff , and suggest that the PA-X causes selective degradation of host mRNAs by discriminating some aspect of Pol II-dependent RNA biogenesis in the nucleus .
Inhibition of host gene expression , termed “host shutoff” , is thought to enable viruses to simultaneously inhibit innate immune responses and provide preferential access for viral mRNAs to the cellular translation machinery . Influenza A virus ( IAV ) has long been known to carry out host shutoff , and multiple shutoff mechanisms have been reported for this virus , including translation blockade [1] , inhibition of polyadenylation and nuclear export of host pre-mRNAs by the IAV NS1 protein [2] , and degradation of the host RNA polymerase II complex [3] . Because some of these mechanisms are specific to certain IAV strains [4] , it has long been suspected that more universal IAV host shutoff mechanisms exist . The recent discovery of the highly conserved RNA endonuclease PA-X [5] has prompted re-examination of established models of IAV host shutoff . Viruses from several divergent families use virus-encoded RNA endonucleases to broadly degrade host mRNAs and reduce host protein production [5–9] . Although host shutoff ribonucleases ( RNases ) generally have broad specificity in vitro , several studies have shown unexpected selectivity for different types of host transcripts [10–15] . PA-X limits accumulation of host mRNAs and proteins in infected cells and suppresses host responses to infection [5 , 16–19] , but the mechanistic determinants of selectivity , cleavage and degradation are not yet known . IAV is a negative strand RNA virus with a genome consisting of eight segments . PA-X protein is encoded on IAV genome segment 3 , which also produces polymerase acidic protein ( PA ) , one of the three subunits of the viral RNA-dependent RNA polymerase ( RdRp ) . It is generated by ribosomal pausing on a rare CGU codon that results in a +1 frameshift and read-through of an alternative open reading frame ( ORF ) [5 , 20] . PA-X comprises the amino-terminal 191 amino acids of the polymerase subunit PA fused to a carboxy-terminal domain ( termed “X-ORF” ) of either 41 or 61 amino acids that result from the frameshift [5 , 21] . Consequently , PA-X lacks the carboxy-terminal domain of PA responsible for its recruitment into the RdRp complex . The shared PA/PA-X amino-terminal domain includes an RNA endonuclease domain that is required for PA-X shutoff activity [5] . Thus , PA-X has a function analogous to host-shutoff proteins from other viruses that trigger RNA degradation . These factors include SARS coronavirus nsp1 [9 , 22] , herpes simplex virus 1 ( HSV-1 ) vhs [6 , 23] and Kaposi’s sarcoma-associated herpesvirus ( KSHV ) SOX [8] . In general , these proteins are unrelated at a molecular level , although both PA-X and the SOX family of proteins belong to the PD- ( D/E ) XK nuclease superfamily [24–28] . All known host-shutoff RNases use a similar mechanism of action , causing endonucleolytic cleavage of the RNA and relying on host enzymes to complete RNA degradation [14] . Moreover , they are all selective for translatable RNA polymerase II ( Pol II ) transcripts , but spare non-coding RNAs ( ncRNAs ) synthesized by Pol I and Pol III [14] . This selectivity has been linked to the process of translation or loading into translation initiation complexes [14 , 22 , 29–31] . RdRp-transcribed IAV mRNAs share essential features with host mRNAs like a 5’ 7-methyl guanosine ( m7G ) cap and a 3’ poly-adenylate ( poly ( A ) ) tail . 5’ m7G caps are acquired by “cap-snatching” , whereby the PA subunit cleaves nascent host Pol II-transcribed RNAs at a position 10–14 nucleotides downstream from the 5’ cap [32] . The RdRp complex uses these fragments to prime viral mRNA synthesis . The poly ( A ) tail is generated by RdRp “stuttering” , allowing reiterative copying of a short poly-uridine sequence at the 5’ end of the template genome segment [33] . The acquisition of m7G caps and poly ( A ) tails allows efficient loading of ribosomes and translation of IAV mRNAs; however , the similarity between host and viral mRNAs raises the question of their susceptibility to PA-X-mediated degradation . Here , we demonstrate that PA-X selectively targets host RNA transcribed by the RNA Pol II complex for cleavage , and degradation is completed by the host 5’->3’ exonuclease Xrn1 . By contrast , host transcripts generated by other Pol complexes resist PA-X-mediated degradation . Interestingly , selective targeting is not linked to translation , as we observe that PA-X also degrades non-coding Pol II transcripts , and may instead be linked to distinct features of Pol II transcript biogenesis in the nucleus . Consistent with these observations , we show that PA-X is recruited to the nucleus via X-ORF interactions that involve highly conserved basic residues previously shown to be important for the shutoff function [34] . Accordingly , we find that IAV mRNAs generated by viral polymerase complexes are intrinsically resistant to PA-X-mediated degradation . In addition , we demonstrate that PA-X is required for efficient translation of viral mRNAs , as viral protein accumulation is significantly diminished in cells infected with PA-X deficient viruses . Taken together these findings suggest that IAV PA-X hijacks cellular RNA biogenesis processes to direct the degradation of host RNAs , and that the distinct biogenesis mechanism for viral mRNAs provides a convenient way to discriminate host and viral products . Thus , although PA-X shares some mechanistic properties with other host shutoff RNases , it also displays distinctive features that advance our understanding of host shutoff .
Diverse RNA species are generated by three host DNA-dependent RNA polymerases; Pol II transcribes mRNAs and some non-coding RNAs ( ncRNAs ) , whereas Pol I transcribes the rRNA precursor 47S and Pol III transcribes a variety of short ncRNAs including the 5S rRNA . To determine whether PA-X could trigger the degradation of distinct host RNA species , we transfected HEK 293T cells with plasmid reporters encoding an RFP gene driven by a Pol II promoter and a GFP gene under the control of a Pol I- , Pol II- , or Pol III-driven promoter . We observed that ectopic expression of PA-X derived from the A/PuertoRico/8/34 H1N1 ( PR8 ) virus together with these reporters consistently inhibited the accumulation of Pol II-driven RFP and GFP transcripts , whereas Pol I and Pol III-driven GFP transcripts accumulated to high levels ( Fig 1A ) . These data suggest that PA-X activity is specific for Pol II transcripts . To determine whether PA-X could similarly selectively target endogenous Pol II-driven transcripts , we constructed HEK 293T- and A549-based cell lines that stably express PA-X in a doxycycline-inducible manner ( iPA-X cells ) . In comparison to an inducible RFP control , PA-X expression in the HEK 293T cell lines dramatically reduced endogenous mRNA levels for actin , EEF1A , GAPDH , GUSB , HIST1H3C , tubulin and RPS6 , but had a much smaller effect on RPS18 and POLR2A mRNAs ( Fig 1B ) . By contrast , consistent with our results with reporter constructs ( Fig 1A ) , PA-X did not decrease the levels of the Pol I transcript 47S or several Pol III transcripts ( 5S , 7SK and 7SL ) . Both the specificity for Pol II transcripts and the variable level of down-regulation of different mRNAs were also observed in the A549 lung carcinoma cell line ( Fig 1C ) . In addition , expression of the PA-X RNase domain catalytic mutant D108A [5 , 35] in A549 cells did not affect RNA levels ( Fig 1C ) . This result indicates that the mRNA down-regulation is likely due to increased degradation , as it requires an intact RNase domain . Multiple independently isolated HEK 293T and A549 cell lines generated similar results ( S1A and S1B Fig ) . Thus , PA-X expressed in isolation decreases the levels of a broad range of host mRNAs with varying efficiency , but Pol I- and Pol III-transcribed RNAs resist degradation , pointing at a mechanism of action similar to that of previously described host-shutoff RNases from other viruses [14] . We previously found that herpesvirus and coronavirus host shutoff endonucleases employ analogous mechanisms to trigger widespread RNA degradation , whereby endonuclease cleavage of target RNA is followed by processive 5’->3’ degradation by the host exonuclease Xrn1 [14] . To determine whether PA-X requires host 5’->3’ exonucleases to complete RNA degradation , we tested plasmid reporters that contain a pseudo-knot forming sequence from West Nile virus ( SLII [36] ) in either the GFP coding region or the 3’ UTR ( Fig 2A ) ; the presence of the SLII element leads to protection of the downstream RNA from digestion by 5’->3’ exonucleases [14 , 29] . Northern blotting using a probe against the 3’ untranslated region ( UTR ) of GFP showed that ectopic expression of PA-X reduced the levels of full-length GFP mRNA and that a SLII-protected species appeared , which indicated that host 5’->3’ exonucleases were required for full degradation of target mRNAs ( Fig 2B ) . To determine whether Xrn1 in particular is required for full PA-X-initiated RNA degradation , Xrn1 expression was silenced in HEK 293T cells by doxycycline-inducible shRNA expression ( Fig 2E ) . As expected from previous observations ( Fig 1A ) , ectopic PA-X expression reduced GFP mRNA levels in control cells , but induction of Xrn1 shRNA expression reversed this phenotype ( Fig 2C ) . Using a northern blotting approach with a probe specific for GFP mRNA , we found that Xrn1 knock-down in wild-type PA-X-expressing cells caused the appearance of a heterogeneous population of partially digested GFP mRNA products ( Fig 2D , lane 7 ) . This is consistent with a model in which PA-X cleavage occurs throughout the length of the RNA with no discrete target site , followed by full digestion by host 5’-3’ exonucleases . As expected , neither a decrease in full-length GFP RNA levels nor the GFP fragments were detected when the D108A catalytic mutant of PA-X was expressed ( Fig 2D , lanes 3 and 8 ) . The lack of an apparent discrete primary endonuclease cleavage site makes PA-X unique amongst host-shutoff endonucleases studied to date , all of which either reveal sequence-specific cleavage sites on mRNAs , or target the 5’ end of transcripts [14 , 15 , 22 , 29 , 37] . Indeed , analysis of KSHV SOX cleavage products in Xrn1-deficient cells reveals the accumulation of a specific degradation product , reflecting cleavage at a discrete location in the mRNA ( Fig 2D , lane 10 ) , consistent with published reports [29] . Moreover , SARS nsp1 , which is reported to cut RNAs close to the 5’ end [22] did not cause the appearance of partially digested GFP mRNA products ( Fig 2D , lane 9 ) . Together , these data indicate that PA-X degrades host mRNAs in concert with Xrn1 and perhaps other cellular exonucleases , and unlike other host-shutoff endonucleases , lacks obvious sequence or location specificity , consistent with existing in vitro data [38] . Many Pol II-transcribed RNAs are translated . Association with components of the translation apparatus has been shown or proposed to be a determinant of selective mRNA targeting by other host-shutoff RNases , such as HSV-1 vhs [30 , 31] and KSHV SOX [29] . Moreover , SARS nsp1 requires active translation of RNAs for degradation [22] . To investigate the relationship between translation and RNA targeting by PA-X , we employed a Pol II-driven reporter that is not translated ( S2A Fig ) due to the insertion of a hairpin close the 5’ cap ( hp-GFP ) . This reporter mRNA associates with the translation initiation machinery , but cannot be translated because the hairpin blocks ribosome association and/or scanning [39] . We observed that PA-X decreased the levels of the hp-GFP and a control GFP mRNAs to a similar extent , suggesting that targeting is independent of mRNA translation ( Fig 3A ) . Conversely we used a dual-construct T7 polymerase system to direct transcription of luciferase RNA by overexpressed T7 polymerase [40] , rather than cellular Pol II . The T7-synthesized luciferase RNA is actively translated due to the presence of an EMCV internal ribosome entry site ( S2B Fig ) , but luciferase mRNA levels are unaffected by PA-X ( Fig 3B ) . These results indicate that synthesis by RNA Pol II , rather than translatability of the RNA , is a major determinant for targeting by PA-X . In addition to mRNAs , Pol II also transcribes several ncRNAs , including the long ncRNAs MALAT1 and TP53TG1 and the precursor of the small nuclear RNA U2 . Based on our reporter data ( Fig 3A and 3B ) , we hypothesized that Pol II transcribed ncRNAs may also be targeted by PA-X . Consistent with this hypothesis , we found that levels of the MALAT1 and TP53TG1 ncRNAs were reduced in 293T ( Fig 3C , S2C Fig ) and A549 iPA-X cell lines ( Fig 3D , S2D Fig ) . The effect of PA-X on the levels of U2 was variable; however this small nuclear RNA was unaffected in A549 iPA-X cells and some 293T iPA-X clones ( Fig 3C and 3D , S2C and S2D Fig ) . Half-life measurements for the actin mRNA and MALAT1 ncRNA in the presence of wild-type and catalytically inactive PA-X confirmed that PA-X directly affects the stability of the different RNA species ( S2E Fig ) . These results indicated that PA-X targets at least some Pol II synthesized ncRNAs for degradation and that , unlike herpesviral host shutoff endonucleases and SARS nsp1 , association with the protein synthesis machinery is not a prerequisite for targeting by PA-X . To determine whether PA-X selectively targets Pol II-transcribed RNA in the context of IAV infection , A549 cells were infected with PR8 or a mutant PR8-PA ( fs ) virus that should not produce PA-X due to codon optimization that prevents ribosome pausing and frameshifting ( Fig 4A ) . As we previously reported [41] , at later times post-infection PR8 virus causes dramatic depletion of cytoplasmic polyadenylated RNA which drives the nuclear relocalization of poly ( A ) binding protein ( PABP ) . By contrast , in cells infected with the PR8-PA ( fs ) mutant virus , nuclear PABP relocalization is significantly delayed , and only becomes detectable at 12 hours post-infection ( hpi , Fig 4B and 4C ) . The fact that PABP relocalization , a known PA-X-dependent phenotype , is markedly reduced in cells infected with the PR8-PA ( fs ) virus confirms that this virus is PA-X deficient; however , we note that nuclear PABP relocalization was still observed at later times post-infection . Other host shutoff mechanisms and/or leaky expression of PA-X in PR8-PA ( fs ) virus-infected cells may cause late nuclear localization of PABP . For this reason , we selected the 12 hpi time point for the analyses of host transcript levels . Consistent with our observations from PA-X ectopic expression experiments ( Fig 1 , S1 Fig ) , actin and GAPDH mRNA levels were reduced in a PA-X-dependent manner ( Fig 4D ) in PR8 IAV infected cells . We also detected a decrease in the levels of tubulin , POLR2A , and , to a lesser extent , RPS6 and RPS18 , but in these cases the decrease was only partially dependent on PA-X function , and may be due in part to other host-shutoff mechanisms . Interestingly , the Pol II-transcribed ncRNA MALAT1 and the histone mRNA HIST1H3C were strongly down-regulated in both PR8 wt and PR8-PA ( fs ) infected cells ( Fig 4D ) , suggesting that these RNAs are subject to regulation by other viral proteins . In contrast to our results with Pol II transcripts , the levels of Pol I and Pol III-transcribed ncRNAs ( 47S and 7SK , respectively ) were not altered in PR8 or PR8-PA ( fs ) infected cells . Collectively , these data show that during an IAV infection , PA-X selectively targets Pol II-transcribed RNAs for degradation . When determining the titers of recombinant PR8-PA ( fs ) virus stocks generated in our laboratory , we consistently observed reduction in plaque size compared to the parental wild-type PR8 strain ( Fig 5A ) . This suggests that PA-X function is important for efficient multi-round replication of IAV in culture , consistent with previous reports [19] . However , to date the exact contribution of PA-X to virus fitness remains poorly understood . In order to compare viral RNA and protein production between the wild-type PR8 virus and the PA-X deficient PR8-PA ( fs ) strain , we infected A549 cells with the same number of virions and collected parallel samples for immunofluorescence staining , western blotting for viral proteins , and total RNA isolation . Immunofluorescence staining for the viral PA protein confirmed that in our experiments a similar number of cells were infected with either wild-type PR8 or the mutant PR8-PA ( fs ) virus ( Fig 5B ) . Similar PA protein accumulation was also detected in wild-type and mutant virus-infected cell lysates at multiple time points ( Fig 5C ) . However , the accumulation of the viral proteins M1 , NS1 , and especially M2 was significantly slower in cells infected with PR8-PA ( fs ) mutant virus compared to wild-type virus-infected cells ( Fig 5C and 5D ) , which may be the cause for the reduction in plaque size . The reduced viral protein accumulation in the absence of PA-X was unexpected; if PA-X were able to degrade viral RNAs , we would expect an increase in viral protein levels in PR8-PA ( fs ) infected cells . These findings suggest potential secondary consequences of PA-X-mediated RNA degradation on viral protein accumulation . Importantly , we saw comparable levels of most viral mRNAs and genomic vRNAs in PR8 and PR8-PA ( fs ) infected cells ( Fig 5E ) . Only M1 mRNA accumulated to significantly higher levels in PR8-PA ( fs ) infected cells at later time points , with roughly 1 . 6-times more M1 transcript at 12 and 15 hpi ( Fig 5E and 5F ) . M2 and NEP mRNAs , which are generated through splicing of M1 and NS1 transcripts respectively , were slightly reduced in PR8-PA ( fs ) infected cells ( Fig 5E and 5F ) . Metabolic pulse-labeling of nascent RNAs by Click-IT chemistry revealed that the changes in M1 and NEP total mRNA levels resulted from increased ( M1 ) or decreased ( NEP ) synthesis rates , with relative total mRNA levels at 12 hpi overall matching the altered rates of synthesis at 9 hpi ( Fig 5F and 5G ) . Taken together , these data show that unlike cellular mRNAs synthesized by host Pol II , RdRp-generated viral mRNAs are not subject to PA-X mediated degradation . In fact , viral RdRp-generated mRNAs are translated more efficiently in the presence of PA-X , which may be due to reduced competition with host mRNAs for access to translation machinery . The selectivity of PA-X for Pol II transcripts could be linked to the RNA polymerase directly , or to processing events that are specific to Pol II transcripts . In particular , termination of transcription by Pol II is normally followed by addition of a non-templated poly ( A ) tail at the 3’ end of the RNA , which is absent from Pol I and Pol III transcripts , and the canonical polyadenylation signal serves to direct termination , cleavage of the RNA and polyadenylation . An alternative stem loop termination signal is used for histone mRNAs [42] , whereas some ncRNAs have distinct 3’ end processing mechanisms [43 , 44] . To test whether PA-X targeting of host Pol II transcripts is coupled to canonical 3’-end processing , we used a set of constructs that had altered 3’ ends . The 3’ polyadenylation signal and the 3’ untranslated region ( UTR ) of the GFP reporter were replaced by a self-cleaving hammerhead ribozyme ( HR ) or the histone stem loop termination region ( hisSL ) ( Fig 6A ) [45] . The HR sequence obviated the need for the cleavage and polyadenylation machinery in truncating the RNA 3’ end . We found that PA-X was unable to degrade the GFP-HR construct ( Fig 6B and 6C ) , which was also not translated as previously reported ( Fig 6D ) . Addition of a 60 nt templated stretch of adenosines that mimicked a poly ( A ) tail ( GFP-A60-HR ) restored translation of the GFP-HR construct ( Fig 6D ) while GFP-A60-HR mRNA levels were still unaffected by PA-X ( Fig 6B and 6C ) , suggesting that the lack of cellular 3’ end processing of this RNA prevented targeting by PA-X . By contrast , replacement of the 3’ poly ( A ) signal with a histone stem loop termination region promoted down-regulation by PA-X ( Fig 6B and 6C ) , consistent with the fact that the histone HIST1H3C mRNA was also down-regulated by PA-X expression ( Fig 1B and 1C , S1A and S1B Fig ) . Northern blotting confirmed that the size of the GFP RNA species is consistent with the expected processing route ( Fig 6C ) . In addition , we found that similarly to PR8 PA-X , PA-X variants from two other human strains of IAV , a pre-pandemic 2006 H1N1 strain and a 2009 pandemic H1N1 strain could degrade endogenous actin mRNA and transfected poly ( A ) -tailed GFP mRNA , but not GFP-A60-HR mRNA ( Fig 6E ) . Previous examples of protection of specific RNAs from host shutoff RNases have focused on the presence of protective elements . This is the case for SARS CoV mRNAs , which are protected from nsp1-mediated degradation [46] or for the host IL-6 mRNA during KSHV infection , which is protected from SOX-mediated degradation through 3’ UTR elements [47 , 48] . However , our results strongly suggest that resistance of IAV mRNAs to PA-X is due to their differential biogenesis pathway , in particular the 3’ maturation mechanism . These results , together with our viral mRNA data ( Fig 5E and 5F ) , suggest that the process of polyadenylation or the presence of the poly ( A ) tail per se are not sufficient or necessary for PA-X targeting of mRNAs . Instead they hint at a more universal feature of the 3’ end processing of the Pol II transcribed RNAs that earmark target RNAs for PA-X mediated cleavage . Since mRNA processing and maturation are linked to nuclear export , we compared the PA-X mediated changes in reporter transcript levels between cytoplasmic and nuclear RNA fractions . Importantly , PA-X was able to target both the cytoplasmic and the nuclear pools of the susceptible reporter RNAs ( poly ( A ) -tailed GFP and GFP-hisL ) , while we failed to detect down-regulation of the HR bearing constructs in either fraction ( S3A Fig ) . This demonstrates that PA-X is able to function in the nucleus and target transcripts prior to their export in the cytoplasm . Although all other host shutoff RNases are proposed to work in the cytoplasm , our results link PA-X targeting to 3’ end processing ( Fig 6 ) and show degradation of reporter transcripts in the nucleus ( S3A Fig ) . This suggests that PA-X may degrade nascent RNAs shortly after transcription . To test this hypothesis , we measured changes in the levels of endogenous transcripts in the nuclear and the cytoplasmic RNA fractions upon PA-X induction in A549 iPA-X cells ( Fig 7A ) . Consistent with the reporter RNA data , both nuclear and cytoplasmic actin and GAPDH mRNAs were down-regulated ( Fig 7A ) . Next , we compared PA-X dependent decrease in the levels of unspliced pre-mRNAs and mature mRNAs for these genes , as well as the MALAT1 ncRNA , in the nuclear fraction . We found that unlike processed mRNAs and MALAT1 , pre-mRNAs were not down-regulated ( Fig 7B ) . Because most splicing is thought to occur cotranscriptionally [49] , unspliced RNA levels likely reflect nascent RNA levels . Thus , this result reaffirms our finding that a later step of processing specific to Pol II transcripts is crucial for targeting by PA-X . Previous studies have demonstrated that the C-terminal domain of PA-X created by the frameshift , the X-ORF , is important for PA-X shutoff activity [34 , 50] . In particular the recent study by Oishi et al . [34] highlighted the importance of 6 highly conserved basic residues within the first 15 amino acids of X-ORF for PA-X function . Since we demonstrated that PA-X can function in the nucleus , we set out to examine the subcellular localization of PA-X and the effect of the X-ORF on the nucleocytoplasmic distribution of this protein . To this end we constructed a series of GFP fusion constructs containing the full-length wild type ( PA-X-GFP ) or catalytic mutant PA-X protein ( PA-X ( D108A ) -GFP ) , the N-terminal nuclease domain of PA-X ( PA-N191-GFP ) , the X-ORF ( GFP-X61 ) or shortened version of the X-ORF that mimics variants of PA-X with a 41 amino acid tail ( GFP-X41 , Fig 7C and 7D ) . In addition , we created a GFP X-ORF fusion protein in which four out of the six functionally important basic residues were mutated to alanines ( GFP-X61 ( 4A ) , Fig 7C and 7D ) . We observed that GFP-tagged PA-X and all fusion proteins containing the first 41 amino acids of X-ORF , although found throughout the cell , concentrated in the nucleus ( Fig 7E , S4A Fig ) . By contrast , the subcellular distribution of PA-N191-GFP and GFP-X61 ( 4A ) was highly similar to that of GFP alone ( Fig 7E and 7F and S4A Fig ) . Although GFP is small enough to show some nuclear accumulation on its own , the nuclear accumulation of the X-ORF fusion proteins was much more robust , and analysis of cells that expressed lower levels of the fusion proteins showed almost exclusive nuclear accumulation of the protein ( Fig 7E and S4A Fig ) . PA-X-GFP was efficient at blocking expression of co-transfected luciferase reporter constructs despite the large tag ( S4B and S4C Fig ) . Moreover , consistent with previous reports , the lack of the X-ORF severely impaired host-shutoff activity ( S4B and S4C Fig , PA-N191-GFP ) . As expected , despite nuclear accumulation , the PA-X ( D108A ) -GFP mutant , GFP-X61 or GFP-X41 did not possess the shutoff activity due to lack of the functional nuclease domain ( S4B and S4C Fig ) , although a previous report has argued that overexpression of X-ORF peptides alone has effects on gene expression [51] . Finally , we compared the shutoff function of myc-tagged constructs of the wild type PA-X , a PA-X with the four K/R-to-A mutations in the X-ORF ( PA-X ( 4A ) ) , and the N-terminal nuclease domain alone ( Fig 7G and 7H ) using luciferase reporter assay . In this assay , PA-X no longer inhibited luciferase expression when the four basic residues required for the nuclear localization function of the X-ORF ( Fig 7F ) were mutated to alanine , similarly to the complete deletion of the X-ORF ( Fig 7G and 7H ) . Collectively these data reveal a striking correlation between nuclear accumulation of the PA-X-GFP fusion proteins and their shutoff activity . More specifically , it shows that X-ORF acts as a nuclear accumulation domain and that the conserved basic residues important for PA-X shutoff function participate in key molecular interactions governing X-ORF-mediated nuclear accumulation .
Our study provides new mechanistic insights into the specificity of PA-X , the most recently identified viral host shutoff nuclease . We demonstrate here for the first time that PA-X is recruited to the nucleus , selectively targets a subset of host transcripts , and is not active against viral mRNAs . Moreover , we uncover a novel route of target discrimination by a viral RNase , which takes advantage of the divergent mRNA biogenesis mechanisms that generate viral and host transcripts . We demonstrate that PA-X shares some specificity features with other host shutoff nucleases , as it selectively targets products of cellular RNA Pol II , while sparing Pol I and Pol III transcripts , and requires host RNases to complete RNA degradation . Interestingly , the mechanism for PA-X targeting of Pol II transcripts is not related to the translatability of the mRNAs . Instead , PA-X targeting is directly linked to synthesis by RNA Pol II complex or early processing events unique to Pol II transcripts . Moreover , we find that PA-X likely degrades RNAs in the nucleus , because it accumulates in this compartment and affects both the nuclear and cytoplasmic fraction of its target RNAs . Both the host shutoff activity of PA-X and the nuclear localization function of the C-terminal X-ORF are dependent on the presence of a set of basic residues in the X-ORF , indicating a correlation between nuclear localization and RNA targeting . Thus , the mechanism of PA-X targeting may be different from other shutoff RNases because it is tightly linked to the mechanism of biogenesis of host and viral mRNAs in the nucleus . Like other host shutoff endonucleases SOX , vhs and nsp1 [22 , 29 , 37] , PA-X carries out an initial cleavage of its targets and then relies on host enzymes to complete degradation of RNA fragments . Nonetheless , there are key differences between PA-X and these previously described enzymes . PA-X cleavage is not directed by specific sequence elements , as previously reported for the gamma-herpesviral host-shutoff RNases [14 , 15 , 29] , or by proximity to the 5’ end of the message , like SARS CoV nsp1 [22] or alphaherpesvirus vhs [37] . Analysis of the GFP reporter upon Xrn1 knock-down suggests that PA-X cuts RNAs throughout the transcript ( Fig 2D ) . Whether this is true for endogenous mRNA targets remains to be determined . Moreover , whereas a tight relationship with translational machinery has been reported for some of the endonucleases ( vhs , nsp1 [22 , 30 , 31] ) , we find that translation is not a key determinant of PA-X targeting . Pol II-transcribed RNAs that are not translated , like the 5’ hairpin containing GFP reporter and the endogenous ncRNAs MALAT1 or TP53TG1 are down-regulated by PA-X ( Fig 3A , 3C and 3D , S2A , S2C–S2E Fig ) . By contrast , translated viral mRNAs ( Fig 5C–5F ) and the translated GFP-A60-HR and T7-polymerase driven luciferase reporter constructs ( Figs 3B , 6B–6E , S2B and S3A Figs ) are not affected . Although we cannot exclude that other processes connected to translation are important for RNA degradation , our data strongly indicate that translation per se is not required . Our data point instead to the processing of Pol II transcripts in the nucleus as the link between PA-X and its target . Importantly , our examination of endogenous and reporter Pol II transcripts that differ in their 3’ end processing suggests that this step in biogenesis may be recognised by PA-X to find its targets . In our experiments , GFP reporters bearing the canonical polyadenylation signal ( PAS ) or a histone stem loop at the 3’ end were efficiently inhibited by PA-X . By contrast , the same reporter with a templated poly ( A ) stretch followed by a self-cleaving hammerhead ribozyme sequence was not degraded by PA-X despite being translated ( Fig 6 ) . Replication-dependent histone mRNAs and polyadenylated mRNAs are processed very differently , but some proteins participate in both mechanisms including the scaffold protein symplekin and the cleavage and polyadenylation specificity factor ( CPSF ) complex subunit CPSF-73 [52 , 53] . Nuclear localisation and targeting of nuclear mRNA pools by PA-X demonstrated in our study also strongly hint at a mechanism that involves association with target mRNAs prior to their engagement in translation in the cytoplasm . The fact that in our experiments nuclear unspliced pre-mRNAs were largely unaffected by PA-X ( Fig 7B ) also suggests that the later stages of mRNA maturation are responsible for PA-X recruitment . A host protein involved in Pol II transcript 3’ end processing could in theory earmark the RNA for cleavage by PA-X , although our data on the nuclear MALAT1 ncRNA hints at a more complex model . The MALAT1 gene includes a canonical PAS , but the most abundant MALAT1 transcript is processed by RNase P at a location upstream of the PAS , and is therefore not polyadenylated [43] . It is possible that the presence of the PAS could still promote co-transcriptional recruitment of cellular factors necessary for PA-X targeting , regardless of whether it is used for 3’-end processing , because the PAS can direct transcription termination independently of 3’ cleavage and polyadenylation [54] . In general , determining whether PA-X physically interacts with transcription and RNA processing machinery will be key to understanding how PA-X targets Pol II-transcribed RNAs in the nucleus . A major challenge for future studies will be to disentangle the highly interconnected steps of RNA metabolism to better understand PA-X function . Our data also demonstrate a new cellular function for the C-terminal X-ORF , whose required function in host shutoff [34 , 50] is poorly understood . We found that the conserved 41-amino acid portion of the X-ORF is sufficient to cause nuclear accumulation of GFP fusion proteins and acts as a nuclear targeting sequence ( Fig 7C–7F and S4A Fig ) . Moreover , nuclear localization of PA-X is tightly linked to reporter shutoff efficiency ( Fig 7G and 7H ) . At present it remains to be determined whether the X-ORF contains a functional NLS or whether it causes nuclear accumulation through interactions with other NLS-containing proteins . The latter mechanism appears more likely , because the region responsible for the nuclear localization function of X-ORF lacks homology to well-characterized NLS consensus sequences ( Fig 7D ) . Also , because increased in vitro activity has been reported for PA-X vs . the N terminal RNase domain alone [38] , the X-ORF probably has additional roles in RNA degradation by PA-X . The fate of viral mRNAs and proteins during host shutoff has been a topic of intense study , as the expectation is that viral products should be protected from host shutoff to confer a selective advantage . For example , SARS CoV transcripts are protected from RNA degradation by nsp1 by a common 5’ leader sequence [46] . Interestingly , some viruses , like gamma-herpesviruses , have no mechanism for widespread protection of viral mRNAs [55] . Early studies of IAV host shutoff demonstrated that the virus selectively inhibits translation of host mRNAs , while IAV mRNAs appear resistant [1] . One of the well characterised host shutoff mechanisms employed by IAV is mediated by the NS1 protein from human IAV strains that binds and inactivates cellular CPSF30 , preventing polyadenylation and nuclear export of host pre-mRNAs [2] . IAV mRNAs are exempt from CPSF30 inactivation because they rely exclusively on RdRp for addition of the poly ( A ) tail . Remarkably , we discovered that mRNA 3’ end processing in the nucleus may serve as the basis for the protection of viral mRNAs from PA-X as well . Future studies will determine whether similarly to NS1 , PA-X is recruited to its target RNAs through direct interaction with one of the subunits of 3’ end processing machinery . We observed no major effects of PA-X on the levels of viral mRNAs and vRNAs comparing viral RNA levels between wild-type and PA-X deficient mutant viruses ( Fig 5E ) . The only exceptions are the M1 mRNA , which is transcriptionally up-regulated , albeit to a small degree , and the spliced M2 and NEP mRNAs , which are synthesized at slightly lower rates ( Fig 5E–5G ) . One recent study reported an increase in RdRp activity and levels of PA protein in PA-X-deficient viruses [17] . We note that we do not see an increase in PA levels in the PR8-PA ( fs ) mutant virus-infected cells ( Fig 5B–5D ) . In fact , several of the IAV PR8 proteins accumulated to lower levels in the absence of PA-X , including M1 , despite an increase in its mRNA levels . This suggests that , like HSV-1 vhs [56] , PA-X may also function to reduce competition for translational machinery between host and viral mRNAs . In turn , changes in viral protein accumulation could explain the transcriptional and splicing changes we detected , because viral proteins have roles in regulating viral gene expression . For example , NS1 is required for the production of correct levels of spliced M2 mRNA [57] , and we detect both changes in NS1 protein accumulation and in M2 mRNA and protein levels . The reduced accumulation of M2 in particular may be responsible for the small plaque phenotype that we have observed with PR8-PA ( fs ) virus , as mutations in M2 have been reported to give rise to a small plaque phenotype [58] . Previous studies using different strains of IAV have reached different conclusions on the role of PA-X in replication of the virus in tissue culture and virulence in animal models [5 , 16–19] , suggesting that the effect of PA-X on viral replication and disease is strain dependent . In this study we used the mouse adapted PR8 strain that possesses NS1 protein lacking the ability to bind CPSF30 and block host mRNA polyadenylation . Because PR8 lacks at least one of the other IAV host-shutoff mechanisms , PA-X could play a more important role in host shutoff and/or have stronger effect on viral fitness in our model . Additional studies examining the interplay of the PA-X and NS1-mediated host shutoff mechanisms will further advance our understanding of IAV host shutoff and its role in viral replication , pathogenicity , and species adaptation .
pCR3 . 1-PA-X-myc and pCR3 . 1-PA-N191-myc were generated by inserting the PCR-amplified full length PA-X or the PA nuclease domain sequences from pCR3 . 1-PA-X plasmid [41] into the pCR3 . 1-myc vector [59] between KpnI and MluI sites . Subsequently , pCR3 . 1-PA-X ( D108A ) -myc and pCR3 . 1-PA-X ( 4A ) -myc vectors were generated by site-directed mutagenesis of the pCR3 . 1-PA-X-myc plasmid . In order to create the GFP-tagged constructs pCR3 . 1-PA-X-GFP , pCR3 . 1-PA-X ( D108A ) -GFP , and pCR3 . 1-PA-N191-GFP , the myc tag sequence flanked by MluI and XhoI sites was replaced with the PCR-amplified EGFP ORF . pCR3 . 1-EGFP control vector was created by inserting EGFP ORF between BamHI and XhoI sites of pCR3 . 1-myc plasmid . An AgeI restriction site was introduced by PCR immediately upstream of the EGFP ORF stop codon to enable insertion of the C-terminal X-ORF sequences . Subsequently , full-length ( X61 ) , shortened ( X41 ) , and mutant [X61 ( 4A ) ] X-ORF sequences were PCR-amplified from the pCR3 . 1-PA-X-myc and pCR3 . 1-PA-X ( 4A ) -myc templates without the inclusion of the myc tag and inserted between AgeI and XhoI sites to create pCR3 . 1-GFP-X61 , pCR3 . 1-GFP-X41 , and pCR3 . 1-GFP-X61 ( 4A ) vectors . pCR3 . 1–2006 H1N1 PA-X-myc and pCR3 . 1–2009 TN H1N1 PA-X-myc were generated by subcloning PA-X ( including the 5’ UTR ) from A/Hong Kong/218847/2006 ( H1N1 ) and A/Tennessee/1-560/2009 ( pandemic H1N1 ) IAV segment 3 constructs that were a kind gift from R . Webby ( St Jude’s Children Research Hospital , Memphis , TN ) into the SalI and MluI sites of pCR3 . 1-myc . A single nucleotide was deleted to obtain the PA-X coding sequence using overlap extension . pd2-eGFP-N1 was purchased from Clontech . pd2-eGFP-N1-CMVd1 ( used in Fig 1A ) was generated by deleting 457 bp from the CMV IE promoter in pd2-eGFP-N1 , in order to reduce the constitutive levels of GFP mRNA and protein . GFP-3’SLII , GFP-codingSLII , GFP-HR , GFP-A60-HR , GFP-hisSL and hp-GFP constructs are based on pd2-eGFP-N1 and were previously described [14 , 29 , 45] . Pol I GFP , Pol III GFP , pCDEF3-SOX , pCAGGS-nsp1 , and pCDNA3 . 1-DsRedExpress-DR were previously described [8 , 14 , 29] . pTRIPZ-shNS and pTRIPZ-Xrn1 were purchased from Thermoscientific ( shXrn1: clone V2THS_89028/RHS4696-99704634 , targeting sequence: TATGGTGAGATATACTATG ) . pTRIPZ-PA-X-myc was generated by replacing the RFP sequence in pTRIPZ-shNS with the PA-X-myc coding sequence using the AgeI and ClaI restriction sites . pTRIPZ-RFP-SV40-3’UTR was generated by substituting the shRNA sequence downstream of the RFP coding sequence in pTRIPZ-shNS with the SV40 3’ UTR from pd2-eGFP-N1 . The RFP sequence was then replaced with PA-X-myc or PA-X ( D108A ) -myc to generate pTRIPZ-PA-X-myc-SV40-3’UTR and pTRIPZ-PA-X ( D108A ) -myc-SV40-3’UTR using the AgeI and ClaI restriction sites . All PA-X constructs also include the viral 5’ UTR of PR8 IAV segment 3 . The T7 polymerase and T7-driven EMCV-luciferase constructs [40] were a kind gift from E . Heldwein ( Tufts University School of Medicine , Boston , MA ) . They were co-transfected in 293T iPA-X cells to obtain luciferase transcription by T7 polymerase . Luciferase RNA and luciferase activity were only detected when the T7 polymerase expression construct was co-transfected , confirming that the luciferase RNA is solely transcribed by T7 . pTRE2-Fluc and pTRE2-Rluc vectors were previously described [60] . pGL4 . 32 and pGL4 . 74 reporter plasmids were purchased from Promega . Sequence information for the vectors generated in this study is available upon request . Mouse embryonic fibroblasts ( gift from Dr . Kedersha , Brigham and Women’s Hospital , Boston , MA ) , HeLa Tet-Off ( Clontech ) , HEK 293A ( ATCC ) , HEK 293T ( ATCC ) , A549 ( ATCC ) , and their derivative 293T and A549 “iPA-X” cells were maintained in Dulbecco’s modified Eagle’s medium ( DMEM; high glucose , Life Technologies ) supplemented with 10% fetal bovine serum ( FBS , Hyclone ) at 37°C in 5% CO2 atmosphere . HEK 293T iPA-X cells were generated by lentiviral transduction using pTRIPZ-PA-X-myc . A549 iPA-X were generated by lentiviral transduction using pTRIPZ-PA-X-myc-SV40-3’UTR or pTRIPZ-PA-X ( D108A ) -myc-SV40-3’UTR . Clonal populations were selected and several lines were used for all experiments . PA-X expression was induced by addition of doxycycline ( Fisher , 0 . 2–1 μg/ml final concentration depending on the cell line ) for approximately 18 h prior to harvesting , with the exception of half-life experiments , in which doxycycline was added for 4 . 5 h before actinomycin D addition . HEK 293T shNS ( also referred to as “iRFP” cells ) and shXrn1 cells [15] were generated using pTRIPZ-shNS ( Thermo scientific ) and pTRIPZ-shXrn1 ( Thermo scientific ) respectively . To induce expression of the shRNAs , cells were treated with 1 μg/ml doxycycline for 4–5 days prior to harvesting . For experiments using reporter constructs , 800 ng of DNA was transfected in 12-well plate wells using polyethylenimine ( Fisher ) . For the experiment in Fig 6E , FugeneHD ( Promega ) was used for transfection , in order to achieve high transfection efficiency ( estimated at > 75% based on GFP fluorescence ) . RNA and proteins were harvested as detailed below 1 day after transfection . For measuring the levels of the Firefly and Renilla luciferase reporter expression the Dual-Luciferase Reporter Assay Kit ( Promega ) was used according to manufacturer protocol . The A/PuertoRico/8/34/ ( H1N1 ) ( PR8 ) and the recombinant mutant PR8-PA ( fs ) viruses are described in [41] . Virus stocks used for experiments were produced and titrated by plaque assays as described [41] . Two independent recombinant virus rescues were performed as described in [61] for both wild-type PR8 and PR8-PA ( fs ) and used in experimental infection replicates . The genomic RNA segment 3 of all virus stocks was verified by sequencing . For each infection , after 1-hour inoculation with virus dilutions , cells were washed with PBS and cultured in infection medium ( 0 . 5% bovine serum albumin ( BSA ) in DMEM ) and incubated at 37°C in 5% CO2 atmosphere . Cells grown on glass coverslips were fixed and immunostained according to the protocol in [62] using mouse monoclonal antibody to PABP1 ( sc-32318 , Santa Cruz Biotechnology ) ; goat polyclonal antibody to influenza virus ( ab20841 , Abcam ) , or rabbit antibody to PA ( GeneTex-125932 ) at manufacturer-recommended dilutions . Nuclei were stained with Hoechst dye ( Invitrogen ) . AlexaFluor-conjugated secondary antibodies ( Molecular Probes ) were used at 1:1 , 000 dilution . Images were captured using Zeiss Axioplan II microscope . For steady-state measurements of RNA levels in cells ectopically expressing PA-X protein , RNA was harvested 1 day after transfection or 18 h after PA-X induction in iPA-X cells . For half-life measurements , PA-X ( wt or D108A ) was induced in A549 iPA-X cells for 4 . 5 h , followed by addition of actinomycin D at 10 μg/ml final concentration . RNA was collected 0 , 2 , 4 , 6 h after actinomycin D addition , as well as prior to doxycycline treatment . RNA for Northern blotting was harvested and extracted using Trizol reagent ( Life Technologies ) following manufacturer’s protocol . To isolate cytoplasmic vs . nuclear fractions , cells were lysed in 0 . 1% NP-40 in Dulbecco’s phosphate buffered saline ( DPBS ) and the nuclear pellet was spun down . The supernatant was collected as the cytoplasmic fraction . The pellet was washed in 0 . 1% NP-40 in DPBS , then it was collected as the nuclear fraction . The RNA was extracted from both fractions using Trizol ( Life Technologies ) following manufacturer’s protocol . The RNA for northern blotting was run on a 1 . 8% agarose/formaldehyde gel and transferred by capillary action onto nitrocellulose membrane ( Bio-Rad ) using 10x SSC buffer . Northern blots were probed with probes against the SV40 3’ UTR present in pd2-eGFP-N1 [29] , the first 450 nt of the GFP coding sequence , the endogenous 18S rRNA [45] , 7SL and U2 ncRNAs in Church buffer . Probes were generated and radiolabeled using the Decaprime II kit ( Life Technologies ) . Blots were imaged using a Fujifilm scanner FLA-9000 . Quantification of the blots was carried out using ImageJ [63] . Figures show images representative of multiple biological replicates . Total cellular RNA for RT-qPCR in iPA-X and transfected cells was harvested and isolated using Zymo mini-prep kit ( Zymo Research ) following manufacturer’s protocol . cDNA was generated using iScript ( Bio-Rad ) from DNase-treated RNA . For analyses of mRNA and vRNA levels in A549 cells infected with IAV , total RNA was isolated at 6 , 9 , 12 , and 15 h post-infection using Qiagen RNeasy Plus kit . For cDNA synthesis , Thermo Maxima H Minus kit was used with gene-specific primer for 18S rRNA combined with either oligo ( dT ) ( for mRNA targets ) or influenza vRNA-specific Uni12 primer [64] . To minimize primer-less reverse transcription , after primer annealing reactions were carried out at 65°C according to manufacturers’ protocol . Quantitative PCR analysis was performed using iTaq Universal SYBR Green Supermix ( Bio-Rad ) and Ct values were analyzed using the BioRad CFX Connect Real-Time System qPCR and Bio-Rad CFX Manager 3 . 1 program analysis . S1 Table lists primers used for qPCR . Pre-mRNA measurements were carried out with primer sets that have been previously used in the literature and that are located within predicted introns [65] . All qPCR experiments shown are the average of three or more biological replicates . Within each biological replicate , RNA levels were assessed using the average of at least two technical replicates . One- and two-sample Student’s t-test was used to analyze values for significant differences . Metabolic labeling and isolation of nascent RNA in A549 cells infected with wild-type PR8 or mutant PR8-PA ( fs ) viruses was performed using Molecular Probes Click-iT Nascent RNA Capture Kit . At 8 hpi , 0 . 4 mM Click-iT nucleotide analogue 5-ethynyl uridine ( EU ) was added to the infection media for 1 hr at 37°C . At 9 hpi monolayers were washed twice with PBS and total RNA was isolated as described using Qiagen RNeasy Plus kit . Biotinylation and subsequent purification of EU-labelled RNA was performed according to the kit manufacturer protocol . cDNA synthesis and qPCR was performed on streptavidin beads as directed by the kit manufacturer protocol and described in methods section above . Total cellular protein was collected in protein lysis buffer ( 10 mM Tris pH 8 , 150 mM NaCl , 1% Triton x100 , and cOmplete EDTA-free protease inhibitor cocktail ( Roche ) ) unless specified otherwise in the text . Proteins were separated on SDS-PAGE gels and transferred onto PVDF membranes ( Millipore ) . Western blots were performed in PBST with 5% milk or TBST with 4% bovine serum albumin . The following antibodies were used: Xrn1 ( 1:200 , Santa Cruz-16598 ) , IAV NS1 ( 1:1 , 000 , gift from Kevin Coombs [66] , clone 8C7 ) M2 ( 1:1 , 000 , Abcam ab5416 ) , GFP ( 1:500 , Santa Cruz-8334 ) , β tubulin ( 1:200 , Santa Cruz-9104 ) , actin ( 1:4 , 000 , HRP-conjugated , Cell Signalling #5125 ) , IAV PA ( GeneTex-125932 ) , IAV ( 1:2 , 000 , Abcam ab20841 , recognizes NP , M1 , and ( weakly ) HA proteins of PR8 strain ) . Secondary antibodies were purchased from Southern BioTech ( rabbit , mouse ) or Santa Cruz ( goat ) and used at 1:3 , 000 to 1:5 , 000 dilution . Human genes: β-actin: ACTB/60; β-tubulin: TUBB/ 203068; EEF1A: EEF1A1/1915; Histone cluster 1 H3C: HIST1H3C/8352; POLR2A: POLR2A/5430; MALAT1: MALAT1/378938; TP53TG1: TP53TG1/11257; GAPDH: GAPDH/2597; GUSB: GUSB/2990; RPS6: RPS6/6194; RPS18: RPS18/6222; 7SL: RN7SL1/6029; 7SK: RN7SK/125050; U2: RNU2-1/6066 . IAV genes: PR8 PA-X: PA-X/13229134 . | All viruses depend on host components to convert viral mRNAs into proteins . Several viruses , including influenza A virus , encode factors that trigger RNA destruction . The influenza A virus factor that serves in this capacity is known as PA-X . PA-X limits accumulation of host mRNAs and proteins in infected cells and suppresses host responses to infection , but to date its precise mechanism of action remains obscure . Here we report that PA-X selectively targets cellular mRNAs , while sparing viral mRNAs , thereby compromising host gene expression and ensuring priority access of viral mRNAs to the protein synthesis machinery . We demonstrate that complete degradation of mRNAs cut by PA-X is dependent on the host factor Xrn1 and that PA-X likely works in the cell’s nuclei . Interestingly , PA-X targeting appears to be selective for products of host RNA polymerase II , and canonical mRNA processing is required for cleavage . Even though viral mRNAs are spared from PA-X-mediated degradation , PA-X-deficient viruses displayed defects in the synthesis of certain viral mRNAs and decreased viral protein accumulation . Thus , PA-X-mediated host shutoff influences the efficiency of viral gene expression . These studies significantly advance our understanding of this important viral host shutoff protein and may provide future opportunities to limit the pathogenesis of influenza A virus infection . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"medicine",
"and",
"health",
"sciences",
"nucleic",
"acid",
"synthesis",
"luciferase",
"pathology",
"and",
"laboratory",
"medicine",
"293t",
"cells",
"pathogens",
"enzymes",
"messenger",
"rna",
"rna",
"extraction",
"microbiology",
"biological",
"cultures",
"enzymology",... | 2016 | Selective Degradation of Host RNA Polymerase II Transcripts by Influenza A Virus PA-X Host Shutoff Protein |
Identifying modes of species diversification is fundamental to our understanding of how biodiversity changes over evolutionary time . Diversification modes are captured in species phylogenies , but characterizing the landscape of diversification has been limited by the analytical tools available for directly comparing phylogenetic trees of groups of organisms . Here , we use a novel , non-parametric approach and 214 family-level phylogenies of vertebrates representing over 500 million years of evolution to identify major diversification modes , to characterize phylogenetic space , and to evaluate the bounds and central tendencies of species diversification . We identify five principal patterns of diversification to which all vertebrate families hold . These patterns , mapped onto multidimensional space , constitute a phylogenetic space with distinct properties . Firstly , phylogenetic space occupies only a portion of all possible tree space , showing family-level phylogenies to be constrained to a limited range of diversification patterns . Secondly , the geometry of phylogenetic space is delimited by quantifiable trade-offs in tree size and the heterogeneity and stem-to-tip distribution of branching events . These trade-offs are indicative of the instability of certain diversification patterns and effectively bound speciation rates ( for successful clades ) within upper and lower limits . Finally , both the constrained range and geometry of phylogenetic space are established by the differential effects of macroevolutionary processes on patterns of diversification . Given these properties , we show that the average path through phylogenetic space over evolutionary time traverses several diversification stages , each of which is defined by a different principal pattern of diversification and directed by a different macroevolutionary process . The identification of universal patterns and natural constraints to diversification provides a foundation for understanding the deep-time evolution of biodiversity .
The radiations and extinctions of species are recorded in the patterns of the tree of life [1 , 2] . Those patterns are fundamentally important to testing classical evolutionary hypotheses ( e . g . , the Red Queen [3] and adaptive radiation [4] ) , for understanding species distributions across clades [5] and regions [6] , and , more generally , for piecing together how life evolves on Earth . Accordingly , the development of models for understanding how diversification unfolds using phylogenies of extant taxa has exploded in the last decade [7] , and several meta-analyses have attempted to identify general principles of diversification [8–12] . Despite these recent and rapid developments , we do not yet have a coherent , panoramic view of the patterns of diversification across the tree of life . There is a general consensus that many phylogenies have a signal of slowdown in speciation rates through time , although this consensus is not universally accepted , and the processes behind the general pattern remain equivocal [8–13] . There is also broad agreement that diversification rates vary considerably across lineages [5 , 14] , although approaches for detecting where diversification shifts happen [5 , 15] and what determines these shifts continue to be deliberated and debated [16] . Although model-based approaches have been crucial for testing specific hypotheses , a thorough description of the diversification patterns seen in empirical phylogenetic trees requires approaches free of any a priori assumption of how those trees behave . We developed a novel , model-free framework rooted in spectral graph theory that allows for the direct comparison of phylogenetic trees across various groups of organisms ( see Materials and Methods ) [17] . In essence , we summarize the information contained in tree shape by a distribution—the spectral density—representing the spectra of eigenvalues of the modified graph Laplacian ( MGL ) , which is a matrix constructed from the evolutionary distances between phylogenetic nodes specially designed to retain maximal information from the tree . This allows us to measure the similarities and dissimilarities between trees from different species clades by calculating the dissimilarity between their respective spectral densities . Phylogenetic trees can then be clustered according to how similar their shapes are . In addition , spectral density profiles can be used to compute summary statistics . In particular , the ln-transformed principal eigenvalue ( λ* ) , skewness ( ψ ) , and peak height ( η ) are indicative of phylogenetic expansion , stem-to-tip distribution of branching events , and heterogeneity of branch lengths , respectively [17] . Phylogenetic expansion λ* is highly correlated with measures of phylogenetic diversity and ( less so ) to species richness; the stem-to-tip distribution ψ is loosely correlated with the classical gamma summary statistic; and the heterogeneity of branch lengths as measured by η is not correlated with any known measure of phylogenetic balance but instead is representative of regular branching events in the tree ( such that , for example , trees with localized radiations or diversification rate shifts will appear irregular ) . Importantly , trees constructed from different models of diversification cluster separately according to their spectral densities , and the three spectral density summary statistics are collectively sufficient to distinguish between trees governed by different diversification processes [17] . Thus , empirical trees can be grouped into clusters representing distinct diversification patterns and mapped with minimal information loss onto a three-dimensional space—which we call phylogenetic space—defined by λ* , ψ , and η .
We computed the spectral density profiles of 214 family-level trees representing mammals [18] , birds [19] , squamates [20] , amphibians [21] , and ray-finned fishes [22] ( see Materials and Methods ) , totaling 11 , 930 species . By clustering these profiles , we found that there are five principal ways in which all vertebrate families diversify , significantly supported by both hierarchical and k-medoids clustering ( Fig 1A and 1B , respectively ) . Iterating our analyses on samples from a Bayesian posterior distribution of trees showed that these results were robust to phylogenetic uncertainty ( Methods ) : we identified five clusters 100% of the time; in 75% of the iterations , the distribution of trees across clusters was identical; and in the iteration with maximum mismatch , only 6% of the trees were placed in different clusters . The five distinct clusters are characterized by both global-scale differences related to overall tree shape ( as measured by λ , η , and ψ ) and local-scale differences in branching patterns restricted to specific parts of the tree that are not encapsulated by the three main summary statistics but appear as smaller details in the shape of the spectral density profile ( Fig 1C and 1D ) . Diversification types fall on a positively correlated gradient of λ* and η , with ψ varying considerably and unsystematically between types . The most abundant type ( III ) is defined by an intermediate λ* and η and is distinguished from the second most abundant type ( I ) by a relatively low ψ . The third and fourth most abundant ( II and V ) are virtually antipodal , the former having extremely low λ* and η and the latter being principally defined by a remarkably high ψ . The least abundant type ( IV ) is distinguished by very high λ* and η . Notably , as estimated by the eigengap heuristic , phylogenies were most frequently unimodal ( 40% ) , with 20% having two or three modes and the rest having between four and thirteen ( S1 Data ) . No diversification type shows a prevalence for a number of modalities ( D < 0 . 25 , p > 0 . 1 ) . Despite variation in the abundance of types across families , we see no single diversification type that dominates the vertebrate landscape ( Fig 1E , S2 Data ) . We observe , however , certain diversification types to be over- or under-represented within each vertebrate class , even when accounting for sample size ( S1A Fig , S3 Data ) . Ray-finned fish deviate the least from an even distribution . Type III trees are absent in squamates and amphibians , but only significantly so in amphibians . This may be due , in part , to mean differences in crown age between classes ( S1B Fig ) . Mapping empirical phylogenies in the space defined by λ* , ψ , and η reduces the high dimensionality of phylogenies to only three dimensions , thus allowing the first visualization of vertebrate phylogenetic space ( Fig 2A ) . The geometry of phylogenetic space reveals that vertebrate phylogenies are able to occupy a broad , multidimensional range of tree space and therefore differ from one another along multiple axes of variation rather than fall along a single line . As expected , phylogenies from distinct diversification types cluster together in phylogenetic space and , accordingly , vertebrate classes are differently distributed in space ( S1C Fig ) . Phylogenies differing in the number of missing extant taxa are spread across phylogenetic space , suggesting that undersampling does not bias our representation of this space ( S2 Fig ) . The dimensionality of phylogenetic space can be further reduced to only two dimensions , as revealed by a polytope test that estimates the statistical significance of a given number of vertices ( i . e . , archetypes ) encompassing input data ( see Materials and Methods ) [23] . We identified three archetypes as the minimum number that could significantly delimit phylogenetic space ( p < 0 . 01 ) . These three archetypes present a Pareto-optimal situation: all phylogenies fall within the optimal polytope ( here , a triangle ) and a different phylogenetic property is maximal nearest each archetype ( i . e . , vertex of the triangle ) ( Fig 2B ) . Each archetype is variably contributed to by λ* , ψ , and η—archetype a ( 10% , 89% , 1% ) ; archetype b ( 44% , 6% , 49% ) ; archetype c ( 45% , 4% , 51% ) —such that , as a tree approaches an archetype , its properties change according to the proportional contributions of each property at that archetype . Direct trade-offs between λ* , ψ , and η reveal constraints on how certain phylogenetic properties covary ( Fig 3A ) . The positive linear relationship between λ* and η reveals the tendency for phylogenies to become more regular as they accumulate phylogenetic diversity , with regularity increasing at a slower rate than diversity . The negative logarithmic describing η as a function of ψ draws a steep slope around ψ = 0 , where a wide range of η is possible , but then abates , signifying the tendency for phylogenies ( with ψ > 0 ) to become increasingly irregular as they become more tippy . Taken together , the archetypes underline phylogenetic trade-offs in the movement through phylogenetic space: as trees move from archetype a to b , they become more stemmy , and as they move from b to c , they become more expansive and regular . The trade-offs we observe in phylogenetic space reveal biological rather than structural constraints . Comparing the three-dimensional empirical phylogenetic space to the entirety of the space available to phylogeny-type networks ( i . e . , bifurcating , ultrametric , and rooted trees restricted to the λ* range of vertebrate trees; see Materials and Methods ) shows that empirical phylogenies occupy only a small portion ( 36% ) of this possible space ( Fig 3B ) . In fact , any combination of high , low , and medium values of λ* , η , and ψ can be reached by simulated phylogeny-type networks . The polytope encompassing phylogeny-type networks ranges more extensively in high-η , low-ψ phylogenetic space , resulting in a symmetrical , Pareto-poor geometry ( S3 Fig ) . The difference between empirical and potential space is not linked to the restricted number of taxonomic families we analyzed in comparison with the total number of vertebrate families . Indeed , rarefaction analyses suggest that the sample of trees we analyze saturates empirical phylogenetic space , with coordinates of the two-dimensional polytope reaching an asymptote ( S4 Fig ) . Hence , empirical phylogenetic space is truly evolutionarily constrained , with the parts of space epitomizing trees that are regular ( high η ) and either particularly stemmy ( low ψ ) or particularly tippy ( high ψ ) not reached by diversification trajectories . To understand what types of trees , in diversification terms , occupy these empty spaces , we simulated ultrametric trees under varying diversification parameters . We found that much of the regular space could be occupied by the simulated trees , with the stemmy space occupied by trees with a declining speciation rate and the tippy one occupied by trees with an increasing speciation rate ( S5 Fig ) . There is overlap between the simulated and empirical trees , but most trees simulated with β = ±0 . 1 fall outside of vertebrate phylogenetic space . In order to interpret different regions of phylogenetic space in traditional macroevolutionary terms , we estimated the statistical support for state-of-the-art , processed-based phylogenetic diversification models that are comparable in a likelihood framework ( see Materials and Methods ) . The processes featured by these models were mass extinction events ( constant speciation and extinction punctuated by events when a fraction of species is lost [24] ) , diversity dependence ( diversification rates varying as a function of the number of extant species [25] ) , temperature dependence ( diversification rates varying as a function of temperature [26] ) , and speciation by genetic differentiation ( SGD ) ( speciation emerging from the accumulation of mutations [27] ) . Each diversification type is predominantly ( or exclusively ) supported by a different model . Specifically , 100% of type II phylogenies are best supported by an exponential diversity-dependent model; 100% of type V by a mass extinction model; 80% and 20% of type I by SGD and temperature dependence , respectively; phylogenies of type III and IV show support from all four diversification models , although they are disproportionately supported by SGD and temperature dependence , respectively ( Fig 3C , S4 Data ) . Linear diversity dependence and linear temperature dependence were never selected above their exponential counterparts . On average , corrected Akaike Information Criterion differences ( ΔAICc ) between models showing the best and second-best likelihood support was modest ( ΔAICc = 1 . 49±1 . 22; S6 Fig , S1 Data ) , which illustrates the well-recognized difficulty of unambiguously disentangling different diversification scenarios with model-based phylogenetic comparative methods [7] . Trees simulated under the same four diversification processes , plus a trait-dependent process [28] , and mapped onto phylogenetic space ( see Materials and Methods ) corroborate that certain processes coincide with certain diversification types , but that the matching between type and process is not perfect . Trees generated under diversity-dependent and mass extinction models are the most localized in phylogenetic space , whereas other models are more widespread ( S7 Fig ) . Nonetheless , the phylogenetic space of model-based trees largely corresponds ( ~85% overlap ) with empirical phylogenetic space , and different regions of phylogenetic space are disproportionately associated with different diversification processes . Accordingly , these processes have shaped the geometry of vertebrate phylogenetic space by constraining family-level vertebrate phylogenies to diversify in particular ways . We estimated the trajectory that , on average , families traverse through phylogenetic space over time by binning empirical phylogenies by crown age ( see Materials and Methods ) . The trajectory traverses four diversification types starting from type II ( the diversity-dependent type ) followed by type I , III , and finally V ( the mass extinction type ) ( Fig 4A ) . This is confirmed by a significant age dependence to each type ( S8A Fig ) , although there is considerable variation within each bin ( S8B Fig ) . The trajectory is characterized by an exponential increase in λ* and a linear increase in ψ ( Fig 4B ) , but no significant trend in η or , despite its relevance to λ* [17] , species richness ( p = 0 . 34 ) . However , when we look at taxonomic classes separately , we see that each takes a divergent average trajectory through phylogenetic space ( S9 Fig ) , which conforms with their different distributions in space ( S1B Fig ) and suggests phylogenetic constraints to how diversification evolves . The trajectory for birds traverses a path involving types II/III , I , and V that resembles that of the global trajectory , whereas the mammalian trajectory takes a different path , traversing types II , III , and IV . It is notable that phylogenies that predate the Cretaceous–Paleogene ( K–Pg ) boundary ( ~66 million years [my] ) , which here comprise only non-mammalian families , fall within the region of phylogenetic space associated with mass extinction , whereas the most ancient mammalian phylogenies ( <60 my ) fall within the region associated with temperature dependence .
Finding general patterns in the way that clades diversify has been notoriously difficult , limiting our ability to understand how biodiversity emerges in deep time . Mapping clades in phylogenetic space allows us to have a much clearer view of the landscape of diversification . Importantly , we find that there are constraints to diversification . Vertebrate phylogenetic space fills only a third of all possible tree space—akin to the constrained morphospace of Raup‘s mollusc shells [29]—due to ostensible limitations to how vertebrate families successfully diversify . Biodiversity might well follow , if not entirely predictable , at least constrained trajectories through evolutionary time . We find that there are five principal types of diversification , which are bound in phylogenetic space by three vertices characterized by: tippy trees; stemmy , irregular trees; and large , regular trees ( vertices a , b , and c from Fig 2A , which correspond to type V , II , and IV trees , respectively ) . The bird family Acanthizidae is an example of archetype a; the mammalian family Muridae is an example of archetype b; and the mammalian family Nycteridae is an example of archetype c , while the mammalian family Leporidae is not optimized at any archetype but rather at an intermediate position in phylogenetic space . The support for mass extinction events among tippy trees suggests that the tippy part of phylogenetic space ( high ψ ) is occupied by clades that radiated after mass extinction left niche space unoccupied . The support for diversity-dependent models among disproportionally irregular stemmy trees ( type II trees ) suggests that the slowdown in speciation is differentially distributed across lineages due , perhaps , to differences in species abundances . Finally , support for temperature-dependent speciation among regular trees ( type IV trees ) suggests that radiation events resulting from environmental factors may similarly and pervasively affect all lineages . As certain regions of phylogenetic space are dominated by particular biotic or abiotic processes , it is ultimately the combined effect of intrinsic and extrinsic limiting factors on each family that shapes the geometry of phylogenetic space . The three vertices bounding phylogenetic space are the result of trade-offs between various characteristics of the trees . There is a long history of work showing how phenotypic evolution is constrained in every instance: there are always trade-offs between optimizing different fitness strategies [30 , 31] . As a result , species adapt to one fitness optimum at the expense of others , or to a concessionary optimum that serves several while optimizing none [32] . While phylogenetic trees are not adaptive in the sense that phenotypes are , there are trade-offs and combinations of tree characteristics that are not seen in nature , suggesting that certain diversification patterns are unstable ( i . e . , they lead to clade-wide extinction ) . Trees simulated with a speciation rate > α * e±0 . 1 , for example , fall overwhelmingly outside of vertebrate phylogenetic space . Therefore , successful diversification in vertebrate families is approximately bounded by these speciation rates , and the reason we see trade-offs in phylogenetic properties and a constrained phylogenetic space is because families that surpass those bounds go extinct . Consequently , we may expect phylogenies of extinct clades to violate these rules and fall outside of phylogenetic space . Specifically , there are trade-offs between both stemminess and expansiveness with regularity among vertebrate families . Trees become more irregular as they become tippier . If tippyness is indeed the mark of mass extinction events , this trade-off could be the result of ( mass ) extinction unevenly culling parts of trees . We also observe a lower boundary for stemminess , which suggests that clades that diversify only early in their history go extinct . Finally , trees that are expansive tend to be more regular , suggesting that the various processes generating irregularity ( e . g . , runaway diversification in one lineage versus others , multiple slowdowns in diversification ) result in contracted clades or widespread extinction . As a consequence of these phylogenetic trade-offs , there are parts of phylogenetic space unoccupied by vertebrate families . Those parts are predominantly characterized by both stemmy and tippy regular trees and can be occupied by trees simulated with considerably decreasing and increasing rates of speciation , respectively . The fact that vertebrate family phylogenies are wedged between these two extremes of speciation provides the first evidence for the range of speciation possible at this level of vertebrate evolution . Such limitations to species diversification must be the consequence of a set of limiting factors such as minimum generation time , responsiveness to abiotic events , spatiotemporal availability of niche space , or some other facets of vertebrate speciation that favor certain diversification patterns while prohibiting others . There is a significant positive correlation between crown age and both expansiveness and tippyness . The former correlation is unsurprising and suggests , firstly , that expansiveness is not bounded by any kind of upper ecological limit and , secondly , that branch length accumulates at a more or less similar rate across families , such that older families are typically more expansive than younger ones simply as a result of their having had more time to expand . The observation that tippyness also increases as a function of crown age is somewhat more surprising , although it is consistent with the observation that young clades tend to carry a stronger signal of early burst radiations ( captured by stemminess ) than older , more inclusive ones [9 , 19] . This suggests that young clades show a signal of early burst radiations , with speciation rates declining as ecological niches are filled , but also that this limitation in niche space is not absolute and radiations can happen later in a clade‘s history . As clades age , new opportunities for radiations may arise if , for example , ecological opportunity is recovered after a mass extinction event , or if new factors allow the exploitation of a part of ecological space that was not previously explored ( e . g . , new interspecific interactions prompting character displacement , fluctuations in temperature producing new niches under favorable environmental conditions for speciation ) . The distinct distributions in and evolutionary trajectories through phylogenetic space of birds and mammals may reveal something about evolutionary differences between classes . Bird families are distributed nearly homogeneously in phylogenetic space and their average trajectory is similar to that of squamate , amphibian , and fish families . Mammals , on the other hand , are disproportionately underrepresented at archetype a and show an average trajectory through phylogenetic space orthogonal to those of other vertebrates . The relative preponderance of mammal families in diversification type IV is consistent with the metabolic theory of ecology [33] as well as work suggesting that mammalian diversification is inversely correlated with temperature [34] . We can speculate that the evolutionary novelty swaying mammalian biodiversity towards temperature dependency is connected to endothermic viviparity , which has led to considerable reproductive diversity in mammals [35] and has been shown to directly impact diversification in other vertebrates [36] . It is also worth noting that the clear age dependence of diversification types may bias the distribution of mammal families in phylogenetic space and , therefore , their relative absence near archetype a may be explained , at least in part , by their median youngness . The benefit of generating an absolute phylogenetic space for species trees is that it provides a reference frame for describing and contextualizing phylogenies generally . It furthermore advances the idea that phylogenies are not simply the result of a single biotic or abiotic factor , but that a time-sensitive series of factors differentially influences diversification during a clade’s evolutionary history . By mapping non-vertebrate taxa in this space , it will be interesting to see whether the constraints observed here are unique to vertebrates or extend to other animal or even all organismal life .
We compiled a dataset of dated , family-level ultrametric trees from an exhaustive literature search of species-level vertebrate phylogenies . In order to avoid biased comparisons between trees , we included only phylogenies that were ≥80% sampled , which is shown to be a sufficient sampling estimate to make use of the MGL [17] . We restricted our analyses to family-level trees because , after discarding insufficiently sampled trees , this provided us with the largest number of trees of adequate size ( >20 species ) while allowing us to cover all vertebrate classes . We used the most recent molecular mammalian tree with 4 , 160 extant species ( v . 1001 ) [18] . We used a molecular bird tree with 6 , 670 extant species constructed on the Hackett backbone [19] . To avoid known ( and unknown ) polytomies and negative branch-lengths in the mammal and bird phylogenies , we computed maximum clade credibility ( MCC ) trees in BEAST 2 [37] that selected trees with the highest posterior probability product . We used a phylogeny of 4 , 161 extant squamate species [20] and one of 3 , 309 extant amphibian species [21] , both of which were constructed with a supermatrix analysis of molecular data . Finally , we used a phylogeny of 7 , 822 extant actinopterygians ( ray-finned fishes ) , which was constructed with a maximum-likelihood approach and made ultrametric using penalized likelihood [22] . Our compilation resulted in 72 Mammalia , 102 Aves , 7 Squamata , 9 Amphibia , and 24 Actinopterygii family-level phylogenies for a total of 214 trees . To assess phylogenetic uncertainty , we randomly sampled 100 trees from posterior distributions of 1 , 000 mammal trees [18] and 10 , 000 bird trees ( birdtree . org ) and from each sample parsed family-level trees . Following the non-parametric approach described in detail in [17] , given a phylogenetic tree , we defined its standard MGL as the difference between its degree matrix ( the diagonal matrix where diagonal element i is the sum of the branch lengths from node i to all the other nodes in the phylogeny ) and its distance matrix ( where element ( i , j ) is the branch length between nodes i and j ) . We then obtained the spectral density profile of the tree by convolving the spectra of eigenvalues of the MGL with a Gaussian kernel . Finally , we measured the distance between phylogenies using the Jensen–Shannon index [38] and subjected these distances to unsupervised hierarchical and k-medoids clustering . For hierarchical clustering , significance was calculated with bootstrap resampling and clusters were considered significant at α ≥ 0 . 95 . For k-medoids clustering , the number of clusters was estimated by optimum average silhouette width . Average silhouette widths greater than 0 . 5 were considered significantly supported . These analyses are implemented in the R package RPANDA [39] . To assess the effect of phylogenetic uncertainty on clustering results , we repeated both clustering protocols over 100 iterations by randomly drawing from the mammal and bird trees sampled from posterior distributions . We tallied the number of times we duplicated the original result using the MCC trees versus those where at least one tree was placed in a different cluster . When there were mismatches , we measured the percentage of trees that were placed in a different cluster . We determined whether the distribution of trees across clusters was more or less than expected by chance for each taxonomic class . Given a specific taxonomic class represented by x trees , we iteratively distributed x hypothetical trees in the five clusters , where the probability of being placed in a particular cluster was defined by the relative size of that cluster ( as calculated above ) . For each cluster , 500 iterations of this process yielded a null distribution of the expected number of trees in the cluster . We deemed significant the deviation of the actual number of families assigned to each cluster from the expected if the actual number fell in the lower or upper tail of the distribution ( α < 0 . 05 ) . We computed the spectral density profile summary statistics ( λ* , ψ , and η ) of phylogenies as described in [17] ( S1 Data ) . Briefly , each eigenvalue computed from the MGL of a phylogeny describes the distance between nodes , such that larger and smaller eigenvalues describe greater and shorter distances , which in practice represent speciation-poor and -rich regions of the tree , respectively . The largest distance , therefore , is denoted by the largest eigenvalue , λ* . When the spectrum of eigenvalues is convolved with a Gaussian kernel , it is possible to identify patterns in its distribution . The skewness , ψ , of the distribution indicates whether the eigenvalues of the MGL are preponderantly large or small and thus whether the phylogeny is comprised of mostly speciation-rich or -poor regions . Likewise , the peak height , η , demonstrates whether branch lengths in the phylogeny are heterogeneous ( low η ) or homogeneous ( high η ) —that is , irregular or regular , respectively . These statistics were used to map empirical phylogenies into a three-dimensional space . Spectral density profile summary statistics were regressed against each other and best-fit slopes were selected using a stepwise AIC . Modality in the MGL , which was not used to define phylogenetic space , was calculated using the eigengap heuristic ( S1 Data ) [17] , which determines the number of modes of division in a phylogeny by the position of the greatest difference between ranked eigenvalues . We determined the geometry of the space by finding the polytope with the smallest number of vertices ( i . e . , archetypes ) that could explain the most variance ( or beyond which little explanatory power was added ) in the distribution of phylogenies [32] . Archetypes were positioned around the data using the SISAL algorithm and allowing for up to six archetypes . Best fits at p < 0 . 01 were determined by t-ratio tests . The phylogenetic features optimized at each archetype were determined by the relative contributions of each summary statistic at that archetype [23] . To assess how sufficiently our sample of phylogenies captured the entirety of vertebrate phylogenetic space , we performed rarefaction analyses . We calculated the maximum coordinates of the archetypes for bootstrapped samples of phylogenies while fixing the number of dimensions at three and fit saturation curves for those coordinates as a function of sample size . Bootstrapped samples were replicated 50 times . To assess the entirety of phylogenetic space available to trees consistent with the empirical distribution of λ* , we extensively generated phylogeny-type networks ( i . e . , bifurcating , ultrametric , rooted networks ) while constraining λ* to be in the empirical distribution . Given a λ* from the empirical distribution , we constructed a graph Laplacian by populating a symmetric , positive , semidefinite M-matrix with zero-sum rows and columns . We computed a distance matrix from the graph Laplacian using the classical Dijkstra algorithm , which finds distances between nodes in a graph given an adjacency matrix [40] . We obtained the corresponding non-ultrametric tree from the distance matrix using complete linkage clustering [41] , and the resulting tree was made ultrametric using mean path length [42] . Finally , the tree was rooted at the stem and , where necessary , polytomies were resolved using the R package ape [43] . We constructed 10 , 000 such trees . The proportion of phylogenetic space available to trees that is occupied by empirical phylogenies was calculated as the volume of three-dimensional space occupied between the most distal empirical points along each axis as a percentage of the total space occupied by the simulated trees . The geometry of tree-type space was determined with a polytope analysis as above , using 1 , 000 trees randomly sampled from the 10 , 000 for computing purposes . To interpret phylogenetic space in terms of statistical diversification processes , we simulated ultrametric birth–death trees using the R package TESS [44] . We simulated 500 trees at ages 10–60 my under constant speciation rates ( b = 0 . 1–0 . 25 ) , decreasing speciation rates ( α * e−βt for α = 0 . 01–−1 and β = 0 . 05–−0 . 2 , and increasing speciation rates ( α * eβt for α = 0 . 01–−0 . 05 and β = 0 . 05–−0 . 2 ) ; the extinction rate was held constant ( d = 0–0 . 1 ) . We discarded trees outside the range of species richness for empirical phylogenies ( 20–700 ) . The final set of trees for each birth–death model had a comparable mean species richness to the empirical set . We then computed their spectral density profiles and plotted them in phylogenetic space as above . We selected phylogenetic diversification models meant to directly model a process ( rather than a temporal trend ) and that were comparable in a likelihood framework . This resulted in five models featuring mass extinction events , linear and exponential diversity dependence of speciation rates , exponential temperature dependence , and SGD . The mass extinction model was fitted using the |bd . shifts . optim| function in the R package TreePar , which maximizes the likelihood of a model with constant speciation and extinction rates and one or more sampling events ( i . e . , mass extinctions ) at discrete time points t1 , t2 , … , tn in the history of the clade [24] . We allowed the mass extinction events to occur at any time . Diversity-dependent models were fitted using the dd_ML function in DDD [25] . We tested models with speciation rates showing both linear and exponential dependence to the number of extant species . Temperature-dependent models were fitted with the fit_env function in RPANDA with constant , linear , and exponential speciation—and extinction-rate dependence to temperature [26] . Temperature data were inferred from benthic foraminiferal δ18O measurements extending to 108 my ago [45] at 0 . 1 my time-steps . Finally , the SGD model was fit using the fit_sgd function in RPANDA with Nelder–Mead optimization [27] . We fitted these models to all empirical phylogenies with ≥50 species ( a total of 62 phylogenies ) . Likelihood values were normalized after [46] in order to make the likelihood estimates directly comparable . Models with the lowest AICc scores were considered the most supportive of the phylogeny [47] . Trait-dependent models were not considered , as their likelihood functions are not , in their current form , comparable to the likelihoods of the other models . We simulated trees under five diversification models—birth–death with mass extinction , exponential diversity dependence ( which was always favored to linear dependence ) , exponential temperature dependence , SGD , and continuous trait dependence [28]—and mapped those trees in phylogenetic space . In order to restrict the parameter space to realistic values , we first estimated parameters by fitting models to the 62 empirical phylogenies with ≥50 species as above . We fitted the continuous trait-dependent model with ln-transformed body mass ( or , for ray-finned fish , maximum length ) as trait data using QuaSSE in diversitree [48] . Data were collected from the literature for mammals [49] , birds [50] , and ray-finned fish [51] . For each empirical parameter estimate , 30 trees were simulated , resulting in 1 , 500 simulated trees per diversification model . We calculated the distribution of simulated trees in phylogenetic space by drawing polygons around each empirical diversification type and counting the number of trees simulated under a diversification model falling in each polygon . Polygon vertices were selected using the most distal points in each type . The volume of phylogenetic space for simulated trees was estimated as above and the percentage correspondence of space between empirical and simulated trees was calculated as the volume of overlapping space over the total volume occupied by both empirical and simulated trees . We binned empirical phylogenies by crown age using a natural breaks optimization algorithm [52] without specifying a number of desired bins . For each bin , we calculated the arithmetic mean for λ* , ψ , and η and then plotted those values in empirical phylogenetic space . The diversification type corresponding to each bin was determined by drawing polygons around each diversification type as above . Median phylogenies ( Fig 4C ) were defined as those falling nearest the arithmetic means for λ* , ψ , and η . The same was done for mammals and birds separately . There were too few trees in the other taxonomic classes to analyze them similarly . | Are there universal laws in the evolution of biodiversity ? Why do some clades go extinct and others flourish ? These questions are fundamental to our understanding of present-day biodiversity . In a meta-analysis of nearly 12 , 000 species spanning ~500 million years of evolution , we find that there are five principal patterns of diversification to which all vertebrate families hold , and that these patterns can be mapped into a multidimensional phylogenetic space . Importantly , because certain diversification patterns invariably lead to extinction , clades do not explore all possible phylogenetic space , and thus the evolution of biodiversity is constrained by a set of loose but inviolable rules . We characterize the biotic and abiotic factors precipitating those rules with important implications for our knowledge of the emergence and maintenance of the diversity of life around us . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"taxonomy",
"ecology",
"and",
"environmental",
"sciences",
"vertebrates",
"animals",
"mammals",
"animal",
"phylogenetics",
"phylogenetics",
"data",
"management",
"phylogenetic",
"analysis",
"speciation",
"paleontology",
"molecular",
"biology",
"techniques",
"paleogenetics",
... | 2016 | Natural Constraints to Species Diversification |
Progress in dengue vaccine development has been hampered by limited understanding of protective immunity against dengue virus infection . Conventional neutralizing antibody titration assays that use FcγR-negative cells do not consider possible infection-enhancement activity . We reasoned that as FcγR-expressing cells are the major target cells of dengue virus , neutralizing antibody titration assays using FcγR-expressing cells that determine the sum of neutralizing and infection-enhancing activity , may better reflect the biological properties of antibodies in vivo . We evaluated serum samples from 80 residents of a dengue endemic country , Malaysia , for neutralizing activity , and infection-enhancing activity at 1∶10 serum dilution by using FcγR-negative BHK cells and FcγR-expressing BHK cells . The serum samples consisted of a panel of patients with acute DENV infection ( 31% , 25/80 ) and a panel of donors without acute DENV infection ( 69% , 55/80 ) . A high proportion of the tested serum samples ( 75% , 60/80 ) demonstrated DENV neutralizing activity ( PRNT50≥10 ) and infection-enhancing activity . Eleven of 18 serum samples from patients with acute secondary DENV infection demonstrated neutralizing activity to the infecting serotype determined by using FcγR-negative BHK cells ( PRNT50≥10 ) , but not when determined by using FcγR-expressing cells . Human serum samples with low neutralizing activity determined by using FcγR-negative cells showed DENV infection-enhancing activity using FcγR-expressing cells , whereas those with high neutralizing activity determined by using FcγR-negative cells demonstrate low or no infection-enhancing activity using FcγR-expressing cells . The results suggest an inverse relationship between neutralizing antibody titer and infection-enhancing activity , and that neutralizing activity determined by using FcγR-expressing cells , and not the activity determined by using FcγR-negative cells , may better reflect protection to DENV infection in vivo .
Dengue fever ( DF ) and dengue hemorrhagic fever ( DHF ) is caused by infection with dengue virus ( DENV ) , a flavivirus , which consists of four serotypes ( DENV-1 , DENV-2 , DENV-3 and DENV-4 ) . DENV affects up to 100 million people annually living in the tropics and sub-tropical areas . Clinical manifestations of DENV infection ranges from asymptomatic and relatively mild dengue fever ( DF ) , to severe , life-threatening illness , dengue hemorrhagic fever ( DHF ) and dengue shock syndrome ( DSS ) [1] , [2] . In endemic regions , the risk for developing severe infection was speculated to be higher as compared to non-endemic regions due to the higher possibility of secondary exposure to heterologous DENV serotypes [3] , [4] . The number of dengue patients has increased in Malaysia over the past 10 years with 7 , 103 cases and 45 deaths in 2000 , to 41 , 486 cases and 88 deaths in 2009 , to , 46 , 171 cases and 134 deaths in 2010 [5] , [6] . All four DENV serotypes co-circulate in Malaysia [7] , [8] . High prevalence of severe dengue virus infections and dengue-related deaths in recent years is speculated to be associated to rapid urbanization and global travel , leading to the spread of dengue virus , and thus to higher prevalence of infected individuals [9]–[11] . Primary infection with one DENV serotype does not confer protection to infection with a heterologous serotype [12] , [13] . Epidemiological studies have demonstrated that DHF occurs at a higher rate in secondary infection than in primary infection [14]–[17] . DENV sub-neutralizing , infection-enhancing antibodies induced during primary infection is speculated to play a central role in the pathogenesis of DHF [18]–[21] . During secondary infection , sub-neutralizing antibodies form infectious immune-complexes with DENV , resulting in higher levels of viral progeny in FcγR-expressing cells , a phenomenon known as antibody-dependent enhancement ( ADE ) [22] , [23] . It has been speculated that ADE may play a role not only in causing DHF but in worsening a spectrum of DENV illness [24] . We previously demonstrated that higher neutralizing antibody titers were detected using FcγR-negative BHK cells as compared to FcγR-expressing BHK cells [25] . In the present study , we examined DENV infection-enhancing activity in serum samples with varying levels of DENV neutralizing activity using FcγR-expressing BHK cells .
Eighty serum samples obtained from 80 residents in Perak , Malaysia were used in the study . Perak is located in north-western region of Peninsular Malaysia , and is endemic for dengue , and other flavivirus infections [26]–[28] . Incidence of DENV infection in Perak was 2 , 288 in 2010 , 2 , 734 in 2009 , and , 4 , 119 in 2008 [6] . The serum samples were collected in 2008 and were provided by National Public Health Laboratory , Malaysia . Characteristics of the patient population sampled are summarized in Table 1 . The study protocol was approved by the ethics committee of the National Institute Infectious Diseases , Japan ( Reference no . 210 ) . Patients were de-identified and study data was analyzed anonymously . Patient serum samples used in the present study were examined for the presence of dengue viral RNA by reverse-transcriptase polymerase chain reaction ( RT-PCR ) , and NS1 antigen by enzyme-linked immunosorbant assay ( ELISA ) . Viral RNA was extracted using High Pure RNA extraction kit ( Roche Diagnostics , Germany ) and DENV serotypes were determined by serotype-specific reverse transcriptase polymerase chain reaction ( RT-PCR ) [29] . For serological tests , serum samples were heat-inactivated at 56°C for 30 minutes before use . Detection of the NS1 antigen was performed using Platelia Dengue NS1 Antigen ( Bio-Rad Laboratories , France ) according to manufacturers' instructions . After reaction was terminated , optical density readings ( OD ) were obtained with a spectrophotometer at wavelengths of 450 nm/620 nm and the index of each sample was calculated with the following formula: OD of samples/OD of calibrators . The index value of each sample was interpreted according to manufacturer's instructions; index values of <0 . 9 , 0 . 9–1 . 1 , and , >1 . 1 were considered negative , equivocal , and positive , respectively [30] . Dengue virus type 1 ( DENV-1 ) , 01-44-1HuNIID strain ( GenBank accession no . AB111070 ) , dengue virus type 2 ( DENV-2 ) D2/Hu/OPD030NIID/2005 strain ( GenBank accession no . AB219135 ) , dengue virus type 3 ( DENV-3 ) CH53489 strain ( GenBank accession no . DQ863638 ) , and , dengue virus type 4 ( DENV-4 ) TVP360 strain were used . BHK cells , a hamster kidney cell line ( Japan Health Science Research Resources Bank , Japan ) and FcγR-expressing BHK cells were used . Cells were cultured in Eagle's Minimum Essential Medium ( EMEM ) ( Sigma , USA ) , supplemented with heat inactivated 10% fetal calf serum ( FCS , Sigma ) at 37°C in 5% CO2 . FcγR-expressing BHK cells were cultured in EMEM ( Sigma ) , supplemented with heat inactivated 10% FCS ( Sigma ) and 0 . 5 mg/ml neomycin ( G418 , PAA Laboratories GmbH , Austria ) at 37°C in 5% CO2 [25] . Heat-inactivated human serum samples were serially diluted 2-folds from 1∶5 to 1∶1250 with EMEM/2% FCS . As the amount of serum samples was limited , two replicates were tested for each of the serum samples to four dengue serotype . Virus-antibody mixture was prepared by mixing 25 µl of DENV-1 , DENV-2 , DENV-3 , or , DENV-4 at titers of 2000 PFU/ml with 25 µl of serum samples serially diluted 2 folds from 1∶5 to 1∶1250 . For infection with DENV alone , the mixture was prepared by mixing 25 µl of DENV-1 , DENV-2 , DENV-3 , or , DENV-4 strains at titers of 2000 PFU/ml with 25 µl of EMEM supplemented with 10% FCS . After incubation at 37°C for 60 minutes , 50 µl of virus-antibody mixture was inoculated on FcγR-negative BHK and FcγR-expressing BHK monolayers in 12-well plates . The plates were then incubated for 60 minutes at 37°C in 5% CO2 . After virus absorption , the cells were washed once with 1 ml of EMEM ( Sigma ) and overlaid with 1 ml EMEM ( Nissui Pharmaceutical , Japan ) containing 2% FCS ( Sigma ) and 1% methylcellulose ( Wako Pure Chemical Industries , Japan ) . The plates were incubated at 37°C in 5% CO2 for 5–7 days , when plaque formation could be confirmed by naked eye . Cells were then fixed with 10% formalin ( Wako Pure Chemical Industries ) and stained with methylene blue ( Wako Pure Chemical Industries ) . Number of plaques was counted with naked eye . Plaque-reduction neutralizing test ( PRNT50 ) end points are expressed as the last serum dilution showing a 50% or greater reduction in plaque counts as compared to the number of plaques determined from wells of cells infected in the absence of antibodies [31] . For enhancement assay against 4 DENV serotypes , 25 µl serum samples diluted at 1∶5 with EMEM/2% FCS were used . Virus-antibody mixture was prepared by mixing 25 µl of DENV-1 , DENV-2 , DENV-3 , or , DENV-4 strains at titers of 1000–2000 PFU/ml with 25 µl of 1∶5 diluted serum samples and incubated at 37°C for 60 minutes . For negative controls , virus mixture was prepared by mixing 25 µl of DENV-1 , DENV-2 , DENV-3 , or , DENV-4 strains at titers of 1000–2000 PFU/ml with 25 µl of 2% FCS/EMEM and incubated at 37°C for 60 minutes . Fifty microliters of the virus-antibody mixture was then applied to FcγR-expressing BHK monolayers in 12-well plates . The plates were then incubated for 60 minutes at 37°C in 5% CO2 . After virus absorption , the cells were washed with 1 ml of EMEM and overlaid with 1 ml EMEM containing 2% FCS and 1% methylcellulose . The plates were incubated at 37°C in 5% CO2 for 5 days . Cells were then fixed with 10% formalin and stained with methylene blue . Plaques were counted with naked eye . Fold enhancement values were determined using the following ratio: ( mean plaque count at 1∶10 serum dilution ) / ( mean plaque count in the absence of human serum samples , negative control ) . The sum of the mean of the negative control plus two times of the standard deviation ( SD ) value obtained from 4 wells of negative control was used as cut-off to differentiate enhancing and non-enhancing activity [32] , [33] . Enhancing activity was defined as positive when values are greater than the mean plaque count in the absence of human serum samples plus a greater than 2 times SD .
Eighty serum samples were tested for the presence of neutralizing antibody to each of the four DENV serotypes by using BHK cells ( Table 1 ) . Neutralizing antibody ( PRNT50≥10 ) was detected in 69% ( 55/80 ) to DENV-1 , 60% ( 48/80 ) to DENV-2 , 56% ( 45/80 ) to DENV-3 and , 20% ( 16/80 ) to DENV-4 . Of the 80 samples , 12 samples and 13 samples were obtained from patients with acute DENV-1 and DENV-3 infections , respectively ( Table 1 ) . Neutralizing activity to DENV-1 was detected in 58% ( 7/12 ) of the DENV-1 infected patients and that to DENV-3 was detected in 62% ( 8/13 ) of the DENV-3 infected patients . In contrast , enhancing-activity to DENV-1 was detected in 33% ( 4/12 ) of the DENV-1 infected patients and that to DENV-3 was detected in 61% ( 8/13 ) of the DENV-3 patients ( Table 2 ) . Five DENV-1 infected patients and two DENV-3 infected patients demonstrated neither neutralizing nor infection-enhancing activity to any of the four DENV serotypes ( Table S4 ) . Four ( 58% ) of the 7 serum samples from acute secondary DENV-1 infected patients demonstrated infection-enhancing activities to DENV-1 serotype and similarly , 8 ( 73% ) of 11 serum samples from acute secondary DENV-3 infected patients enhanced DENV-3 infection . The results indicate that serum samples from dengue patients and non-dengue patients in dengue endemic region , Malaysia , possess neutralizing and infection-enhancing activities ( Table S1 , Table S2 ) . The history of past dengue infections of each of the patients was however , not known . Because Malaysia is endemic for dengue infection , and some of the serum samples exhibited high levels of neutralizing antibody titers to DENV , it is highly likely that the patients with high neutralizing antibody activity has been previously exposed to DENV infection . The 18 samples derived from acute secondary dengue patients which demonstrated both neutralizing and infection-enhancing activities were examined for neutralizing antibody titers to each of the 4 DENV serotypes . The serum samples were from 7 acute secondary DENV-1 patients and 11 acute secondary DENV-3 patients ( Table 1 ) . Using FcγR-negative BHK cells , 7 serum samples possessed neutralizing antibody titers only to DENV-2 and 11 samples showed neutralizing antibody to more than one serotype ( Table 3 , Table S3 ) . All the serum samples except one ( 17 of 18 ) demonstrated neutralizing antibody titers to DENV-2 at titers of 1∶10 to 1∶1280 . Using FcγR-positive BHK cells , 13 serum samples obtained from patients with DENV-1 or DENV-3 infection possessed neutralizing antibody titers only to DENV-2 ( 1∶10 to 1∶160 ) , and 2 samples possessed neutralizing antibody titers only to DENV-1 ( 1∶10 to 1∶20 ) . Some of the non acute dengue patients ( serum sample no . 23 , 28 , 30 , 38 , 73 , 74 , 77 , 78 , 79 , Table 4 ) exhibited heterotypic neutralizing activity using BHK cells but monotypic neutralizing activity using FcγR-expressing cells , as those observed in patients with acute secondary dengue infection ( Table 3 ) . Interestingly , serum samples from patients with acute secondary infection ( #46 , 49 , 56 , 40 , 42 , 43 , 44 , 45 , 52 , and 54 ) demonstrated neutralizing antibody titers of 1∶10–1∶80 to the infecting serotype as determined using BHK cells , but these serum samples showed no neutralizing antibody titers to the infecting serotype as determined using FcγR-expressing BHK cells ( Table 3 ) . The results indicate that neutralizing antibody titers determined by using FcγR-expressing BHK cells were lower than those determined by using FcγR-negative BHK cells , and that for some of the samples , neutralizing antibody titers were detected only by using FcγR-negative BHK cells , but not by using FcγR-expressing BHK cells . Major target cells of DENV in vivo are FcγR-expressing cells such as monocyte-lineage cells [34]–[37] . The results suggest that the titers determined by using FcγR-expressing BHK cells may , thus , reflect actual biological activities of antibodies in vivo . Fold-enhancement to DENV-1 ranged from 0 . 7–5 . 6; DENV-2 , <0 . 1–2 . 1 , DENV-3 , 0 . 9–4 . 9 , and DENV-4 , 1 . 1–7 . 0 with serum sample from patients with infecting serotype of DENV-1 . Fold-enhancement to DENV-1 ranged from <0 . 1–5 . 5; DENV-2 , <0 . 1–2 . 3; DENV-3 , 0 . 6–6 . 7 , and DENV-4 , 0 . 9–7 . 1 with serum samples from patients with infecting serotype of DENV-3 ( Table 4 , Table S3 ) . Some of the samples from non acute dengue patients also exhibited enhancing activity to DENV ( Table S1 ) . The results indicate that sera from patients with DENV infection possess the ability to enhance the infection by the infecting serotype . The ability of serum samples with DENV neutralizing activity to enhance each of the 4 DENV serotypes at 1∶10 serum dilution was analyzed ( Figure 1 ) . Serum samples with neutralizing titers of ≥1∶10 determined by using BHK cells demonstrated lower levels of fold-enhancement to the homotypic serotypes . Fold enhancement to DENV-1 of serum samples with DENV-1 neutralizing antibody ( NA ) titer of ≥1∶10 was 0 . 8±0 . 9 versus DENV-1 NA titer <1∶10 = 4 . 5±1 . 6 ( P<0 . 01 ) . Fold infection-enhancement to DENV-2 using serum samples with DENV-2 neutralizing antibody ( NA ) titers of ≥1∶10 was 0 . 6±0 . 9 , while that of samples with DENV-2 NA titers <1∶10 was 1 . 9±1 . 1 ( P<0 . 01 ) . Fold enhancement to DENV-3 of serum samples with DENV-3 neutralizing antibody ( NA ) titer of ≥1∶10 was 0 . 9±1 . 1 , while DENV-3 NA titers <1∶10 = 3 . 9±2 . 1 ( P<0 . 01 ) . Fold enhancement to DENV-4 of serum samples with DENV-4 neutralizing antibody ( NA ) titer of ≥1∶10 was 3 . 6±2 . 0 , while that of samples with DENV-4 NA titers <1∶10 was 4 . 9±1 . 9 ( P = 0 . 02 ) ( Figure 1 ) . Serum samples with high levels of neutralizing activity to DENV-2 ( 40 , 42 , 44 , 49 , 54 and 58 ) exhibited peak fold enhancement ranging from 5 . 8–7 . 6 at higher serum dilutions of 1∶100–1∶1000 ( Figure 2 ) . Of the 55 serum samples from non acute dengue patients , 42 exhibited infection-enhancement activity to DENV ( Table S1 ) . Enhancement activities against DENV-1 and DENV-3 using serum samples obtained from 7 patients with acute DENV-1 infection ( mean fold-enhancement = 3 . 1 , P = 0 . 02 , and mean fold-enhancement = 3 . 2 , P = 0 . 02 respectively ) were significantly higher than to those of samples from non-acute DENV patients with multitypic neutralizing activity to ≥3 DENV serotypes ( mean fold-enhancement = 0 . 6 to DENV-1 , mean fold-enhancement = 0 . 8 to DENV-3 ) . Similarly , enhancement activities against DENV-3 of samples from 11 patients with acute DENV-3 infection ( mean fold-enhancement = 2 . 4 , P = 0 . 04 ) were higher than those of samples from non acute DENV patients with multitypic neutralizing activity to ≥3 DENV serotypes ( Table S5 ) . Using serum samples at 1∶10 dilution , neutralizing activities were higher in samples from patients with multitypic neutralizing activity to ≥3 DENV serotypes ( mean percentage of DENV-1 plaque reduction = 92% , DENV-3 plaque reduction = 92% ) than in those from patients with acute secondary DENV-1 infection ( DENV-1 plaque reduction = 43% , P<0 . 01; DENV-3 plaque reduction = 34% , P<0 . 01 ) and , than in those from patients with acute secondary DENV-3 infection ( DENV-1 plaque reduction = 63% , P = 0 . 02; DENV-3 plaque reduction = 55% , P<0 . 01 ) ( Table S6 ) . Neutralizing titers were higher in samples from non acute dengue patients with neutralizing activity to multitypic DENV than in those from patients with acute primary and secondary DENV infection . In contrast , enhancing activity was significantly lower in non acute dengue patients with neutralizing activity to multitypic DENV than in those from patients with acute secondary DENV infection ( Table S5 ) . The results suggest that serum samples with high neutralizing activity possess no or only low levels of infection-enhancing activity to respective serotypes at low serum dilutions . In contrast , serum samples without neutralizing activity possess high levels of enhancing activity .
The relationship between neutralizing activity determined using FcγR-negative cells and infection-enhancing activity determined using FcγR-expressing cells was examined in the present study . We determined that a high proportion of the serum samples from residents of a DENV endemic country , Malaysia , possessed DENV neutralizing and infection-enhancing activity ( 75% , 60/80; Table 2 ) . The neutralizing antibody titers were higher when determined by using FcγR-negative cells than when determined by using FcγR-expressing cells ( Table 3 ) . Mammalian cells such as Vero cells and BHK cells are commonly used in plaque reduction neutralizing tests [39] . However , in the absence of FcγR , these cells exclusively detect neutralizing antibody titers . Other investigators have suggested that the use of FcγR-expressing cells , including monocyte-lineage THP-1 cells and K562 cells , dendritic cells , FcγR-expressing CV-1 and BHK cells , and , DC-SIGN expressing RAJI cells and U937 cells may better reflect the in vivo neutralizing titers [40]–[46] . Surrogate plaque titration assays are , however , required to determine virus titers in non-adherent monocyte-lineage cells . Infection-enhancement was also detected using an FcγR-expressing cell line , THP-1 cells , for a subset of serum samples ( #39 , 40 , 28 and 77 ) which exhibited high infection-enhancement activity to DENV-4 using FcγR-expressing BHK cells ( data not shown ) . Thus , the results of ADE assays were consistent between THP-1 cells and FcγR-expressing BHK . In addition , serum samples ( #46 , 40 , 42 , 43 , 44 , 45 , 52 and 54 ) that demonstrated neutralizing antibody titers to DENV-2 ( 1∶10 to >1∶1280 ) along with neutralizing activities to other serotypes in FcγR-negative BHK cells exhibited neutralizing antibody only to DENV-2 in FcγR-expressing BHK cells . Cross-reactive neutralizing activity determined using FcγR-negative cells was not detected using FcγR-positive cells . This could be due to hampered neutralizing activity to heterologous serotypes by infection-enhancing activity [25] , [42] , [47] . Importantly , our results suggest that the FcγR-expressing BHK cells provides more informative data on serotype-specific neutralizing activity , and may better reflect protection in vivo . Presence of infection-enhancing activity in antibodies with neutralizing activity to the infecting serotype has been reported previously [37] , [38] , [47] . Interestingly , despite the result that 11 serum samples from patients with DENV infection exhibited neutralizing activity to the infecting serotype at 1∶10–1∶80 as determined by using FcγR-negative BHK cells , the neutralizing titer to the respective infecting serotypes as determined by using FcγR-expressing BHK cells was <1∶10 in all of the 11 serum samples tested . It has been reported that the main target cells of DENV infection in vivo are FcγR-expressing cells , such as monocytes and macrophages [34]–[36] . The results suggest that DENV-antibody complexes which are incapable of infecting FcγR-negative cells , may retain the ability to infect FcγR-expressing cells due to the presence of FcγR . However , further studies are required to identify the relationship between infection-enhancing activity and neutralizing activity during infection in vivo . All four DENV serotypes are found co-circulating in Malaysia . Although data was not available for the pre-dominant dengue virus serotype in Perak prior to 2008 , DENV-2 was the dominant circulating serotype in peninsular Malaysia between 1998–2000 and 2006–2007 ( Chua et al . , unpublished data ) . High levels of neutralizing activity against DENV-2 were detected in some of the serum samples ( Table 3 ) , indicate that these individuals may have been previously exposed to DENV-2 . Previous studies have suggested that ADE activity influences disease severity in patients with secondary DENV infection [32] , [48] . As low levels of serum dilutions may better reflect in vivo conditions , serum samples at dilutions of 1∶10 were used in the present study . The results showed infection-enhancing activity ( fold enhancement ) of 0 . 7–6 . 7 to the infecting serotype at serum dilutions of 1∶10 . The assay using FcγR-expressing BHK cells detected infection-enhancing activity to infecting serotypes at low serum dilutions in serum samples with neutralizing activity , suggesting the presence of ADE activity in vivo . Interestingly , there was an inverse relationship between infection-enhancement activity to a DENV serotype and high neutralizing activity as determined by using FcγR-negative BHK cells to the respective serotypes ( Figure 2 , Table 3 , Table 4 ) . Previous studies showed that higher dilutions of patient serum samples , in quantities that are not sufficient to support neutralization , enhance DENV infection [32] , [47] . Patient serum samples #49 , 58 , 42 , 45 , and 54 , require higher concentrations for neutralization in the presence of FcγR than in the absence of FcγR . In addition , serum samples #57 , 23 , 28 , 30 , 74 , 77 , and 78 exhibited similar neutralizing antibody titers in both FcγR-expressing BHK cells and FcγR-negative BHK cells [25] . Some antibodies may require lower threshold occupancy for neutralization , and , thus , virus neutralization may occur at similar concentrations both in FcγR-negative and FcγR-expressing cells . Alternatively , binding of some antibodies may lead to DENV conformational changes [49] , and therefore , result in virus neutralization both in the presence and absence of FcγR , at similar antibody concentrations . The FcγR-expressing BHK cell-based assay system is unique as antibody neutralizing activity could be analyzed and compared simultaneously using one cell line ( BHK cell line ) either in the absence or presence of FcγR by a conventional plaque assay . It is known that primary infection with one DENV serotype induces long-term protection to infection with the same serotype [50] . By using serum samples from a DENV-endemic area , we demonstrated that infection-enhancing activity which was determined only by using FcγR-expressing cells hampers neutralizing activity that was determined using FcγR-negative BHK cells . Moreover , infection-enhancing activity was also detected in serum samples with low or negative neutralizing activity that were determined using FcγR-negative BHK cells . Although in vitro systems may not faithfully reflect all aspects of DENV infection in vivo , the results suggest that as compared to the neutralizing activity determined by FcγR-negative cell culture system , the sum of neutralizing and enhancing activity determined by the FcγR-expressing cells may better reflect protection to DENV infection in vivo . Further studies are , however , needed to define the relationship between protection and neutralizing titers determined by using FcγR-positive cells . | Dengue has become a major international public health concern in recent decades . There are four dengue virus serotypes . Recovery from infection with one serotype confers life-long protection to the homologous serotype but only partial protection to subsequent infection with other serotypes . Secondary infection with a serotype different from that in primary infection increases the risk of development of severe complications . Antibodies may play two competing roles during infection: virus neutralization that leads to protection and recovery , or infection-enhancement that may cause severe complications . Progress in vaccine development has been hampered by limited understanding on protective immunity against dengue virus infection . We report the neutralization activity and infection-enhancement activity in individuals with dengue in Malaysia . We show that infection-enhancement activity is present when neutralizing activity is absent or low , and cross-reactive neutralizing activity may be hampered by infection-enhancing activity . Conventional assays for titration of neutralizing antibody do not consider infection-enhancement activity . We used an alternative assay that determines the sum of neutralizing and infection-enhancement activity in sera from dengue patients . In addition to providing insights into antibody responses during infection , the alternative assay provides a new platform for the study of immune responses to vaccine . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [
"medicine",
"vaccination",
"infectious",
"diseases",
"clinical",
"immunology",
"immunity",
"immunology",
"viral",
"diseases"
] | 2012 | Dengue Virus Infection-Enhancing Activity in Serum Samples with Neutralizing Activity as Determined by Using FcγR-Expressing Cells |
Understanding the principles governing the dynamic coordination of functional brain networks remains an important unmet goal within neuroscience . How do distributed ensembles of neurons transiently coordinate their activity across a variety of spatial and temporal scales ? While a complete mechanistic account of this process remains elusive , evidence suggests that neuronal oscillations may play a key role in this process , with different rhythms influencing both local computation and long-range communication . To investigate this question , we recorded multiple single unit and local field potential ( LFP ) activity from microelectrode arrays implanted bilaterally in macaque motor areas . Monkeys performed a delayed center-out reach task either manually using their natural arm ( Manual Control , MC ) or under direct neural control through a brain-machine interface ( Brain Control , BC ) . In accord with prior work , we found that the spiking activity of individual neurons is coupled to multiple aspects of the ongoing motor beta rhythm ( 10–45 Hz ) during both MC and BC , with neurons exhibiting a diversity of coupling preferences . However , here we show that for identified single neurons , this beta-to-rate mapping can change in a reversible and task-dependent way . For example , as beta power increases , a given neuron may increase spiking during MC but decrease spiking during BC , or exhibit a reversible shift in the preferred phase of firing . The within-task stability of coupling , combined with the reversible cross-task changes in coupling , suggest that task-dependent changes in the beta-to-rate mapping play a role in the transient functional reorganization of neural ensembles . We characterize the range of task-dependent changes in the mapping from beta amplitude , phase , and inter-hemispheric phase differences to the spike rates of an ensemble of simultaneously-recorded neurons , and discuss the potential implications that dynamic remapping from oscillatory activity to spike rate and timing may hold for models of computation and communication in distributed functional brain networks .
Our understanding of the biophysical mechanisms governing the dynamics of single neurons has increased dramatically over the past decades . In contrast , a principled understanding of the mechanisms governing ensembles of interacting neurons – from local cortical microcircuits to discrete functional areas to large-scale brain networks – remains elusive . Recent imaging advances have generated detailed structural maps that span from the micro-scale of local synaptic connectivity [1]–[3] to the macro-scale of hierarchical , long-range cortical networks [4]–[6] . However , a comprehensive description of the brain connectome [5] – the structural connectivity of the nervous system – is a necessary but not sufficient condition for understanding the dynamic coordination of brain networks . To survive in a complex world , agents must switch quickly from one task to another – for example , switching from thinking about an article to dodging a speeding car when crossing an intersection . Different tasks require the differential activation of separate functional networks , including ( but not limited to ) changes in the mean activity of multiple brain areas ( nodes ) as well as transient modulation of the effective strength of connectivity between areas ( directed links ) . Importantly , the modulation of distinct networks required for task switching occurs fast enough that structural connectivity can be considered relatively fixed . How is this dynamic coordination of networks accomplished ? Several groups have proposed that neuronal oscillations play a critical role in the dynamic coordination of multi-scale brain networks [7]–[22] . In this view , oscillations or brain rhythms may influence both local cortical computation [7] , [8] as well as long-range communication [21] . Furthermore , Lakatos et al . [17] proposed that a hierarchy of interacting oscillations , where low frequency phase modulates high frequency power , may serve to coordinate information flow across multiple spatial and temporal scales [18] . Most hypotheses about how oscillations may influence neural coding and network coordination predict that the spiking of single neurons is statistically dependent upon one or more distinct frequency bands of the local field potential ( LFP ) . We refer to this statistical dependence between the micro-scale of single neurons and the meso- and macro-scale of oscillatory network activity as cross-level coupling ( CLC ) . Importantly , if neuronal oscillations play a role in coordinating distributed networks , then we would expect to observe CLC between the neurons embedded in those networks and the different brain rhythms associated with their functional activation . That is , failure to observe CLC would be evidence against oscillations playing a role in dynamic coordination . Conversely , if CLC is present , then a quantitative descriptive model of the CLC observed within neuronal ensembles across different tasks may help distinguish between different mechanistic accounts of how brain rhythms modulate activity in distributed functional networks . That is , characterizing the distribution and stability of CLC parameters across neurons and tasks is an essential step that may prove useful for selecting between different possible mechanisms of oscillatory network control . CLC between spikes and internally-generated brain rhythms remains less well understood than the coupling between spikes and externally-associated factors such as visual orientation [23] , [24] or movement direction [25] , [26] . Nonetheless , empirical evidence for CLC between single neurons and a variety of different brain rhythms has been observed in several brain areas across a number of behavioral tasks [22] , [27]–[41] . This evidence includes spike-field coherence between single neurons and the gamma ( 30–80 Hz ) rhythm in the hippocampus [33] , [37] , basal ganglia [42] , and a variety of different cortical areas [22] , [29] , [43] . CLC in the theta ( 4–8 Hz ) band has been observed in the hippocampus [34] and between the hippocampus and prefrontal cortex [38] , with theta-phase precession of hippocampal place cells [44] serving as a prototype of dynamic CLC . Event-related changes in the power and synchronization of the gamma and theta bands suggests that these rhythms are related to functional activation [45] . For example , gamma synchronization in visual cortex is modulated by attention [39] and predicts the speed of change detection [46] , while increased hippocampal theta power precedes successful memory encoding in humans [47] . Furthermore , the role of oscillatory phase ( distinct from oscillatory power ) in neuronal coding and communication has emerged as a topic of growing interest . It was recently shown that spikes from prefrontal neurons occurring at particular phases of 32-Hz filtered LFPs were more informative about object coding during a working memory task than were spikes occurring at other phases [22] . Similarly , posterior parietal neurons coherent with 15-Hz beta activity were predictive of reaction times during a coordinated reach-saccade task , while other actively spiking neurons were not [32] . In primary visual cortex , gamma phase modulates orientation selectivity and noise correlations [43] , while entrainment of delta ( 1–4 Hz ) phase in visual cortex increased response gain and speeded reaction times [48] . However , compared to other rhythms such as theta and gamma , the cellular/network origins [49] , [50] and functional role [51]–[56] of sensorimotor beta oscillations are less well understood and remain subjects of considerable debate [16] . In primary motor cortex , Murthy and Fetz [27] were among the first to show that spike timing is dependent on the phase of the motor beta ( 10–40 Hz ) rhythm , with stronger phase-locking occurring when beta power is high . Reimer and Hatsopoulos [57] showed that precise spike timing depends on the combined influence of both external events as well as internally-generated ongoing beta activity – that is , motor cells are tuned to internal as well as external events . Intriguingly , beta activity is a mesoscopic phenomenon arising from microscale network interactions and results in propagating spatiotemporal waves that can encode information about upcoming movements [58] , [59] and action goals [60] . The strong coupling between single cells and the beta rhythm , on the one hand , and between beta and experimental task demands [52] , [61] , [62] , on the other , suggests that beta-band CLC in the motor system may prove useful in understanding the connection of single cells to the dynamic activation of functionally-defined neuronal ensembles . Therefore , a primary purpose of this study was to provide a quantitative characterization of CLC between the motor beta rhythm ( 10–45 Hz ) and a large ensemble of simultaneously-recorded neurons in primary motor cortex ( M1 ) across two distinct but related behavioral tasks ( Figure 1 ) . Such a characterization of CLC is required both for understanding the functional role of the beta rhythm in the motor system , on the one hand , as well as for evaluating the hypothesized role oscillations may play in coordinating large-scale networks more generally . Importantly , several aspects of CLC remain unclear . First , the degree of heterogeneity of CLC parameters across a population remains uncertain . Many prior studies employed acute recordings of single cells or small ensembles of simultaneously-recorded cells , pooling cells recorded serially over different days in order to make inferences about the distribution of CLC parameters over the neuronal population . This ergodic assumption makes it difficult to distinguish the case where a wide range of CLC parameters holds in a stable fashion for an ensemble over time , on the one hand , from the case where each cell is described by similar CLC parameters , but those parameters evolve dynamically over time . An advantage of the chronically-implanted microelectrode arrays used here is that a large ensemble of neurons can be recorded simultaneously , with identified single units followed over several days , permitting an unbiased assessment of the population diversity of CLC that holds during a given task . Second , it remains unclear how stable within-neuron CLC parameters are across different tasks . Few studies have investigated CLC in a given neuron over different tasks – that is , it remains unclear how variable CLC is within a given neuron as the subject switches from one task to another . This study examines CLC within the same neuron for two different tasks that have similar high-level goals , but are associated with distinct activation patterns – namely , a delayed center-out movement task performed via motion of the hand ( Manual Control , MC ) or via modulation of neuron firing rates ( Brain Control , BC; see Materials and Methods ) . Third , the relative importance of different aspects of oscillatory activity remain unclear . CLC has most often been assessed using spike- or cycle-triggered averages , or via spike-field coherence , but both of these methods combine oscillatory amplitude and oscillatory phase into one measure . Furthermore , these measures do not assess the dependence of spiking upon other factors such as synchronization between different regions . By computing these dependencies for the same set of neurons across conditions , the relative importance of different oscillatory factors to a given neuron can be made clear . Thus , characterizing the CLC of a large population of simultaneously-recorded neurons over two different tasks enables us to describe both the within-task , cross-neuron diversity of coupling to different aspects of oscillatory activity , as well as the cross-task , within-neuron stability ( or dynamic remapping ) of coupling that may occur between spiking and beta activity in the motor cortex . Here we present several findings . First , we provide a descriptive ( phenomenological ) model of the coupling between the instantaneous spike rate of a given cell and frequency-specific oscillatory activity . Importantly , this model accounts separately for the influence on rate of oscillatory amplitude , phase , and the interaction between amplitude and phase . Second , we show that this model describes the coupling for a large ensemble of cells , but that a wide range of model parameters holds across the population during a given task . In particular , some cells were more sensitive to amplitude than phase or vice versa , or had differential sensitivity to the interaction between amplitude and phase . Third , for a given cell the coupling of beta activity to spiking was stable across multiple sessions of a given task , but was often remapped when subjects switched to a different task . The parameter changes induced across the ensemble by this reversible remapping were reliable across multiple datasets . Fourth , it appears that this rhythm-to-rate mapping and task-dependent remapping have properties that would prove useful for the causal control of functional networks interactions . We conclude with a discussion of how these empirical results point to potential mechanisms for the control of neuro-computational processes . In sum , cross-level coupling between micro-scale spiking and meso- and macro-scale network activity appears to be a robust , flexible bridge linking together the different levels of brain organization required for effective perception , cognition , and action .
Two adult male rhesus monkeys ( Macaca mulatta ) were chronically implanted with multiple microelectrode arrays . Each array consisted of 64 Teflon-coated tungsten microelectrodes ( 35 mm in diameter , 500-mm interelectrode spacing ) arranged in an 8 × 8 array designed to target cortical layer V ( Innovative Neurophysiology , Durham , NC ) . Monkey P was implanted bilaterally in the arm area of primary motor cortex ( M1 ) , and in the arm area of left hemisphere dorsal premotor cortex ( PMd ) , for a total of 192 electrodes across 3 implants . 95 identified single units from this monkey were examined . Monkey R had bilateral implants in the arm area of M1 and PMd , for a total of 256 electrodes across four implants . 86 identified single units from this monkey were examined . Localization was performed using stereotactic coordinates [63] . Implants targeted layer-5 pyramidal tract neurons and were positioned at a depth of 3 mm ( M1 ) or 2 . 5 mm ( PMd ) . Intraoperative monitoring of spike activity guided electrode depth . See [64] for full experimental details . A 128 channel multi-acquisition processor ( MAP ) system ( Plexon Inc . , Dallas , TX ) was used to record unit activity . Only single units that had a clearly identified waveform with a signal-to-noise ratio of at least 4∶1 were used . An on-line spike-sorting application ( Sort-Client; Plexon Inc . , Dallas , TX ) was used to sort activity prior to recording sessions . Large populations of well-isolated units and up to 128 LFP channels ( 1 kHz sampling ) were recorded during daily sessions for both monkeys . Conducted procedures were in compliance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals and approved by the University of California at Berkeley Institutional Animal Care and Use Committee . Monkeys were trained to perform a delayed center-out reaching task using either their natural arm inside a Kinarm exoskeleton ( BKIN Technologies , Kingston , Ontario ) ( Manual Control , MC ) , or under direct neural control through a brain-machine interface ( BMI ) and irrespective of overt physical movement ( Brain Control , BC ) [64] . Monkeys self-initiated trials by bringing the cursor to the center for a hold period ( MC , 500 ms; BC , 100 ms ) , followed by the presentation of a GO cue ( color change of center cue ) . A trial error occurred if the cursor failed to reach the target within 10 s after a GO cue . The goal was to perform a center-out task , moving the cursor from the center to one of eight peripheral targets distributed over a 14-cm diameter circle . Required hold times at target were 400 ms for MC and 50 ms for BC . Target radius was typically 0 . 75 cm . A liquid reward was provided after a successful reach to each target . During training and recording animals sat in a primate chair with their heads restrained . During BC sessions the Kinarm was removed and the arms restrained to the primate chair . Analyses were done using MATLAB ( Mathworks ) . Filtering to extract beta amplitudes and phases was performed by convolving signals with Gabor time-frequency basis functions ( Gaussian envelope ) . A Gabor time-frequency atom is fully defined by three parameters; namely , the center time t0 , the center frequency v0 , and the duration parameter s0 . In the time domain , the Gabor atom g is given as g ( t | t0 , v0 , s0 ) = 21/4 exp[ ( − 1/4 ) s0 − p ( t − t0 ) 2 exp[ − s0] + 2p v0 ( t − t0 ) ] . Since there was no significant difference in the frequency corresponding to the power spectral peak ( power of −46 10*log10 ( µV2/Hz ) at a frequency of 28 Hz ) , a fixed center frequency v0 of 28 Hz centered on the observed PSD peak and a duration parameter s0 of −5 . 075 ( frequency domain standard deviation of 3 . 57 Hz ) were used to extract the “beta signal” this study . For the amplitude-to-rate , phase-to-rate , phase-difference-to-rate , and amplitude-to-weight mappings , a spatial average of all LFPs recorded from the 64 electrodes in one microelectrode array was generated and used as the raw input signal . Each 8 × 8 microelectrode array covers an area of 3 . 5 × 3 . 5 mm2 , and therefore this spatial average is similar in scale to a single ECoG macroelectrode . Two average signals sL and sR , were generated for left and right M1 , respectively , prior to additional analyses . After concatenating separate recording blocks , the BC dataset had a duration of 410 minutes for monkey P ( 97 minutes for monkey R ) , while the MC dataset had a duration of 172 minutes for monkey P ( 58 minutes for monkey R ) . To compute event-related potentials ( ERPs ) and event-related time-frequency amplitude maps , the signal indices of go cue onsets were identified for BC and MC . For ERPs , trial epochs −1000 ms before to 10000 ms after go cue indices were extracted from signals sL and sR were averaged . For time-frequency analyses , first the signals sL and sR were filtered around a given center frequency as described above . 40 center frequencies spaced semi-logarithmically from 1 to 300 Hz were employed . Second , the amplitude of each filtered signal was normalized such that the mean amplitude across all data was 1 . Third , trial epochs −1000 ms before to 10000 ms after go cue indices were extracted from the amplitude time series and averaged . RT-sorted single-trial analyses ( e . g . , Figure 2C , F ) were performed similarly , but rather than averaging all trial epochs together , a sliding window of 250 trials was used after sorting all trials by movement duration . Event-related PSTHs were computed similarly to ERPs , using a binary time series representing spike times . To generate the beta amplitude-to-rate mapping , first the time series of instantaneous amplitudes was extracted from one of the average signals ( sL or sR ) described above . Call this N × 1 vector of amplitudes xA . Amplitudes were normalized such that the mean amplitude across all data was 1 . Second , the spike times from one neuron were used to generate a N × 1 binary vector xS , where xS[t] equals 1 for spike times t and equals 0 otherwise . Third , a N × 2 matrix M1 = [xA xS] was formed . Fourth , this matrix M1 was truncated to allow reshaping of the array in a future step – given the number of amplitude bins to be used later ( nb ) , this matrix M1 was truncated to form the Nt × 2 matrix M2 , where Nt is the largest integer less than or equal to N for which nb * P = Nt for some integer P . Fifth , the rows of the matrix M2 were sorted according to the amplitude values in the first column of M2; M2 = sortrows ( M2 , 1 ) . Sixth , a 3-dimensional array of size P × nb × 2 was created by reshaping the sorted matrix M2: M3 = reshape ( M2 , [P nb 2] ) . Seventh , the mean amplitude for each bin was computed: A = mean ( M3 ( : , : , 1 ) , 1 ) , where A is a 1 × nb vector of amplitudes . Eighth , the average spike rate for each bin was computed: R = ( SR/P ) *sum ( M3 ( : , : , 2 ) = = 1 ) , where SR is the sampling rate and R is a 1 × nb vector of spike rates . The rate values R over the amplitude support A describe the empirically-observed amplitude-to-rate mapping . Ninth , this histogram-based mapping was fit with 4-parameter sigmoidal function FS using the MATLAB function lsqcurvefit . m , where FS ( a ) = p1 + p2 tanh ( ( a − p3 ) / ( 2 p4 ) ) , where a<0 is beta amplitude and tanh is the hyperbolic tangent function . In order to assess task-related changes , amplitude-to-rate mappings were computed separately for the full BC and MC datasets . In order to assess within-task stability of the mappings , the BC dataset was split into two disjoint datasets , BC1 and BC2 consisting of odd and even trials , respectively , and the above procedure performed separately on each . Similarly , split-half reliability during MC was assessed using two disjoint datasets MC1 and MC2 . The empirical-observed estimates of the beta phase-to-rate and phase-difference-to-rate mappings were produced in an identical way , where the xA time series of step 1 was replaced with a N × 1 vector of instantaneous phases from one hemisphere ( phase-to-rate ) or a N × 1 vector of inter-hemispheric phase differences ( phase-difference-to-rate ) . For the phase-to-rate mapping , a cosine-type function was used in fitting: FC ( q ) = p1 + p2 cos ( θ − p3 ) , where θ is beta phase , p2>0 , and q , p3 are in the interval [ − p , p ) . For the phase-difference-to-rate mapping , a von Mises-type function was used in fitting: FD ( j ) = p1 + p2 exp[p3 cos ( φ − p4 ) ] , where φ is the phase difference , p3>0 , and φ , p4 is in the interval [ − p , p ) . For target-specific and trial-stage-specific analyses , the data was presorted to extract only relevant time intervals . To determine the amplitude-to-weight mapping that describes the multiplicative gain effect beta amplitude has on the phase-to-rate mapping , a procedure similar to that describe above was performed , but sorting datapoints jointly by amplitude and phase . That is , first the amplitude time series xA , the phase time series xP , and the spike time series xS were combined into a matrix W = [xA xP xS] , such that each row represents the amplitude , phase , and spike status of one sample point . Second , the rows of W were sorted according to the values in the amplitude column and partitioned into nab ( amplitude ) bins , where each bin has the same number of sample points . Third , the data in each ( amplitude ) bin was further sorted into npb ( phase ) bins . Fourth , the spike rate for each ( amplitude , phase ) bin was computed , generating a nab × npb matrix of spike rates . Fifth , this matrix was used to constrain the fitting of the 7-parameter function describing the full beta-to-rate mapping: FB ( a , θ ) = RBETA ( a , θ ) = p1 + p2 tanh ( ( a − p3 ) / ( 2 p4 ) ) + ( p5 a + p6 a2 ) cos ( θ − p7 ) . Given the amplitude-to-rate mapping RAMP ( a ) and the phase-to-rate mapping RPHASE ( θ ) described above , the quadratic weight factor or amplitude-to-rate mapping wAMP ( a ) = b1 a + b2 a2 can be extracted from the relation RBETA ( a , θ ) = RAMP ( a ) + wAMP ( a ) RPHASE ( θ ) . The modeling approach described above is inherently univariate and does not extend easily to multivariate approaches . For the multivariate analysis we used a procedure similar to that described in [41] , but adapted to account for both amplitudes and phases . First , the Nchannel LFP signals from a training dataset were filtered to generate a Nchannel × Nsamples complex-valued matrix , where for each entry the absolute value gives the beta amplitude and the argument gives the beta phase . Second , this matrix was used to fit the parameters describing the complex multivariate Gaussian distribution [65]: p ( x ) = β exp[ − 1/2 ( xB − μB ) H RB−1 ( xB − μB ) ] , where x is a Nchannel × 1 vector of complex values , xB = [x; conj ( x ) ] is the 2Nchannel × 1 ( augmented ) vector of complex values , μB is a 2Nchannel × 1 vector of complex values representing the mean of xB , RB is the 2Nchannel × 2Nchannel ( augmented ) covariance matrix of xB , b = 1/ ( πN sqrt ( det ( RB ) ) ) is a normalization term , conj ( x ) returns the complex conjugate of x , and the superscript H represents the conjugate transpose operation . Call this distribution , fit using all data , the baseline distribution pBASE ( x ) . Third , perform another distribution fitting using LFP data from spike times only; call this the spike-triggered distribution pST ( x ) . Fourth , from a new training dataset of filtered LFP signals , extract the Nchannel × 1 vector representing each sample point and compute the log-likelihood ratio L ( x ) = log[pST ( x ) /pBASE ( x ) ] , generating a 1 × Nsamples time series of log-likelihood ratio values . Call this time series L . Fifth , compute the L-to-rate mapping for this training dataset , as was described above for the amplitude-to-rate mapping . Sixth , find the best 4-parameter sigmoid fit FS ( L | p ) for the L-to-rate mapping , where p is a 4 × 1 parameter vector ( see sigmoid function definition above ) . Seventh , given a novel test dataset of filtered LFP signals , extract the Nchannel × 1 vector representing each sample point and compute the predicted instantaneous spike rate estimate REST ( x ) = FS [log[pST ( x ) /pBASE ( x ) ]] . Eighth , evaluate the prediction by computing the estimated-rate-to-measured-rate mapping , computed as was done to estimate the amplitude-to-rate mapping .
Given the strong event-related changes in beta amplitude during both MC and BC ( Figure 2B , E ) , we first investigated the dependence of spike rates on beta amplitude alone ( neglecting beta phase or beta phase differences between sites ) . We term this functional dependence between beta amplitude and spike rate the amplitude-to-rate mapping , consistent with the idea that a given neuron responds to both internal and external factors ( Figure 3 ) . This analysis revealed two key findings . First , within a given task such as BC , the population of simultaneously-recorded neurons exhibited a wide range of responses to changes in beta amplitude , with some neurons increasing firing , some exhibiting no change , and other decreasing their spike rate ( Figures 4A , H–I; S2A–F; S10 ) . Second , a single neuron may exhibit a task-dependent remapping of the amplitude-to-rate function – for example , a given neuron may increase firing as beta amplitude goes up during BC , but decrease firing when beta amplitude increases during MC ( Figures 4B–G , J–K; S1A–F; S2G–H ) . In more detail below , we consider i ) the diversity of within-task amplitude-to-rate mappings observed across the full neuronal ensemble , and ii ) the diversity of amplitude-to-rate remappings that can occur within a single neuron when switching from one task to another . Across both the MC and BC tasks , a statistical dependence between spike rate and beta amplitude was observed for 86 . 7% of the cortical motor neurons examined ( p<0 . 01 uncorrected randomized permutation test; c . f . Table S1 ) . That is , for a given neuron the spike density conditioned on low beta amplitude is different than the spike density conditioned on high beta amplitude . We found that the empirically-observed amplitude-to-rate mapping was well-described by a 4-parameter sigmoidal function ( see Materials and Methods ) . Across the neuronal population , this mapping was described by a wide range of model parameters . For example , Table S1 shows that as beta amplitude increased during BC , 58 . 6% of all neurons exhibited a decrease in spike rate while 28 . 2% showed an increase in spike rate . As beta amplitude increased during MC , 35 . 4% ( 40 . 3% ) of neurons decreased ( increased ) their firing rate . Importantly , while the neuronal population exhibited a wide range of parameter values during a given task , the parameters for single neurons showed high stability across different sessions of the same task ( Figures 4H , I; S2C , F ) . Figure 4A shows the range of amplitude-to-rate mappings during BC for twelve example neurons from monkey P . Dots show the empirically-observed spike rates conditioned on beta amplitude , computed separately for 25 non-overlapping amplitude bins . Note that adaptive binning was employed such that each bin includes the same number of sample points , resulting in non-uniform bin spacing . Lines indicate the best-fit sigmoidal functions describing the observed amplitude-to-rate mapping ( see Materials and Methods ) . Some neurons show a decrease in spike density with increasing beta amplitude while others show an increase in spike density . Furthermore , as shown for a different set of neurons in Figure S10 , removing the offsets due to baseline spike rates reveals that the rate of change of the mapping ( neglecting sign ) is large for some neurons ( purple , gold ) , moderate for others ( green , cyan ) , and small for yet others ( black , red ) . Critically , for a given neuron the sign and slope of the amplitude-to-rate mapping is not correlated with the cell's baseline firing rate , and are different for distinct cells . Thus , a change in beta amplitude does not imply a stereotyped change in spike density that applies to all neurons uniformly; different neurons exhibit differential responses to changes in beta amplitude . That is , while averaging across all recorded cells reveals a negative correlation between ensemble spike rate and beta amplitude , investigating each cell separately reveals that some neurons exhibit a strong negative correlation with beta amplitude , others a strong positive correlation , while yet others show only a weak or negligible dependence on amplitude ( c . f . Figure 4H , I ) . Interestingly , while there is variation in the exact crossover point for the population of amplitude-to-rate mappings – that is , the amplitude value where the sigmoid function intersects the baseline rate – the amplitude-to-rate mappings for many neurons intersect near the mean beta amplitude ( Figure S10 ) . In all analyses presented here the mean beta amplitude has been normalized to 1 . However , while most baseline-free amplitude-to-rate mappings intersect near the same beta amplitude ( the mean value ) , adding different baseline rates can shift the crossover point for pairs of amplitude-to-rate mappings . For example , Figure 4A shows the intersection of the amplitude-to-rate mappings for neuron sig060a ( cyan , bottom ) with the amplitude-to-rate mappings of three other cells: sig099a ( blue ) , sig104a ( orange ) , and sig031a ( purple ) . Vertical lines mark the amplitude values where these curves intersect . For low beta amplitudes , sig060a ( cyan ) has a higher spike rate than the other 3 neurons , while for high beta amplitudes this neuron has a lower rate . For intermediate amplitude values sig060a has a higher rate than some neurons but not others . Large differences in baseline rates result in no overlap of amplitude-to-rate functions and thus no change in the relative rank ordering of neurons in terms of spike rate . Thus , for an ensemble of neurons with similar baseline rates , the common intersection point of the amplitude-to-rate mappings near the mean beta amplitude results in two distinct spike density regimes for the ensemble . That is , when the instantaneous beta amplitude is above its mean value , then there is an associated rank ordering of the spike rates across the population of neurons ( relative to the tonic baseline rate for each neuron ) . For example , given the seven example neurons shown in Figure S10 , a beta amplitude above the mean is associated with the rank order gold , cyan , red , blue , black , green , and purple ( ranked highest-to-lowest in terms of change in spike rate relative to the baseline rate for each neuron ) . In contrast , when the instantaneous beta amplitude is below the mean , then this rank ordering is reversed . However , for neurons with different tonic baseline spike rates ( Figure 4A ) , the amplitude-to-rate mappings of each pair of neurons may cross at beta amplitude values far from the mean value . That is , each pair of neurons in an ensemble may switch their spike rate rank order at any of a wide range of amplitudes , greatly expanding the set of rank-order states possible for the ensemble . Furthermore , each rank-order ensemble state – where an ensemble state is defined as an ordered list of neurons sorted in terms of spike rate – is indexed by a finite range of beta amplitudes . The vertical lines in Figure 4A show three transition points between such states; across both BC and MC conditions , the amplitude-to-rate mappings for the 12 neurons shown in Figure 4A and Figure S2 establish a set of 41 such states , each of which is associated with a finite interval of beta amplitudes . The within-task diversity of amplitude-to-rate mappings observed across the ensemble of recorded neurons is complemented by a different type of diversity – within-neuron , cross-task diversity – that is associated with task switching . Interestingly , while the amplitude-to-rate mappings for a given neuron are relatively stable across multiple datasets as long as all recordings are acquired under the same task conditions ( Figures 4H , I ) , switching from one task to another is associated with a reliable remapping of the amplitude-to-rate relation within a single neuron ( Figures 4B–G , J , K; S1A–F , S2 ) . That is , a given neuron may exhibit one stable amplitude-to-rate mapping during a session performed under MC , then switch to a distinct ( and stable ) amplitude-to-rate mapping during a second session conducted under BC , and finally return to the original amplitude-to-rate mapping when performing a third session under MC . Figures 4B–G show six example neurons that give an indication of the range of task-dependent remapping that can occur . For example , the single unit sig119b ( Figure 4F ) shows a strong decrease in spike density associated with increasing beta amplitude during MC ( blue ) , but little to no change in spike density during BC ( red ) . Unit sig045a ( Figure 4C ) shows a different response , with a moderate decrease in spike rate during MC but a large decrease in rate under BC . Unit sig038a ( Figure 4E ) exhibits a positive correlation between spike rate and beta amplitude , with a larger total change of rate under MC . Other cells exhibit the direction reversal shown by units sig104a ( Figure 4B ) and sig112b ( Figure 4D ) , with rate increasing under one task and decreasing in another . As shown by Table S2 , of the 64 . 1% ( 116/181 ) of neurons across both monkeys that exhibited significant amplitude-to-rate mappings during both MC and BC , 27 . 6% showed a direction reversal when comparing the MC and BC amplitude-to-rate mappings ( similar to neuron sig104a in Figure 4B ) . Of the cells positively correlated with beta amplitude under both BC and MC , 61 . 8% of cells exhibited a larger modulation depth for MC compared to BC ( similar to neuron sig038a in Figure 4E ) , while the remaining 38 . 2% showed a smaller modulation depth for MC compared to BC . For the 43 . 1% of cells exhibiting a negative correlation between rate and amplitude for both MC and BC , 56 . 0% of cells exhibited a larger modulation depth for MC compared to BC , while the remaining 44 . 0% showed a larger modulation depth for BC compared to MC ( similar to neuron sig045a in Figure 4C ) . In addition to task-dependent remapping , the amplitude-to-rate mapping also exhibits changes as a function of trial-stage , as shown for three example neurons in Figure S9 ( c . f . , Figure S5 ) . Importantly , task-related differences are observed during similar trial sub-stages . For example , considering data from segments of the trial where the goal was to move the cursor to the center cue , or toward one of the peripheral targets , still resulted in differences in the amplitude-to-rate mapping between MC and BC ( Figure S9A–B , D–E , G–H ) . Periods where goal-directed activity was presumably minimized , such as the period around reward delivery , also resulted in task-related differences ( Figure S9C , F , I ) . This suggests that task-related changes are distinct from trial-substage related changes , although further experiments will be required to fully disentangle the influence of top-down task demands from bottom-up trial-stage-related changes on the amplitude-to-rate mapping . Importantly , both the within-task amplitude-to-rate mappings and the cross-task amplitude-to-rate remappings are stable across multiple data sets ( Figure 4H–K ) . For example , Figures 4H–I shows the stability of these mappings when the data is divided into two disjoint datasets for each task , and the amplitude-to-rate mappings are computed separately for each dataset . That is , the correlation between parameters defining the amplitude-to-rate mappings computed from two blocks of data from the same task ( BC1/BC2 , Figure 4H , red; MC1/MC2 , Figure 4I , blue ) is higher than the parameter correlation between amplitude-to-rate mappings estimated from different tasks ( BC1/MC2 , Figure 4J; BC2:MC1 , Figure 4K ) . Therefore , the distribution of amplitude-to-rate mappings observed across the neuronal ensemble is stable from one dataset to another as long as the same task is being performed ( within-task stability of mappings ) , but exhibits a reliable and reversible shift when switching from one task to another ( cross-task reliability and reversibility of remappings ) . This association of the beta amplitude-to-rate mapping with an ordered sequence of discrete ensemble states – and the ability to create a new sequence via task-dependent remapping of the amplitude-to-rate relations that hold within an ensemble – has intriguing implications for neuronal computation , which we explore further in the discussion . As with beta amplitude , cortical motor neurons exhibited a spike density dependence on beta phase . During BC ( MC ) , the spike density of 87 . 3% ( 91 . 2% ) of neurons exhibited cosine modulation when conditioned on beta phase ( p<0 . 01 , uncorrected randomized permutation test ) . That is , when considering beta phase alone ( neglecting beta amplitude for now , but see beta-to-rate mapping below ) , the change in spike rate as a function of beta phase ( phase-to-rate mapping ) was well-fit by a 3-parameter cosine-type function ( see Materials and Methods ) . Figures 5A–H show eight example neurons recorded from right primary motor cortex ( M1 ) that exhibit this phase-to-rate mapping for BC ( red ) and MC ( blue ) , superimposed on the fits for all 95 simultaneously-observed neurons from monkey P ( grey ) . For all neurons that exhibited sensitivity to beta phase , this mapping was unimodal , with no neurons exhibiting multimodal dependence on beta phase . As with the amplitude-to-rate mappings , most neurons exhibited stable within-task phase-to-rate mappings ( evaluated using different datasets recorded under the same task conditions ) as well as reliable and reversible cross-task phase-to-rate remappings ( evaluated using datasets recorded under BC or MC; c . f . Figures S4 ) . For a given neuron , both the modulation depth ( maximum rate minus baseline rate ) and preferred beta phase ( beta phase exhibiting the maximum spike rate ) could change from one task to another . For example , some neurons show few cross-task changes ( e . g . , Figure 5A ) , while others primarily exhibit a change in preferred phase alone ( Figure 5D ) , or a change in modulation depth alone ( Figure 5E ) , or a change in both modulation depth and preferred phase ( Figure 5F ) . Interestingly , for all 53 neurons from right M1 of monkey P , the preferred beta phase during BC was earlier than the preferred phase under MC , as shown in Figure 5I . This systematic shift in preferred beta phase was uncorrelated to the tonic baseline firing rates of neurons in either task ( Figure S3D ) , and was also uncorrelated with the change in baseline rates from one task to another . In other words , this systematic shift in preferred phase is not due to a simple rate-to-phase conversion of the type shown in in vitro studies [66] . Furthermore , this preferred beta phase was unrelated to the strength of motor direction tuning or preferred movement direction ( Figure S3H , L ) . Given the peak in the LFP power spectrum at a center frequency of 28 Hz , the average beta cycle occurs over ∼36 ms . Therefore , we can convert a set of preferred phases into a set of most-probable spike times relative to a fixed point in the cycle of the ongoing beta rhythm . That is , if a neuron is going to spike only once in a given beta cycle , it is most likely to do so at its preferred beta phase , which corresponds to a fixed temporal lag relative to the peak of the beta waveform . The ordered set of these lags across the population imposes a probabilistic rank-ordering of spike times across the ensemble that spans ∼10 ms ( sorted red dots in Figure 5J and 5L ) . This is not a strong deterministic ordering of spike times but is rather a probabilistic or stochastic effect . Nonetheless , the knowledge of the preferred beta phase for a pair of cells can be informative about their relative spike timing . As a concrete example , consider neurons sig038a ( Figure 5C ) and sig045a ( Figure 5F ) . The phase-to-rate mapping for cell sig038a peaks earlier within the beta cycle than does the phase-to-rate mapping for sig045a ( 3 . 42 radians vs . 4 . 22 radians , respectively ) . Therefore , given a particular cycle where each cell spikes exactly once during the cycle , we would expect sig038a to spike earlier than sig045a . In fact , considering only the cycles where each cell fires exactly once , sig038a fires before sig045a 61 . 0% ( 40684/66678 ) of the time . Interestingly , however , due to the ensemble-wide , task-dependent shifts in the preferred beta phases , several pairs of neurons exchange their most probable firing order when switching from one task to another . For example , the unit sig064b ( Figure 5H ) is most likely to fire before unit sig043c ( Figure 5E ) during BC ( red ) , but the order switches during MC ( blue ) . Finally , the phase-to-rate mapping is strongly affected by the magnitude of beta amplitude ( Figure 6 ) . For example , Figures 5A–D show the phase-to-rate mappings for 9 example cells , computed separately for data falling into 4 beta amplitude bins ( corresponding to 0–25 , 25–50 , 50–75 , and 75–100th amplitude percentiles ) . The relative sizes of the phase-to-rate mappings shown in Figures 5A–D are largely similar across the four amplitude bins , with all cells exhibiting an increase in phase-to-rate modulation depth ( note vertical scale in Figure 6A–D ) . This increase phase-to-rate modulation depth as a function of beta amplitude is shown explicitly in Figure 6E . However , all cells do not change their modulation depth at the same rate – as beta amplitude increases , some cells increase their phase-to-rate modulation depth at a faster rate than others . Thus , while sig045a ( black ) starts out with the largest firing rate at low beta amplitudes , at high beta amplitudes it has fallen to rank 3 among the 9 neurons shown . Similarly , sig029a ( blue ) falls from rank 2 to rank 5 , while sig043c ( purple ) moves up from rank 4 to rank 2 . In addition to these differential changes in total phase-to-rate modulation depth as function of beta amplitude , there can also be shifts in the phase value associated with the crossover point of the phase-to-rate mappings for a pair of neurons . This can be the case even if each neuron maintains the same preferred beta angle . For example , the phase-to-rate mappings for sig043c ( purple ) and sig029a ( blue ) cross near −3 radians at low beta amplitudes ( vertical line ) , but shifts to approximately −1 . 3 radians for high beta amplitudes , even though the preferred beta phase for each cell remains the same . The cosine-shaped phase-to-rate mapping is a nonlinear function of phase , and combined with differential changes in vertical scale as a function of amplitude , these exchanges of spike-rate rank order as a function of beta phase can occur even when each cell has no change in its preferred beta phase angle . While the ( within-task ) phase-to-rate preferred beta angle is largely stable for most neurons across the full range of beta amplitudes ( Figure S4 ) , the change in phase-to-rate modulation depth can be described by a quadratic function of beta amplitude ( Figure 6F ) . We call this term describing the gain control of the phase-to-rate modulation depth the weight factor or the amplitude-to-weight mapping . This weight factor is sublinear function of amplitude for some neurons ( sig029a , blue ) , near linear for others ( sig038a , yellow ) , and supralinear for yet others ( sig027a , black ) . Despite this variability , across all neurons larger beta amplitudes are associated with larger modulation depths . That is , beta amplitude appears to act as a gain control for beta phase , such that beta phase is more predictive of spike timing when beta power is high than when beta power is low . However , just as the heterogeneity of tonic baseline rates and sigmoidal amplitude-to-rate functions interact to establish an ordered set of ensemble rank-order states that are indexed by beta amplitude , ensemble heterogeneity in phase-to-rate modulation depth and the quadratic weight factor can interact to establish an ordered set of rank-order states indexed by beta phase . That is , for a given beta amplitude , a set of overlapping phase-to-rate mappings have crossover points that occur at particular beta phases ( vertical line in Figure 6A ) . As beta amplitude changes , however , these crossover points can shift to new phases ( vertical line in Figure 6D ) . Furthermore , different tonic baseline rates can force some crossover points to disappear or introduce new crossings . Thus , both the amplitude-to-rate and phase-to-rate mappings , considered across an ensemble of neurons , can be associated with ordered sets of neurons , where cells are sorted according to instantaneous spike rate . Nonetheless , one consequence of this interaction between amplitude and the phase-to-rate modulation depth is that the spike timing preference relative to beta phase becomes stronger for higher beta power . For example , we saw above that neuron sig038a fires before neuron sig045a 61 . 0% percent of the time when looking at all beta cycles where each cell fires once . Sorting these individual cycles according to beta amplitude , however , reveals that sig038a spikes before sig045a only 56 . 2% of the time for cycles in the lowest decile beta amplitudes , compared to 72 . 0% of the time for the cycles in the highest decile of beta amplitudes . Similar effects are seen across the population , and thus variations in beta amplitude influence the probability of observing arbitrarily spike timing sequences within an ensemble . As with the amplitude-to-rate mapping , the phase-to-rate mapping exhibits both task- and trial-stage-related changes ( Figures S6 , S7 ) . As shown by Figure 3 , neurons respond to both internal and external factors , including time-in-trial and target direction . One possibility is that changes in the amplitude- and phase-to-rate mappings may arise from the interaction of multiple external and internal factors each of which influences the overall spike rate . For example , Figures 3B–C show that the target-specific spike rate for neurons can span a wide range of values from the most preferred to least preferred targets . Similarly , Figure 7 shows the target-specific phase-to-rate mappings for 6 example neurons during BC . Each neuron exhibits changes in the baseline rate that is a function of target direction or target ID . For example , Figure 7A shows that sig045b has the highest baseline rate for Target 8 ( black ) and the lowest baseline rate for Target 4 ( green ) . However , removing this target-specific baseline offset , as done for the same cell in Figure 7E , shows that the phase-to-rate modulation depth exhibits target-specific changes as well – in fact , the target-specific baseline rates of sig045b are positively correlated with the target-specific changes in the phase-to-rate modulation depth ( Figure 7M , dots ) . Similarly , sig038a exhibits target-specific variation in both the baseline rates ( Figure 7B ) and the phase-to-rate modulation depth ( Figure 7F ) , also with a positive correlated between them ( Figures 7M , circles ) . However , sig043b and sig020a exhibit little target-specific variation in the phase-to-rate modulation depth ( Figures 7G , H ) despite strong target-specific variation of their baseline firing ( Figures 7C , D ) . In fact , the slope of the regression line between target-specific baseline rates and modulation depths is close to zero ( Figure 7M , diamonds and crosses ) . On the other hand , cells such as sig073b and sig043c exhibit a negative correlation ( Figure 7M , asterisks and squares ) . Thus , while the amplitude-to-rate mapping may exhibit target-specific changes , the functional nature of this relationship remains unclear . Because the spike density of a given neuron depends on the interaction between beta amplitude and phase , the most complete picture of the dependence between spike rates and the beta rhythm is given by the full beta-to-rate mapping ( where the term ‘beta’ here implies both beta amplitude and beta phase ) . That is , the estimated spike rate RBETA ( a , θ ) is a sum of two terms: an amplitude-only rate RAMP ( a ) given by the amplitude-to-rate mapping , and another term that is the product of the phase-only rate RPHASE ( θ ) and a weight factor that is a function of amplitude alone , wAMP ( a ) . Specifically , where RAMP ( a ) is a sigmoidal function of amplitude , wAMP ( a ) is a quadratic function of amplitude , and RPHASE ( θ ) is a weighted cosine function of phase ( see Materials and Methods ) . Alternatively , appears to provide a good fit to data across the ensemble , given sufficient data . The primary result of this study is the finding that parameters describing this beta-to-rate mapping can change reversibly from one set of stable values to a different set of stable values when switching from one task to another . However , while the beta rhythm exhibits strong event-related changes in power ( c . f . , Figure 2 ) , it is possible that different frequency bands may prove even more predictive of cell spiking . An example of this is provided for neuron sig045a in Figure 8A , where the amplitude-to-rate mappings for a wide range of frequencies from 1–300 Hz are shown . Figure 8B shows the amplitude-to-rate mappings for four different center frequencies . For this cell , a center frequency of 6 Hz is as informative as is 27 Hz . However , as shown by Figures 8C–D , for this cell the informative phase-to-rate mappings are restricted to a narrow range of frequencies centered around 34 Hz . Figure 8E shows the range of spike rate variation for this cell as a function of center frequency for both the amplitude- and phase-to rate mappings . Interestingly , the amplitude- and phase-to-rate mappings appear to peak at different but possibly overlapping bands . The center frequency used for this study , 28 Hz , intersects both profiles near their peak , and thus provides information from both types of mapping . Figures 8F–O provide amplitude- and phase-to-rate mappings for 5 example cells over a range of center frequencies from 20 to 40 Hz , showing that the best phase-to-rate center frequency is often higher than the best amplitude-to-rate center frequency – a result that holds across the ensemble . The importance of optimal phase-to-rate center frequencies ∼30–38 Hz , however , is difficult to reconcile with the lack of strong event-related power changes ( Figure 2B , E ) or event-related phase-resetting ( not shown ) in this band . Whether or not the amplitude- and phase-to-rate mappings arise from distinct bands will require further experimental inquiry targeting this question . So far we have only considered the relation between ( micro-scale ) spiking of single neurons to ( meso-scale ) beta LFP activity averaged locally over several millimeters . That is , for a neuron in left M1 , we examine the relation of its spike rate to the average beta activity recorded in left M1 , while for neurons recorded in right M1 we examine the average field potential activity from right M1 . The 8 × 8 electrode arrays used here cover 3 . 5 × 3 . 75 mm2 , such that the LFP signals recorded on opposite sides of the array are generated by distinct cell populations , and the spatial average of all 64 LFP signals from one array is a meso-scale signal similar in scale to the activity recorded from one electrocorticography ( ECoG ) electrode as employed for human neurosurgery . The results above show a clear dependence between micro- and meso-scale phenomena . However , the relation between micro-scale spiking activity and fully macro-scale phenomena – such as phase coupling between the left and right motor cortices – remains unclear . Are some neurons sensitive to the phase difference between left and right M1 , above and beyond the influence that can be attributed to locally-generated field potential activity ? To address this question , we examined the relationship between ( micro-scale ) single-unit spiking and the ( macro-scale ) relative phase difference between the beta activity occurring in the left and right primary motor cortices . While this quantity neglects the beta amplitude in each area , it has the advantage of being statistically independent of the ( absolute ) beta phase local to the neuron . That is , knowing the instantaneous beta phase local to the neuron alone tells us nothing about the instantaneous beta phase in the other hemisphere; however , if we know the local ( absolute ) beta phase as well as the relative phase difference between the hemispheres , then we can calculate the distal ( absolute ) beta phase in the other hemisphere ( Figure 9A–B ) . Thus , if the mapping from the phase difference between left and right M1 to the spike rate of a neuron – the phase-difference-to-rate mapping – is non-uniform , then we can infer that distal phase information is informative about the spike rate of a neuron above and beyond the information gained by knowledge of the locally-generated beta phase ( that is , beta activity generated near the soma ) . Figure 9 shows an example of one such neuron ( sig128a in left M1 ) that exhibits a non-uniform phase-difference-to-rate mapping , as well as task-dependent changes in this mapping . Figure 9D shows the distribution of relative phase differences between the left and right hemispheres during BC , while Figure 9F shows the distribution of beta phase differences under MC . Unlike the distribution of ( single-channel ) absolute phases , which are almost always uniform , the distribution of relative phase differences is often peaked , indicating that the two signals are coupled , either directly or via connections to other , additional areas . While there is a slight shift in the peak of the distribution when moving from one task to the other , both distributions are centered around 0 radians , indicating a tendency to zero-lag phase coupling in both MC and BC sessions . In contrast to the relative stability of the phase-difference distribution across different tasks , the phase-difference-to-rate mappings in Figure 9C–D show that: 1 ) sig128a is sensitive to the relative beta phase difference between the hemispheres , above and beyond the effects of the local beta phase , and 2 ) the phase-difference-to-rate mapping reverses during MC . That is , during BC the cell spikes the least when the instantaneous phase difference is near its most probable value ( the distribution peak in Figure 9D ) , and increases its spike rate when the phase difference moves away from the peak of the distribution . In contrast , during MC the same cell spikes most near the peak of the phase-difference distribution and decreases firing when the phase difference moves to less probable values . Over the population of recorded neurons , 68 . 4% exhibited significant variation in their spike rates as a function of the macro-scale , inter-hemispheric beta phase differences during BC . In order to facilitate comparisons among all recorded neurons , we computed the phase difference between left and right M1 using the average signal from all LFPs in each 8 × 8 electrode array ( one array per area ) – that is , we generated one time series of instantaneous phase differences against which we can examine the activity of all neurons . Interestingly , the phase-difference-to-rate mapping is often stronger when the LFP signal from an electrode proximal to the neuron is used ( data not shown ) . However , using different pairs of LFPs for each neuron makes systematic comparisons across neurons more difficult . An alternative approach to multi-scale coupling – from macro- to meso- to micro-scale – is suggested by Figure S8 , which shows a dependence between the inter-hemispheric beta phase difference between left and right M1 , on the one hand , and the mean beta amplitude in each area , on the other ( Figure S8B ) . Furthermore , Figure S8D shows that the correlation between beta amplitudes recorded in left and right M1 is dependent on the phase difference between them; that is , for one value of the inter-hemispheric phase difference , the amplitude correlation is greater than 0 . 5 , but for other values it is near 0 . 1 ( Figure S8D ) . Thus , it is possible that inter-hemispheric phase differences may influence individual cells using local beta amplitude as an intermediate variable . Nonetheless , these results suggest that single cells may receive information about activity in distal locations – either directly through monosynaptic connections , indirectly through a chain of intermediate variables , or both . Furthermore , spatially averaging field potential activity over several millimeters – as is done here in order to generate a signal comparable to that recorded with the macroelectrodes employed in human electrocorticography – may result in a loss of useful information . That is , neurons may be sensitive to meso-scale spatial patterns in addition to the average activity in a cortical area , a possibility we explore below . The analyses above employ a strictly univariate approach – the univariate ( meso-scale ) signal representing the mean activity in M1 is generated by spatially averaging the individual LFPs from an 8 × 8 microelectrode array ( 3 . 5 mm × 3 . 5 mm; c . f . Figure 10A ) , or the univariate ( macro-scale ) signal of interhemispheric phase differences is extracted . However , this approach ignores any spatial patterns that may occur on the 8 × 8 microelectrode arrays , as well as any neuron-specific preference for different spatial patterns . Furthermore , issues related to direct coupling versus indirect coupling through intermediate variables are difficult to resolve with univariate methods . Thus , investigating spatial patterns requires the use of a multivariate approach . Here we employ a method similar to that used in [41] , but using complex multivariate Gaussian distributions in order to include both the amplitude and phase of multiple LFP signals . This approach uses fewer parameters per channel to characterize the relation between one LFP and the spike rate , but critically it captures the full pattern of covariances between channels and thus provides insight into the influence of meso-scale spatial patterns on the spiking of single cells . For example , Figure 10A shows a schematic of the 8 × 8 electrode array implanted in right M1 . Spikes from a single neuron are recorded on the bottom-leftmost electrode ( coordinates [1 , 1] ) . Groups of electrodes more and more distal to the electrode used to record neuronal spiking are indicated in color – that is , we consider progressively larger groups of electrodes from 4 electrodes ( blue ) , to 16 ( blue and green ) , to 36 ( blue , green , black ) , to 64 electrodes ( all colors ) . Phase coupling between electrodes is a function of inter-electrode distance; with distance d in mm , the von Mises concentration parameter κ between two 28-Hz filtered signals is given by κ = 2 . 67 - 0 . 4435*d . Figure 10B shows the measured and predicted spike rates using these different-sized groups , from 4 electrodes ( blue ) , to 16 ( green ) , to 36 ( black ) , to 64 ( red ) . Importantly , the range of the measured spike rate increases as more and more channels are included . That is , as more distal electrodes are included in the predictive model and it is applied to novel data , a better prediction is generated , both in terms of the range of the measured rate as well as the coefficient of variation ( r2 ) . This is despite the fact that fewer parameters are used per electrode to model the cross-level coupling than for the amplitude- and phase-to-rate mappings considered above . The fact that including spatial information ( in the form of more distal electrodes ) improves the spike-rate predictions for individual neurons suggests that cells may be sensitive to distinct spatiotemporal patterns of population or local network activity . Because of the stability of the within-task mapping from beta activity to single unit spiking , when given training and ( novel ) test data collected under the same task conditions , the spike density of individual cells can be predicted well for a large subset of neurons ( c . f . [41] ) . Most relevant for the current study , however , is the finding that cross-task training and testing is much less effective than within-task training and testing ( on novel data recorded under the same task conditions ) . Figures 10C–H show data from six example neurons where within-task predictions are good ( red ) but the cross-task predictions are poor ( blue ) . Importantly , this loss of predictability is not due to a lack of cross-level coupling ( CLC ) between neurons and distributed LFP signals in one task , since CLC holds in both MC and BC tasks . Rather , while CLC occurs in both tasks , there has been a shift in the mapping to the spike rate of an individual cell from the meso-scale , spatially-averaged beta phase and amplitude , the preferred meso-scale spatial patterns , and macro-scale inter-hemispheric phase differences . That is , there is still coupling across spatial and temporal scales , but the target patterns that a particular cell is sensitive to shift when moving from one task to another . Thus , for a given neuron the cross-level coupling model obtained via training under one type of task ( e . g . , MC ) may in fact hold little or no predictive value for the coupling observed in a different task ( such as BC ) .
Above we showed that the spiking activity of neurons is coupled to multiple aspects of the motor beta rhythm during two different tasks ( MC and BC ) , and that the form of this beta-to-rate mapping changes in a reversible , task-dependent way . For example , as beta power increases , a given neuron may increase spiking during MC but decrease spiking during BC , exhibit a reversible shift in the preferred phase of firing , or remap its sensitivity to relative phase differences between areas . This dependence on beta amplitude was well-fit by a sigmodial function ( Figure 4 ) , while the dependence of spiking on beta phase followed a cosine function ( Figure 5 ) , weighted by beta amplitude ( Figure 6 ) . These results expand on prior findings showing cross-level coupling ( CLC ) between spiking and LFP phase in multivariate signals [41] , here showing an additional , independent coupling to beta amplitude . Critically , this work shows that cells can exhibit task-dependent changes in this coupling . Importantly , the parameters describing this beta-to-rate mapping are stable across multiple datasets of the same task ( within-task stability ) but exhibit reliable changes when moving from one task to another ( cross-task diversity ) . Furthermore , we showed that the ensemble diversity of amplitude-to-rate and phase-to-rate mappings describes a set of discrete ensemble states , where each state is defined by the rank order of instantaneous spike rates . What are the implications of these empirical findings for different hypotheses about the oscillatory control of distributed networks , especially regarding local computation in a given area and long-range communication between areas ? First , there is the question of how the observed beta-to-rate mappings arise – presumably the spike activity of a subset of presynaptic cells is the origin of the amplitude- and phase-to-rate mappings for a given neuron . Rather than speculate on these origins , here we take it as given that the beta-to-rate mapping exists and instead ask what computations are now possible that are not possible or difficult if CLC is absent . We focus on two potential mechanisms that operate over different timescales: first , we consider the impact of CLC on rate-based winner-take-all ( WTA ) competition mediated by recurrent synaptic inhibition . Operating over a timescale of hundreds of milliseconds , modulation of WTA dynamics via the amplitude-to-rate mapping provides one link from cross-level coupling to functional neural computation . Second , operating over a timescale of tens of milliseconds , the phase-to-rate mapping biases ensemble spike timing such that some spike timing patterns are more likely than others . Through this route , cross level coupling may modulate robust temporal coding mechanisms such as synfire chain propagation . When evaluating different neurocomputational mechanisms , it is important to keep the anatomical facts clearly in mind in order to rule out mathematically elegant but biophysically implausible options . In this regard , the recurrent excitatory/inhibitory loops of local cortical circuits appear to provide an ideal platform for winner-take-all ( WTA ) dynamics [67] . Figure 11A presents a simplified schematic of a WTA module , where multiple input paths are converted into the activation of one output path via competitive di-synaptic inhibition . In this module , two excitatory cells , E1 and E2 ( red triangles ) , are reciprocally connected to an inhibitory cell ( blue circle ) that receives input from both E-cells . Both E-cells also receive independent excitatory input from outside the module . None of the cells inside the WTA module need have amplitude-to-rate mappings or any beta sensitivity whatsoever . Next , assume two cells outside the WTA module provide the external excitatory input , and that both of these cells have amplitude-to-rate mappings that intersect . For example , consider the purple and gold cells in Figures 11B–C , which have amplitude-to-rate mappings as shown in Figure S10 . For simplicity , assume these external cells providing WTA input are driven solely by their amplitude-to-rate mappings . Then for low beta amplitudes , the WTA cell E1 becomes active ( Figure 11B ) , whereas high beta amplitudes cause E2 to become active ( Figure 11C ) . In fact , the switch between E1 and E2 occurs at the beta amplitude value corresponding to the intersection of the amplitude-to-rate mappings for the purple and gold input cells . That is , the relative spike rate rank order of the cells providing input to the WTA module is transformed into tonic spiking along one of two possible output paths . Since the evidence presented here shows that within-task amplitude-to-rate mappings are stable , this binary output switch is tuned to a particular value of beta amplitude that is fixed for the duration of the task . Whenever beta amplitude sweeps through this value , this WTA switch changes state . By adding additional cells with amplitude-to-rate mappings that cross at other amplitude values , we can establish a linear , task-dependent sequence of binary WTA switches , each of which is tuned to or indexed by a different value of beta amplitude . Thus , each value of beta amplitude is associated with a binary vector that encodes the ensemble state . Why would this be useful ? First , recall the 12 cells shown in Figure 4A . On the one hand , there are 12 ! = 479001600 possible rank-ordered states for this set of neurons , corresponding to the number of permutations . The ability to generate sequences from such a large set of states would clearly prove computationally useful . However , it is unclear what biological mechanisms are available to quickly identify and activate an arbitrary state selected from the set of all possible ensemble states . On the other hand , if the 12 neurons have fixed baseline rates and flat amplitude-to-rate mappings , then state activation is not a problem since only one state is active at all times . Again , this case is not very computationally useful . In contrast to these extreme cases , an ensemble of neurons with a diversity of amplitude-to-rate mappings ( as shown in Figure 4A ) has both a variety of possible states ( defined by the amplitude-to-rate crossings ) , as well as a method for indexing each state ( every beta amplitude value corresponds to one particular ensemble state ) . More importantly , task-dependent remapping of the amplitude-to-rate functions provide the means to select a different set of ensemble states – where again each state is indexed by beta amplitude . That is , during one task such as BC , the continuous variation in beta amplitude maps to a discrete sequence of ensemble states ( 16 states for the 12 neurons shown in Figures 4A and S2A ) , while switching to another task such as MC maps the amplitude to a different sequence of ensemble states ( 24 states for the same set of neurons ) . In this view , across all tasks the continuous amplitude signal serves as an index function that establishes activation and transition probabilities for ensemble states . However , one task may require a different set of ensemble states than another – thus explaining the task-dependent remapping , as cross-level coupling parameters are tuned to evoke a desired set of ensemble states . In the example above , none of the 16 BC states or 24 MC states are shared across tasks ( Figure S2I ) . Task-dependent remapping thus balances the need for a diversity of ensemble states with the requirement of a simple mechanism for sequential state activation . Therefore , combining WTA dynamics with beta-to-rate mapping and remapping seems to provide a physiologically plausible mechanism for the dynamic linking of distinct sequences of ensemble states to a common , readily-accessible signal representing the overall level of population activity – namely , the beta rhythm . These ideas are consistent with the hypothesis that the functional role of the beta rhythm is to maintain the current computational state in a local network , protecting the local population against irrelevant or contradictory input [16] . That is , beta power remains high if no change in the local network state is needed , or if unwanted changes to local network state must be actively extinguished . Similarly , beta power drops when the local network state must change . The arrival of important but unexpected input may increase or decrease beta power , depending on context and task demands . In this view , beta is an active coordinating rhythm that helps to maintain or release selected patterns of ensemble activity . It is intriguing to speculate that task switching requires remapping coupling parameters in order to evoke a pre-learned sequence of WTA states , while learning involves optimization over the space of WTA sequences in a search for those sequences that prove most task effective . One prediction of this hypothesis is that ensemble amplitude-to-rate mappings will exhibit much more variability during learning than either before or after . The combination of WTA dynamics together with heterogeneous amplitude-to-rate mappings across an ensemble provides a specific and testable mechanism through which the beta rhythm could accomplish this goal of dynamic coordination . Independent of possible functional roles played by the amplitude-to-rate mapping , phase-to-rate mappings may shift the relative probabilities of precisely-timed spike sequences . Simulation studies show that polychronous groups – sets of cells where activity propagates due to precise spike timing relations – can serve as the building blocks for cognitive operations such as working memory [68] , and exhibit activity-dependent growth and decay useful for learning and pattern recognition [69] . Empirically , Havenith et al . [7] showed that relative spike timing in visual cortex reflects properties such as stimulus orientation . Importantly , given inter-connected pools of neurons , synchronous propagation of activity is more stable than asynchronous propagation . In fact , propagating synfire chains yield stable and robust spiking precision in the millisecond range that supports the self-stabilization of synfire chain activity [70] . That is , given the right starting conditions , initially weak synfire chains ( with few active members or poor synchronization ) can recruit additional members and reduce spike-timing variance across the group . However , slightly different initial conditions may force a synfire chain to cross a dynamical systems separatrix between attractors , forcing the synfire chain to quickly decay [71] . Since phase-to-rate mappings can influence spike timing , strong phase-to-rate mappings can increase the likelihood of some synfire chains while rendering others less likely . Since beta amplitude appears to act as a gain control mechanism for the strength of the phase-to-rate mapping , the influence of beta on the probability of different spike sequences can be adjusted by changing beta power . However , Figure 6 shows that a fixed change in beta amplitude will have a differential response on different cells , with some strongly increasing their phase preference while others show only moderate changes . Thus , the mapping from beta amplitude to spike sequence probabilities is not a simple one , but depends on the diversity of CLC parameters that hold across the population . Finally , task-dependent remapping of the preferred phase ( c . f . Figures 5I , J ) provides a mechanism for the selective and task-dependent control of synfire chain activation and propagation . That is , during a given task the relative probabilities of a set of ( function-specific ) multi-neuron spike sequences can be controlled via adjustments in beta amplitude , while switching to another task involves a remapping of CLC parameters in order to call a different set of spike sequences into action . Since motor cortical function involves both rate modulation as well as spike synchronization [72] , [73] , a mechanism to selective control synchronization while leaving rate modulation unchanged may prove useful to a system controlling distributed networks . While the amplitude- and phase-to-rate mappings appear most relevant to local computation within a given cortical area , the phase-difference-to-rate mapping may play a role in the regulation of long-range communication between areas . According to the communication through coherence ( CTC ) hypothesis , the effective gain between interacting areas is a function of the phase difference between them [21] , [51] , [74] . It is difficult to see how the brain could implement CTC control systems without the use of neurons that detect phase differences between areas , on the one hand , as well as neurons than can evoke shifts in the relative phase between distal areas , on the other . Neurons that could serve as phase difference detectors and effectors appear to be fundamental elements required by any distributed system of oscillatory network control . Furthermore , hierarchical predictive coding models suggest that the gamma rhythm is indicative of bottom-up feed-forward processing , while the alpha and beta rhythms serve as signatures of top-down feedback influence [75]–[78] . Distinct phase-difference-to-rate mappings that operate at these frequencies appear to be one way to control the relative balance of feedforward and feedback processing . The phase-difference to amplitude-envelope-correlation relationship shown in Figure S8D appears to support the communication through coherence hypothesis , but further studies targeting the role of spiking neurons in long-range interactions are required to clarify their role in the oscillatory control of distributed networks . Prior work studying neural dynamics in motor cortex has tended to focus on the correlation between spiking activity and “external” factors ( e . g . movement velocity , environmental state , behavior-dependent sensory feedback , etc ) . In contrast , this study focused on “internal” factors that arise from spontaneous , ongoing brain activity – including beta amplitude and phase within an area , or the difference in beta phase between areas . Specifically , we showed that most neurons exhibited a sigmoid dependence on beta amplitude ( considered alone; Figure 4 ) , a cosine dependence on beta phase ( considered alone; Figure 5 ) , and that beta amplitude provided a quadratic gain control for the beta phase preference ( Figure 6 ) . What is the relationship between these “external” and “internal” factors ? Figure 3 provides an example of external and internal tuning for one example neuron , showing how input variables can be mapped to a predicted rate , which can then be compared to a measured rate . For example , Figure 3D shows how time-in-trial and target ID can be mapped to a predicted rate ( color ) , while Figure 3E shows how this predicted rate compares to the rate that actually occurs . Similarly , Figure 3I–J show this for beta amplitude and phase; these figures show that the range of the predicted rate generated from the beta-to-rate mapping ( rinternal ) is about half that of the range of the predicted rate generated from trial information ( rexternal ) . The sum of these terms ( r = rexternal + rinternal ) often has a larger range than either rexternal or rinternal alone . However , this sum assumes that rexternal and rinternal are independent – an assumption that is not appropriate for many neurons . For example , while Figure 7G–H shows neurons where target direction appears independent of the phase-to-rate modulation depth , Figure 7E–F show examples where there is a clear interaction between internal and external factors . The focus of this study was to investigate the dependence of spiking on internal factors , and to determine if this dependence changes from one task to another . Determining the relation between internal and external factors will require further investigation . Nonetheless , the majority of neurons show a dependence on “internal” beta-related factors that is not mediated by external factors such as direction tuning ( Figure S3 ) . A related concern is that the observed changes in CLC are more directly linked to bottom-up demands related to the trial substages ( e . g . hold vs . movement period ) than to top-down modulation associated with the task context . Figure S5 addresses this concern by directly comparing endogenous and exogenous factors; the fact that cross-task , within-stage differences are larger than within-task , cross-stage differences indicates that task context is a factor in determining neuronal responses related to CLC . That is , it appears that cells are influenced both by bottom-up , exogenous input related to the processing demands of the different trial substages as well as by top-down , endogenous input related to the maintenance of task context and rule selection . An interesting aspect of this analysis has been the observation of the strong heterogeneity of neuronal sensitivities to different types of input , considering external vs . internal factors or top-down vs . bottom-up aspects of the experimental demands , compared to the stability of the average population responses . For example , Figure S3A shows the baseline firing rates for each neuron during BC and MC , and makes it clear that many neurons exhibit large task-dependent shifts in the baseline spike rate . The average spike rate over the population , however , is relatively unchanged ( red and blue lines , Figure S3A ) . That is , with a shift in task the neuronal ensemble seems to reassign firing rates around a constant population mean rate . Similarly , the amplitude-to-rate and phase-to-rate mappings computed using spikes from all neurons ( average population mappings ) do not show the strong task-dependent shifts seen in the mappings of individual neurons . Therefore , we would predict that electrophysiological measures that depend on average ensemble activity , such as coupling between beta and the broadband ECoG signal [79] , will be less likely to exhibit strong task-dependent changes than will individual neurons . Finally , the empirical findings reported here are consistent with the hypothesis that dynamic changes in coupling between multiple spatial and temporal scales provide a simple mechanism to bias functional network activity [80] . In particular , coupling between single neurons and the motor beta rhythm exhibits several properties that appear positioned to influence local cortical computation – namely , the phase-regulation of relative spike timing on a scale of tens of milliseconds and the amplitude-regulation of winner-take-all dynamics within neuronal ensembles occurring on a scale of hundreds of milliseconds . Similarly , long-distance communication appears to be modulated by the relative phase difference between areas . The presence of neurons that are sensitive to these properties could provide a mechanistic route for this information about relative phase differences to be detected and actively used in the dynamic regulation of large-scale network activity . While future studies employing casual intervention will be required to fully test the functional role of different oscillatory rhythms , here we have shown that the mapping from beta activity to firing rate changes in a reversible , task-dependent way . Given that beta oscillations are generated by the coordinated population activity of hundreds of thousand cells involved in a distributed network that spans both hemispheres [11] , the results presented here suggests that the relationship of multiscale coupling between single neurons and larger networks is flexible and can be dynamically remapped in order to support new functional roles . | How is the functional role of a particular neuron established within an ensemble ? The concept of a neural tuning curve – the mapping from input variables such as movement direction to output firing rate – has proven useful in investigating neural function . However , prior work shows that tuning curves are not fixed but may be remapped as a function of task demands – presumably via high-level mechanisms of cognitive control . How is this accomplished ? Brain rhythms may play a causal role in this process , but the coupling of single cells to network activity remains poorly understood . We investigated the coupling between rhythmic beta activity and spiking as macaques performed two different tasks . This coupling can be described in terms of a function that maps oscillatory amplitude and phase to instantaneous spike rate . Similarly to direction tuning , this “internal” tuning curve also exhibits task-dependent changes . We characterize these changes across a large ensemble of simultaneously-recorded cells , and consider some of the neuro-computational implications presented by cross-level coupling between single cells and large-scale networks . In particular , relative to the slow time-scale of behavior , the observed beta-to-rate mappings may prove useful for modulating winner-take-all dynamics on intermediate time-scales and relative spike timing on fast time-scales . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [
"computer",
"science",
"computer",
"modeling",
"neural",
"networks",
"computational",
"neuroscience",
"biology",
"computing",
"methods",
"neuroscience",
"neurophysiology",
"coding",
"mechanisms"
] | 2012 | Task-Dependent Changes in Cross-Level Coupling between Single Neurons and Oscillatory Activity in Multiscale Networks |
A critical determinant in chronic gammaherpesvirus infections is the ability of these viruses to establish latency in a lymphocyte reservoir . The nuclear factor ( NF ) -κB family of transcription factors represent key players in B-cell biology and are targeted by gammaherpesviruses to promote host cell survival , proliferation , and transformation . However , the role of NF-κB signaling in the establishment of latency in vivo has not been addressed . Here we report the generation and in vivo characterization of a recombinant murine gammaherpesvirus 68 ( γHV68 ) that expresses a constitutively active form of the NF-κB inhibitor , IκBαM . Inhibition of NF-κB signaling upon infection with γHV68-IκBαM did not affect lytic replication in cell culture or in the lung following intranasal inoculation . However , there was a substantial decrease in the frequency of latently infected lymphocytes in the lung ( 90% reduction ) and spleens ( 97% reduction ) 16 d post intranasal inoculation . Importantly , the defect in establishment of latency in lung B cells could not be overcome by increasing the dose of virus 100-fold . The observed decrease in establishment of viral latency correlated with a loss of activated , CD69hi B cells in both the lungs and spleen at day 16 postinfection , which was not apparent by 6 wk postinfection . Constitutive expression of Bcl-2 in B cells did not rescue the defect in the establishment of latency observed with γHV68-IκBαM , indicating that NF-κB–mediated functions apart from Bcl-2–mediated B-cell survival are critical for the efficient establishment of gammaherpesvirus latency in vivo . In contrast to the results obtained following intranasal inoculation , infection of mice with γHV68-IκBαM by the intraperitoneal route had only a modest impact on splenic latency , suggesting that route of inoculation may alter requirements for establishment of virus latency in B cells . Finally , analyses of the pathogenesis of γHV68-IκBαM provides evidence that NF-κB signaling plays an important role during multiple stages of γHV68 infection in vivo and , as such , represents a key host regulatory pathway that is likely manipulated by the virus to establish latency in B cells .
Murine gammaherpesvirus 68 ( γHV68 ) shares many genetic and biologic properties with its human counterparts , Epstein-Barr virus ( EBV ) and Kaposi sarcoma–associated herpesvirus ( KSHV or HHV-8 ) . For example , it has been shown for both EBV and γHV68 that long-term latency is maintained in memory B cells [1–3] . Identifying the host-dependent requirements for gaining access to the latency reservoir is an important step toward understanding how the virus modulates the host to establish a chronic infection . Such virus–host interactions may lead to dysregulation of normal cellular controls , increasing the risk for the development of lymphomas and other tumors etiologically associated with gammaherpesvirus infections [4 , 5] . Nuclear factor ( NF ) -κB transcription factors are key regulatory molecules of genes involved in innate and adaptive immunity . The absence of particular NF-κB subunits , or upstream regulatory molecules , can result in defects in B-cell development and functions such as activation-induced proliferation ( reviewed in [6–8] ) . The maturation of B cells and survival in the periphery involve NF-κB–mediated upregulation of antiapoptotic bcl-2 , bcl-xL , and bfl-1/A1 genes [7 , 9] . The proliferative response of B cells to stimulation requires NF-κB–mediated upregulation of c-Myc , cyclins , and CDK4 [7 , 10 , 11] . The NF-κB family of transcription factors is comprised of the subunits p65 ( RelA ) , cRel , RelB , p50 ( NF-κB1 ) , and p52 ( NF-κB2 ) that form dimers to mediate sequence-specific regulation of gene expression upon activation . Dimers of NF-κB subunits are retained in the cytoplasm by inhibitory IκB molecules . Cellular activation leads to proteosomal-dependent degradation of IκB molecules and translocation of NF-κB dimers to the nucleus . The engagement of cell surface receptors such as the B-cell receptor , receptors for inflammatory cytokines ( e . g . , tumor necrosis factor [TNF] α ) , and Toll-like receptors lead to the release of p50:cRel and p50:relA dimers through the classic pathway . The phosphorylation and subsequent degradation of the inhibitor of NF-κB , IκBα , are mediated by the IκB kinase ( IKK ) complex ( IKK1 , IKK2 , and NEMO ) . In contrast , engagement of receptors such as the lymphotoxin β receptor and the B-cell–activating factor of the TNF family ( BAFF ) leads to the release of p52:RelB dimers via the alternative IKK1/NIK-dependent pathway ( reviewed in [12] ) . Engagement of primary B cells with CD40L leads to the activation of NF-κB via both the classic and alternative pathways [11] . Given the critical role that NF-κB plays in B-cell biology , lymphotropic gammaherpesviruses have evolved multiple strategies to modulate NF-κB activity in infected cells . For example , the latent membrane protein 1 ( LMP-1 ) encoded by EBV is expressed during the growth latency program and is essential for EBV immortalization of primary B cells [13 , 14] . LMP-1 drives NF-κB activation and is able to partially restore CD40-dependent functions in transgenic mice , indicating that it functions as a constitutively active CD40 receptor [15–17] . In the case of KSHV latency , the virus-encoded vFLIP has been shown to be essential for the proliferation and survival of primary effusion lymphoma cells , driving NF-κB activation via interactions with the TNF receptor–associated factors TRAF2 and TRAF3 [18–20] . Thus , it is likely that manipulation of NF-κB signaling is critical for the establishment and maintenance of gammaherpesvirus latency in vivo; however , to date this remains untested . γHV68 infection of mice represents a tractable animal model in which both viral and host factors that modulate acute replication , latency , and reactivation can be addressed . Intranasal infection with γHV68 results in acute virus replication in the lungs , followed by B-cell–dependent spread of virus to the spleen , in turn followed by acute replication and establishment of latency in the spleen [21–25] . Subsequent to immune clearance of acute virus replication , multiple cell types , including lung epithelial cells , B cells , dendritic cells , and macrophages , are found to harbor the viral genome in the absence of preformed infectious virus [23 , 26–29] . Splenic B cells infected with γHV68 undergo proliferation and bear surface markers of cells that participate in germinal centers , sites of affinity maturation , and differentiation into long-lived memory B cells and antibody-producing plasma cells [2 , 27 , 30–32] . Notably , isotype-class–switched memory B cells are the predominant reservoir for latent γHV68 infection at late timepoints postinfection [2 , 3] . Thus , γHV68 pathogenesis has strong parallels with EBV infection in humans , where EBV latency is characterized by distinct latency programs involving differential expression of specific viral genes that modulate host signaling pathways to facilitate the proliferation of naïve cells into lymphoblasts , followed by differentiation of these lymphoblasts into memory B cells [13] . As with KSHV , γHV68 lacks clear homologs of the EBV latency-associated antigens , yet it has likely evolved a strategy for promoting the activation , proliferation , and differentiation of newly infected B cells into the memory B-cell reservoir . NF-κB likely plays a critical role in multiple aspects of γHV68 pathogenesis . The role of one upstream activator of NF-κB , CD40 , has been examined in the context of an intact immune response . Blackman et al . [3] infected CD40+/CD40− mixed bone marrow chimeric mice and reported that CD40-deficient B cells latently infected with γHV68 rapidly waned compared to latency in CD40-sufficient B cells . This suggests that NF-κB activation upon CD40L stimulation plays a role in the maintenance of latency in B cells [3] . Finally , replication of γHV68 in cell culture is blocked by overexpression of p65 , and it has been proposed that NF-κB activation may promote the establishment of latency by inhibiting lytic cycle initiation [33] . Here we sought to address the role of NF-κB signaling in vivo using virus-directed inhibition of NF-κB activation . This experimental approach avoids the global immune dysfunction present in NF-κB knockout mice , which would not allow the role of NF-κB activation in the establishment of viral latency in B cells to be readily distinguished from its role in the immune response and subsequent control of γHV68 infection . To this end , we report the generation and characterization of a recombinant γHV68 expressing a mutant form of the NF-κB inhibitor IκBα , IκBαM , that functions as a superrepressor of NF-κB activation . We find that while dispensable for in vitro replication , NF-κB activation is a critical host determinant for γHV68 latency in vivo .
To investigate the consequences of inhibiting the NF-κB signaling pathway during the infection of mice with γHV68 , yet avoid global alterations in the immune response , we generated a recombinant γHV68 that would inhibit NF-κB signaling only in infected cells . To accomplish this , we used a mutant form of IκBα , IκBαM , that contains serine-to-alanine substitutions at amino acids 32 and 36 , thus preventing phosphorylation , proteasomal degradation , and release of NF-κB dimers to the nucleus following appropriate upstream stimuli [34] . This “superrepressor” functions as a potent transdominant inhibitor of NF-κB activation in cell culture [35–37] and in vivo [38] . We inserted an IκBαM expression cassette , driven by the HCMV major immediate-early promoter ( MIEP ) , into the intergenic region between ORFs 27 and 29b of γHV68 ( Figure 1A ) . It has previously been shown that insertion of a HCMV MIEP-driven Cre recombinase expression cassette [32] , insertion of a transposon [39] or the mutagenesis of ORF28 in this region [40] does not impair virus replication in vitro [39 , 40] or in vivo [40] , and has no discernable impact on the establishment of , or reactivation from , splenic latency in C57BL/6 mice [32 , 40] . The IκBαM . 1 and IκBαM . 2 recombinant viruses were generated as described in Materials and Methods through allelic exchange in Escherichia coli utilizing a γHV68-BAC [41] followed by removal of BAC sequences through Cre-mediated recombination in Vero-Cre cells . We generated two independent recombinant γHV68 viruses , IκBαM . 1 and IκBαM . 2 , which contain the IκBαM expression cassette in either the rightward or leftward orientation within the viral genome , respectively ( Figure 1A ) . In addition , two independent marker rescue viruses were generated through allelic exchange with IκBαM . 1 , one of which ( IκBαM . MR ) was fully characterized here . As shown in Figure 1B , Southern analysis of BamHI-digested viral DNA isolated from extracellular virions , followed by hybridization with a [32P]-labeled probe specific for IκBαM , detected the IκBαM insert only in the IκBαM . 1 and IκBαM . 2 recombinant viruses ( revealing the expected product sizes of 5 , 477 bp and 5 , 127 bp , respectively ) . A probe containing a portion of ORF27 and the ORF27-29b intergenic region identified the expected 4 , 416-bp fragment in HindIII-digested wild-type ( WT ) and IκBαM . MR viral DNAs . As expected , the introduction of a new HindIII site within the IκBαM insert resulted in a reduction in the sizes of the hybridizing fragments for the recombinant viruses harboring the IκBαM expression cassette in the rightward ( IκBαM . 1; 1 , 682 bp ) or leftward ( IκBαM . 2; 2 , 321 bp ) orientation ( Figure 1B ) . The ability of the IκBαM-expressing viruses to inhibit NF-κB activation in the context of virus infection was examined using an NF-κB–responsive reporter construct in infected cells . NIH 3T12 fibroblasts were infected with WT γHV68 at a multiplicity of infection ( MOI ) of 1 or 10 for 6 h followed by transfection with an NF-κB–dependent luciferase construct , in the presence or absence of an expression construct for MEKK1 . At 24 hours postinfection ( hpi ) , cells were harvested and luciferase activity was quantitated . Notably , infection with WT γHV68 activated the NF-κB–dependent reporter construct in a dose-dependent manner to levels nearly 8-fold over uninfected samples ( Figure 2A ) . This activation synergized with MEKK1 activation to further enhance NF-κB activation of the reporter construct over 100-fold compared to mock infected samples ( Figure 2A ) . In an independent experiment , the synergistic induction of the NF-κB–responsive reporter construct by MEKK1 and γHV68 infection was used to examine the recombinant IκBαM-expressing viruses . Consistent with the results shown in Figure 2A , infection of WT or IκBαM . MR virus enhanced MEKK1 activation of the NF-κB–responsive reporter construct approximately 10-fold above that observed in mock infected cells ( Figure 2B ) . In contrast , both IκBαM-expressing viruses strongly inhibited MEKK1 activation of the NF-κB–responsive reporter construct ( Figure 2B ) . Analysis of NF-κB activity in cells infected with a recombinant γHV68 harboring a CMV MIEP-driven CRE-recombinase expression cassette in the intragenic region between ORFs 27 and 29b revealed no inhibition of NF-κB activation , arguing against interference with NF-κB activation due to insertion of the CMV MIEP-driven expression cassette in this region of the viral genome ( unpublished data ) . Taken together , these data directly demonstrate effective inhibition of NF-κB activation by the recombinant viruses expressing IκBαM , in response to both virus infection and MEKK1-driven activation . To further examine activation of NF-κB upon γHV68 infection of murine fibroblasts , nuclear extracts were prepared from mock infected NIH 3T12 cells , as well as from NIH 3T12 cells infected with either WT γHV68 , IκBαM . 1 , or IκBαM . MR . Electrophoretic mobility shift analysis of nuclear extracts incubated with a consensus NF-κB binding site revealed the induction of NF-κB binding activity in WT and IκBαM . MR virus–infected cells but not in the IκBαM . 1-infected cells ( Figure 2C ) . Furthermore , the induced NF-κB complex could be competed by an unlabeled double-stranded oligonucleotide probe containing a consensus NF-κB binding site but not by a competitor containing a mutated NF-κB binding site ( Figure 2C ) . Electrophoretic mobility supershift analyses of the induced NF-κB complex , using antibodies targeted against specific NF-κB family members , demonstrated that the induced complex contains p65 but does not appear to contain either c-rel or p50 ( unpublished data ) . The ability to easily generate high-titer viral stocks from the initial transfection of γHV68-IκBαM BAC DNA into Vero-Cre cells indicated that insertion of the IκBαM expression cassette into the viral genome does not appreciably impair viral replication in vitro . To quantitatively assess the role of NF-κB signaling during lytic replication , we compared replication of IκBαM . 1 and IκBαM . MR in primary mouse embryonic fibroblasts ( MEFs ) . MEFs were infected for single-step ( MOI = 5 . 0 ) and multistep ( MOI = 0 . 05 ) growth assays , followed by analysis of virus growth at various times postinfection by plaque assay on NIH 3T12 cells . Notably , the kinetics of virus replication in the single-step ( Figure 3A ) and multistep ( Figure 3B ) growth curves were nearly indistinguishable between IκBαM . 1 and IκBαM . MR viruses . The replication of IκBαM . 1 , and IκBαM . 2 , IκBαM . MR , and WT γHV68 in NIH 3T12s was also compared in a multistep growth analysis ( Figure 3C ) . Consistent with the analyses shown in Figure 3A , all viruses replicated with similar kinetics and to comparable virus titers across the 96-h time course ( Figure 3C ) . These data demonstrate that the insertion of the IκBαM expression construct into the viral genome ( in the intragenic region between ORFs 27 and 29b ) , and the resulting inhibition of NF-κB signaling by IκBαM expression , did not negatively impact γHV68 replication in vitro . Upon intranasal infection of mice , γHV68 undergoes a lytic replicative expansion in the lung epithelium prior to trafficking and seeding of other target organs and tissues , such as the spleen . The ability of the recombinant viruses expressing IκBαM to replicate in the lungs of C57BL/6 ( BL6 ) mice during this acute phase was examined ( Figure 3D ) . At days 4 and 9 , following intranasal inoculation with 1 , 000 PFU of virus , lungs were harvested , disrupted and virus titers determined . Acute viral titers in mice infected with the IκBαM . 1 and IκBαM . 2 recombinant viruses were comparable to that observed in mice infected with either WT virus or IκBαM . MR ( Figure 3D ) . Thus , we conclude that NF-κB activation is dispensable for productive replication in the lungs during the acute phase of virus infection . Interestingly , productive replication of the IκBαM-expressing virus following low-dose intranasal inoculation was slightly impaired when compared to IκBαM . MR , although replication of both the IκBαM . MR- and the IκBαM-expressing virus in the spleen was very low ( or not detectable ) at all times postinfection ( Figure 3E ) . We have previously reported that low-dose intranasal inoculation ( 40 PFU ) leads to delayed seeding and clearance of acute virus replication in the lungs and the spleen compared to high-dose inoculation ( 4 × 105 PFU ) [25] . However , IκBαM . 1 was observed to replicate with normal replication kinetics in the lung ( Figure 3D ) , and the low level of lytic replication of IκBαM . 1 in the spleen provided no evidence for delayed clearance of acute replication upon infection with IκBαM . 1 . As discussed below , a defect in acute IκBαM . 1 replication in the spleen was also noted at day 9 post intraperitoneal inoculation and indicates a role for NF-κB during lytic expansion in the spleen . Since impairment of NF-κB signaling had no effect on virus replication in the lung , yet slightly impaired acute replication in the spleen , we next examined the role of NF-κB signaling in the establishment of latency , as well as reactivation from latency , in bulk splenocytes and specific splenocyte subpopulations at early times postinfection . Mice were infected via intranasal inoculation with 1 , 000 PFU of IκBαM . 1 , IκBαM . 2 , IκBαM . MR , or WT γHV68 . At 15 to 16 days postinfection ( dpi ) , splenocytes were isolated and the frequency of viral genome–positive cells was determined by limiting-dilution PCR analysis ( Figure 4A ) . Unsorted , bulk splenocytes were found to harbor viral genome at a frequency of one in 3 , 646 and one in 5 , 570 cells in mice infected with either the IκBαM . 1 and IκBαM . 2 recombinant virus , respectively . This was a significantly lower frequency than that observed in mice infected with either WT γHV68 ( one in 138 cells ) or IκBαM . MR virus ( one in 75 cells ) ( Figure 4A , summarized in Table 1 ) . This approximately 50-fold reduction represented a greater than 96% decrease in the frequency of latently infected cells in mice infected with the IκBαM-expressing recombinant viruses compared to WT and marker rescue control viruses , indicating that inhibition of NF-κB signaling has a substantial impact on the ability of γHV68 to establish latency in splenocytes following intranasal infection . To determine the ability of splenocytes harboring the IκBαM virus to reactivate virus , we quantitated ex vivo reactivation of intact cells by limiting-dilution analysis on MEFs as described in Materials and Methods . Splenocytes from WT and IκBαM . MR-infected mice reactivated at a frequency of one in 3 , 102 and one in 3 , 563 , respectively ( Figure 4B ) . The frequency of reactivation of bulk splenocytes from mice infected with IκBαM . 1 and IκBαM . 2 was significantly lower than control virus reactivation and required a mathematical extrapolation of the best-fit curve , leading to estimated frequencies of one in 180 , 936 and one in 300 , 788 for IκBαM . 1 and IκBαM . 2 , respectively ( Figure 4B ) . The amount of preformed infectious virus detected by plating mechanically disrupted cells in parallel was negligible for all samples . The reactivation frequencies of the IκBαM recombinant viruses are approximately 50-fold lower than the frequency of viral genome–positive cells , in close agreement with the decrease in the frequency of viral genome–positive splenocytes . Thus , the efficiency of virus reactivation is similar for cells infected with the IκBαM recombinant viruses and the control viruses . It is unclear whether the residual levels of latency observed in mice infected with the IκBαM expressing viruses represent the infection of a population of cells that do not require NF-κB activation for establishment of γHV68 latency or whether they arise from latently infected cells in which the levels of expression of IκBαM were not sufficient to block NF-κB activation . Unfortunately , it is not possible at this time to distinguish between these alternatives . We also considered the possibility that infection with the IκBαM-expressing viruses could give rise in vivo to a revertant virus that has lost the ability to express a functional IκBαM . To address this possibility , independent isolates of virus ( IκBαM . R1 and IκBαM . R2 ) were recovered from ex vivo reactivation of splenocytes harvested from IκBαM-infected mice and analyzed for the presence of the IκBαM expression cassette . Southern blot analyses revealed that these viral genomes had retained the IκBαM expression cassette ( unpublished data ) . More important , the viruses recovered from reactivated splenocytes also retained their ability to inhibit NF-κB activation in NIH 3T12 fibroblasts ( Figure 2B ) . Therefore , the small population of cells that is viral genome–positive appears to harbor virus capable of expressing IκBαM . To examine the role of specific splenocyte cell populations , bulk splenocytes were separated by fluorescence-activated cell sorting ( FACS ) into B-cell and non–B-cell populations utilizing CD19 as a marker for B cells . This resulted in cell population of greater than 93% purity for each subset ( unpublished data ) . As shown in Figure 4C and summarized in Table 1 , the CD19+ B-cell population was found to harbor viral genome–positive cells at a similar frequency as the bulk splenocyte population following WT γHV68 infection . This is consistent with previous reports that B cells comprise the major reservoir of γHV68 latency in the spleen [2 , 27 , 42 , 43] . As observed with bulk splenocytes , CD19+ B cells from mice infected with IκBαM . 1 had a drastic reduction in the frequency of latently infected cells ( one in 4 , 093 ) compared to those from mice infected with WT ( one in 98 ) . This 40-fold difference represents a 98% reduction in the establishment of latency upon NF-κB inhibition , which was further substantiated by an estimated 98% reduction in the frequency of CD19+ cells reactivating virus in IκBαM . 1-infected mice compared to mice infected with WT ( Figure 4D ) . The frequency of latently infected CD19− non–B cells was also diminished in mice infected with IκBαM . 1 by 6-fold ( 84% reduction ) compared to those infected with WT virus ( Table 1 ) . At this point , it is unclear whether the diminished levels of latency in the non–B-cell population reflect a dependence on B-cell latency for establishment of latency in the non–B-cell reservoirs or whether there is a role for NF-κB activation in the establishment of latency in the non–B-cell reservoirs . Regardless , these data indicate that NF-κB signaling is critical for the efficient establishment of latency in splenic B cells following intranasal inoculation . Latency in WT-infected mice is characterized by a peak in viral genome–positive splenocytes at 16 dpi , a rapid contraction of latently infected splenocytes that is apparent by 6 wk postinfection , followed by a more gradual rate of decline throughout later time points as the virus reaches a steady-state level in memory B cells [2] . However , some γHV68 mutants that exhibit significant defects in the establishment of splenic latency ( measured at 16 dpi ) fail to contract at the same rate as WT virus and , as such , the frequency of latently infected splenocytes often approaches the level observed with WT virus infection at later times postinfection [44 , 45] . We therefore examined the frequency of latently infected splenocytes at 42 to 49 dpi ( Figure 5A ) . In mice infected with IκBαM . MR , the frequency of viral genome–positive , unsorted splenocytes dropped 16-fold at 42 to 49 dpi from that observed at day 16 , similar to previous reports with WT γHV68 [2 , 25] . In contrast , the frequency of unsorted splenocytes latently infected with IκBαM . 1 decreased only 2-fold . Therefore , the frequency of viral genome–positive unsorted splenocytes infected with IκBαM . 1 ( one in 7 , 017 ) were only 3-fold lower than WT γHV68 ( one in 2 , 210 ) by 42 to 49 dpi ( p = 0 . 0146 ) ( Figure 5A ) . As B cells represent the major long-term reservoir of viral latency in the spleen , the frequencies of genome-positive CD19+ B cells for IκBαM . 1 ( one in 11 , 138 ) and WT virus ( one in 2 , 785 ) were very similar to the frequency of bulk splenocytes , where only a 4-fold difference between the frequencies of IκBαM and WT virus was apparent by 42 to 49 dpi ( p = 0 . 0198 ) ( Figure 5A ) . We hypothesized that the dissimilar rates of contraction of latently infected splenocytes between IκBαM . 1 and WT virus could involve differences in cell turnover due to lytic events and/or the reseeding of latency in the spleen by virus arising from in vivo reactivation ( between day 16 and later time points ) [46] . To assess whether virus reactivation and reseeding of latency might be contributing to the results observed at 42 to 49 dpi , we administered the antiviral drug cidofovir to mice using a drug dosage demonstrated in previous studies to effectively inhibit acute virus replication and seeding of the spleen by day 16 after intranasal infection of C57Bl/6 mice [47 , 48] . In this analysis , the frequency of viral genome–positive cells in control mice ( no cidofovir ) infected with the IκBαM . 1 was 9-fold lower at 39 dpi compared to control mice infected with IκBαM . MR ( Figure 5B ) . Surprisingly , the frequency of cells harboring viral genome in mice treated with cidofovir was nearly indistinguishable from that of the untreated mice infected with either IκBαM . 1 or IκBαM . MR . This argues that the decline in latency of the IκBαM-expressing virus or the control IκBαM-MR virus between 16 and 39 dpi does not involve any significant contribution from virus reactivation and reseeding of splenic latency reservoirs and thus may reflect differences in trafficking of latently infected cells ( see Discussion ) . Bcl-2 is an antiapoptotic molecule whose expression is regulated by NF-κB . Notably , B-cell–specific expression of Bcl-2 in transgenic mice overcomes NF-κB–dependent requirements for mature B-cell development and survival in the periphery but does not restore B-cell proliferation upon mitogen activation [6] . To address the potential role of NF-κB as a factor critical for peripheral B-cell survival during γHV68 latency , we examined whether constitutive expression of Bcl-2 in B cells could rescue the establishment of latency in mice infected with IκBαM . 1 . On 16 dpi , there was an approximately 2-fold increase in the frequency of viral genome–positive unsorted splenocytes in Bcl-2 mice infected with IκBαM . MR compared to the frequency in infected C57BL/6 nontransgenic mice ( Figure 6A and Table 1 ) . However , there was no dramatic increase in the frequency of latency establishment in Bcl-2 mice infected with IκBαM . 1 compared to infected C57BL/6 mice ( Figure 6A and Table 1 ) . Thus , the frequency of bulk , unsorted splenocytes from Bcl-2 mice infected with IκBαM . 1 ( one in 2 , 577 ) was 70-fold lower than the frequency of viral genome–positivity found in IκBαM . MR-infected mice ( one in 37 ) ( Figure 6A ) . This correlated with an absence of splenomegaly in IκBαM . 1-infected mice as determined by spleen weight compared to uninfected mice ( 1 . 1-fold increase ) , while there was notable splenomegaly in the IκBαM . MR-infected mice ( 2 . 2-fold increase ) ( unpublished data ) . Bcl-2 mice are characterized by a higher proportion of B cells in lymphoid organs , which represented 72% of the total splenocytes at 16 dpi in these studies . To further characterize latency in these transgenic mice , splenocytes harvested at 16 dpi were separated into CD19+ and CD19− populations . As expected , the frequencies of CD19+ B cells harboring virus for both IκBαM . 1 ( one in 2 , 168 ) and IκBαM . MR ( one in 39 ) ( Figure 6A ) were nearly identical to the frequencies observed with bulk splenocytes . Since B cells from Bcl-2 transgenic mice are known to survive for a longer period of time in cell culture upon explant , we assessed whether latently infected splenocytes from Bcl2 mice would reactivate at a higher frequency than those recovered from C57BL/6 mice . Unsorted and CD19+ B cells were plated for ex vivo reactivation analyses at 16 dpi . As observed for splenocytes in the C57BL/6 mice , bulk splenoctyes from IκBαM . MR-infected Bcl2 mice had a similar 50-fold reduction in frequency of cells reactivating virus compared to the frequency of cells harboring viral genome ( Figure 6B and summarized in Table 2 ) . The reactivation of purified CD19+ B cells were reduced approximately 3- to 4-fold compared to bulk splenocytes , a decrease that may reflect the time and manipulation required to purify these cells . The frequency of unsorted and CD19+ splenocytes recovered from IκBαM . 1-infected mice was approximately 50-fold lower than observed with the marker rescue virus ( Figure 6B and Table 2 ) . Taken together , these data confirm the major defect in latency establishment by IκBαM . 1 and argue that inhibiting NF-κB regulation of Bcl-2 expression alone does not account for the phenotype of the IκBαM-expressing viruses . In addition , since no difference in the efficiency of spontaneous virus reactivation was observed with latently infected splenocytes harvested from Bcl-2 transgenic mice compared to C57Bl/6 mice , this indicates that survival of latently infected B cells in culture does not play a significant role in dictating the observed frequency of splenic B cells reactivating virus . Extending the analysis of latency in Bcl-2 transgenic mice , we observed at later times postinfection that there was a slightly greater defect in latency in Bcl-2 trangenic mice infected with IκBαM . 1 ( one in 1 , 983 ) compared to IκBαM . MR ( one in 295 ) ( Figure 6C ) than observed in C57BL/6 mice . This defect was also maintained at 235 dpi with a nearly 10-fold difference ( p = 0 . 0181 ) between the frequency of viral genome–positive splenocytes from mice infected with IκBαM . 1 ( one in 6 , 139 ) and IκBαM . MR ( one in 695 ) ( Figure 6D ) . In addition , these data are consistent with the interpretation that there is a steady rate of decay in the B-cell reservoir over time for both IκBαM . 1 and IκBαM . MR once the initial contraction phase between 16 and 41 dpi occurs . Taken together , the lack of complementation by Bcl-2 at any time point after infection indicates that NF-κB signaling likely provides a role distinct from peripheral B-cell survival during γHV68 latency . Recent reports indicate that lung B cells are infected early after virus inoculation and may represent a long-term reservoir for viral persistence [29 , 48 , 49] . As B cells are critical for acute replication and seeding of the spleen at early times after infection [21 , 22] , we hypothesized that the approximately 50-fold reduction of splenic latency in mice infected with IκBαM . 1 might be due to a reduction in seeding of the spleen from γHV68-infected lung B cells . Indeed , acute virus replication in the spleen at 9 and 12 dpi demonstrated that IκBαM . 1 grew more poorly in the spleens of infected mice compared to IκBαM . MR ( Figure 2E ) , even though IκBαM . 1 exhibited WT replication kinetics in the lungs ( Figure 3D ) . This suggested the possibility of a defect in trafficking of infected B cells from the lung to the spleen and led us to investigate the ability of IκBαM . 1 to establish latency in unsorted lung cells and B cells ( CD19+ ) at 16 dpi . At 16 dpi , there were nearly equivalent frequencies of viral genome–positive unsorted lung cells in mice infected with IκBαM . 1 ( one in 422 ) and WT γHV68 ( one in 298 ) ( Figure 7A ) . CD19+ ( B cells ) and CD19− ( non–B cells ) were enriched from the bulk lung cells and analyzed for the presence of viral genome by limiting-dilution PCR . Non–B cells harbored nearly identical frequencies of viral genomes ( Figure 7B ) . However , as shown in Figure 7C , lung B cells from IκBαM . 1 ( one in 1 , 447 ) -infected mice had significantly decreased levels of viral genomes compared to mice infected with WT virus ( one in 138 ) ( p = 0 . 0316 ) . This represents an approximately 90% reduction in the frequency of viral genome–positive lung B cells in mice infected with the IκBαM . 1 recombinant virus ( Table 1 ) . Because it has been shown that increasing the dose of virus inoculation can overcome the defect in latency establishment of a virus containing a stop mutation in the latency-associated M2 gene [44] , we increased the dose of virus 100-fold to 1 × 105 PFU and examined infection of lung B cells by limiting-dilution PCR analysis of FACS-sorted CD19+ cells ( B cells ) . Notably , lung B cells from mice infected with IκBαM . 1 exhibited a 15-fold reduction in the establishment of latency , even with the higher inoculating dose ( Figure 7D ) . These data indicate that mice infected with a virus that inhibits NF-κB activation exhibit a significant reduction in the establishment of latency in B cells of the lung , a phenotype which could not be overcome by increasing the inoculating dose of virus . It should be noted that the magnitude of the defect observed with the IκBαM-expressing virus in the lungs ( 10- to 15-fold ) was less than that observed in the spleen ( approximately 50-fold ) , suggesting that in addition to playing a role in the establishment of B-cell latency , NF-κB activation also plays a role in subsequent events such as trafficking of latently infected B cells from the lung to the spleen and subsequent establishment of splenic latency . NF-κB is well known for its role in recruiting immune cells by driving the expression of inflammatory cytokines , chemokines , adhesion molecules , and inflammatory mediators . To determine whether inhibition of NF-κB signaling in IκBαM-infected mice impacts the global immune response , we assessed the phenotypes of B and T cells in the lungs and spleens 16 and 42 dpi with 1 , 000 PFU of IκBαM . 1 and WT virus . The number of B cells ( 8 . 2% versus 12 . 4% , p = 0 . 0548 ) , CD4+ T cells ( 9 . 6% versus 12 . 9% , p = 0 . 1065 ) , and CD8+ T cells ( 12 . 9% versus 18 . 1% , p = 0 . 0268 ) present in the lungs of IκBαM . 1-infected mice was slightly diminished at 16 dpi compared to WT virus–infected mice ( Table 3; unpublished data ) . However , there was no significant difference in the activation state of T cells as determined by upregulation of CD11a ( unpublished data ) . Notably , the relatively small difference in the number of B cells present in the lungs is unlikely to contribute significantly to the observed decrease in establishment of B-cell latency in mice infected with the IκBαM . 1 virus . To assess whether there were phenotypic differences between the B-cell populations in IκBαM . 1- and WT virus–infected mice , CD19+ lung and splenic B cells were analyzed for markers of isotype-class switching ( IgD− ) , germinal center participation ( PNAhi ) , proliferation ( Ki67+ ) , and activation ( CD69hi ) at 16 and 42 dpi ( Table 3 ) . The total number of B cells and the frequency of IgD and PNAhi CD19+ B cells were slightly lower in the lungs of IκBαM . 1-infected mice compared to control virus . The proliferation of B cells in the lungs of infected mice was also examined by the measuring bromodeoxyuridine incorporation . Mice were administered bromodeoxyuridine in their drinking water from 8 to 16 dpi; no significant difference in the frequency of B cells incorporating bromodeoxyuridine was observed in IκBαM . 1 virus–infected mice compared to IκBαM . MR virus–infected mice ( unpublished data ) . The only significant difference in B cells from mice infected with IκBαM . 1 compared to WT virus at 16 dpi was a decreased frequency of CD19+CD69hi B cells ( Figure 8A ) . In both the lung ( p = 0 . 0003 ) and spleens ( p < 0 . 0001 ) , there was an approximately 3-fold reduction in CD69hi B cells in IκBαM . 1-infected mice compared to mice infected with IκBαM . MR at day 16 ( Figure 8B ) . By 42 dpi , the frequency of CD69hi B cells in mice infected with either WT virus or the IκBαM . 1 virus was not significantly different from that observed in uninfected mice ( Figure 8 ) . Since CD69 is known to play an important role in lymphocyte trafficking , the difference in CD69 levels between IκBαM . 1 and WT virus may contribute to the more severe latency phenotype observed in splenic B cells compared to lung B cells at 16 dpi [50] . However , it is important to note that ( i ) only approximately 1% of B cells at day 16 are γHV68 latently infected and thus the vast majority of the CD69+ B cells are not virus infected and ( ii ) we currently do not have the necessary tools to assess the status of CD69 expression on latently infected B cells . Although increasing the dose of virus used to inoculate mice via the intranasal route did not rescue the defect in establishment of B-cell latency in the lungs observed with the IκBαM . 1 , we addressed whether altering the route of inoculation might have an impact on the establishment of latency . The kinetics of IκBαM . 1 replication in the spleen was compared to IκBαM . MR and WT virus following intraperitoneal inoculation of mice with 1 , 000 PFU of virus . At days 4 and 9 after intraperitoneal inoculation with 1 , 000 PFU , spleens were harvested and titered . Mice infected with IκBαM . 1 had comparable replication kinetics to WT virus at 4 dpi and slightly lower acute viral titers in the spleen compared to those infected with IκBαM . MR ( Figure 9A ) . Surprisingly , there was a significant reduction in detectable virus replication in the spleens of mice infected with IκBαM . 1 at 9 dpi compared to IκBαM . MR ( p < 0 . 0001 ) and WT virus ( p < 0 . 0001 ) ( Figure 9A ) . This indicates that IκBαM . 1 is impaired in the duration of acute replication compared to IκBαM . MR and WT virus in the spleens of mice infected by the intraperitoneal route . Notably , a shortened period of acute phase replication has previously been observed for some other γHV68 mutants that are not impaired in the establishment of splenic latency [51–53] , indicating that a decreased duration of acute replication in the spleen does not directly correlate with the levels of splenic latency that are established following intraperitoneal inoculation . Since no defect in virus replication in the lungs was observed following the same dose of virus administered via intranasal inoculation ( see Figure 3B ) , this argues that there are distinct requirements for acute virus replication at different anatomical sites . To assess the establishment of viral latency , we initially examined latency in peritoneal exudate cells ( PECs ) . PECs were harvested from infected mice at 16 dpi and the frequency of cells harboring virus was determined by limiting-dilution PCR . The frequency of viral genome–positive PECs in mice infected with IκBαM . 1 ( one in 166 cells ) was nearly equivalent to the frequency of observed in mice infected with IκBαM . MR ( one in 108 cells ) ( Figure 9B ) . The frequency of PECs that reactivated virus from mice infected with IκBαM . 1 ( one in 1 , 065 ) , IκBαM . MR ( one in 453 ) , and WT ( one in 587 ) viruses was nearly identical and was approximately 5-fold lower than the frequency of viral genome–positive cells ( Figure 9D ) . As macrophages comprise the majority of the γHV68-infected cells in the peritoneum of mice [26] , these results indicate that NF-κB activation is not likely required for the establishment of latency , or for reactivation , from latently infected macrophages . Importantly , we did not detect any preformed infectious virus in the disrupted PEC samples ( unpublished data ) . Finally , we assessed the consequences of inhibiting NF-κB activation on the establishment of splenic latency at day 16 post intraperitoneal inoculation ( Figure 9C ) . Splenocytes from mice infected with IκBαM . 1 were found to harbor viral genome at a frequency of one in 327 splenocytes , a 4-fold decrease in establishment compared to a frequency of one in 85 splenocytes upon IκBαM . MR infection ( p = 0 . 0437 ) . The frequency of splenocytes reactivating virus from IκBαM . MR- and WT virus–infected mice was nearly identical ( one in 14 , 925 ) , while the frequency of splenocytes reactivating virus from IκBαM . 1-infected mice was estimated to be one in 113 , 987 , an approximately 8-fold decrease in the levels of reactivating virus compared to IκBαM . MR-infected mice ( Figure 9E ) . The latter results are in close agreement with the defect in the establishment of viral latency . Importantly , the defect in establishment of splenic latency following intraperitoneal inoculation of the IκBαM recombinant virus was significantly less than that observed following intranasal inoculation ( see Figure 4 ) , which may be related to differences in virus trafficking , target B-cell populations , and/or transgene expression ( see Discussion ) .
Overexpression of the NF-κB subunit p65 inhibits the progression of γHV68 lytic replication in 293T cells [33] . The inhibition by p65 may relate to its ability to inhibit transactivation of promoters by γHV68 replication and transcription activator ( RTA ) , as reported for KSHV RTA [33 , 54] . Thus , a virus that inhibits NF-κB and does not antagonize RTA might be predicted to have faster replication kinetics , similar to a γHV68 mutant that expresses RTA constitutively [55] . Indeed , the inability to detect γHV68–IκBαM replication in the spleen 9 d after intraperitoneal inoculation could be attributed to a more rapid peak of acute phase replication ( Figure 9A ) . However , unlike the results reported with a dysregulated RTA-γHV68 mutant [55] , IκBαM had no significant alteration in the kinetics of growth in fibroblast cells in culture and did not exhibit heightened peak titers in the lungs after intranasal inoculation ( Figure 3 ) . These data highlight a tissue-dependent requirement for NF-κB during productive replication but do not support an increase in lytic kinetics as an explanation for the substantial decrease in splenic latency upon intranasal infection with γHV68-IκBαM . CD40L , BAFF , and lymphotoxin are costimulatory signals provided by dendritic cells and T cells in germinal centers that are required to drive NF-κB activation for B-cell survival upon activation and differentiation [56] . Given that γHV68 establishes latency in B cells bearing surface markers characteristic of cells proliferating and participating in germinal center reactions [2 , 30 , 31] , NF-κB is likely required upon activation for proliferation of the infected cells . Examination of splenocytes determined that infection with γHV68-IκBαM resulted in a substantial ( 97% ) inhibition in the establishment of latency 16 dpi in comparison to WT virus ( Figure 4 and Table 1 ) . We also determined the frequency of latently infected cells in B-cell and non–B-cell subpopulations; γHV68-IκBαM had a larger deficit in splenic establishment in B cells ( 98% reduction ) compared to non–B cells ( 83% reduction ) . The lack of absolute ablation of latency upon infection with IκBαM might reflect ( i ) the utilization of alternative NIK/IKK1-mediated NF-κB activation , ( ii ) the establishment of virus infection in cell populations that do not rely on NF-κB , and/or ( iii ) inadequate expression of the IκBαM transgene; the HCMV immediate-early promoter might not drive sufficient levels of IκBαM expression in some cells or cell populations ( as such , the residual latency observed at 16 dpi might simply reflect a population of cells in which NF-κB activation was not suppressed ) . Regardless , utilization of the transdominant IκBαM clearly had a substantial impact on the ability of γHV68 to establish latency in vivo . In a mixed CD40+CD40− bone marrow chimeric system , both CD40-deficient and CD40-sufficient B cells were found to harbor viral genomes at 14 dpi at similar frequencies [3] . However , over time , latently infected CD40-deficient B cells were lost . This is consistent with an interpretation for a requirement for CD40-mediated signaling and participation in germinal centers of infected mice for long-term maintenance of γHV68 B-cell latency [3] . In contrast to the failure to maintain latency in CD40-deficient B cells at later timepoints [3] , γHV68-IκBaM–infected B cells exhibit a defect in the establishment of latency at 16 dpi . This indicates that activation of NF-κB plays a critical role in addition to , and prior to , the engagement of infected cells with CD40L in germinal center reactions . Thus , the impairment in the establishment of splenic latency at 16 dpi in mice infected with the recombinant γHV68-expressing IκBαM strongly suggests that NF-κB activation mediated by the classic pathway is essential for the efficient establishment of latency at early time points . Upon encountering antigen , NF-κB functions in mature B cells as a survival factor by mediating the upregulation of several antiapoptotic proteins , including A1 , Bcl-2 , and Bcl-xL [7 , 57] . One explanation for the large defect in latency establishment is that the inhibition of NF-κB results in the loss of expression of antiapoptotic proteins , such as Bcl-2 or Bcl-xL , leading to cell death . Bcl-2 knockout mice lack mature B cells in the periphery , demonstrating a role for Bcl-2 in peripheral B-cell survival [58] . Transgenic mice with targeted expression of Bcl-2 in B cells are characterized by larger numbers of B cells and reduced apoptosis in germinal centers , leading to an accumulation of memory B cells with a low frequency of somatic hypermutation [59–61] . Bcl-2 and Bcl-xL reduce apoptosis in resting and activated crel−/− and nfkb1−/− mature B cells but do not restore proliferative responses in response to mitogenic stimuli [62 , 63] . Although Bcl-2 complemented BAFF−/− mice for peripheral B-cell survival , it did not restore differentiation and marginal zone B cells [64] . Similarly , BAFF-R–deficient mice expressing the Bcl-2 transgene remain defective in the formation of lymphoid follicles and germinal centers [65] . Taken together , Bcl-2 transgenic mice might be expected to rescue a latency defect for γHV68-IκBαM if NF-κB is required for survival in the periphery but would be unlikely to rescue a defect in activation-induced proliferation and participation in germinal center reactions . As shown here ( see Figure 6 ) , in the presence of both the endogenous virally encoded bcl2 homolog ( M11 ) and constitutive expression of the cellular Bcl-2 in B cells , the deficit in latency establishment was still observed . γHV68-IκBαM exhibited a 99% reduction in latency establishment in these mice ( both in bulk splenocytes and in purified B cells ) compared to IκBαM . MR at 16 dpi ( Table 1 ) . γHV68-IκBαM viral load remained 10-fold lower than IκBαM . MR at 6 wk and even through 7 mo after infection . The lack of latency restoration for γHV68-IκBαM in the Bcl-2 transgenic mice indicates that NF-κB plays a critical role ( s ) in addition to cell survival . The complementation of cell survival , but not proliferation , in response to mitogenic stimulation by the bcl-2 transgene in c-rel−/− B cells indicates that NF-κB likely functions to promote cell cycle progression in a Bcl-2–independent manner [63] . c-rel−/− B cells fail to enter S phase upon B-cell receptor activation , a block in cell cycle progression that is attributed to the failure to induce the transcription of G1 cyclins D3 and E , which in turn leads to a loss of cdk activity and delayed pRB phosphorylation [66] . NF-κB activation likely mediates important roles in cell cycle progression that are required during the proliferative expansion of γHV68-infected B cells in the spleen [31 , 32] . An examination of the maintenance of long-term latency at 6 wk postinfection demonstrated , as previously observed , a significant contraction in splenic latency from 16 to 42 dpi for WT and marker rescue , in both C57BL/6 and Bcl-2 transgenic mice . However , a similar magnitude of contraction of splenic latency was not observed in γHV68-IκBαM–infected mice . The end result is that the latency defect of the γHV68-IκBαM virus is diminished at later times postinfection . Similar observations have been made for other γHV68 mutants , including viruses with targeted mutations in the M2 , M4 , gp150 , and viral TK genes [44 , 45 , 51 , 67 , 68] , all of which exhibit severe latency establishment phenotypes at day 16 postinfection but less substantial latency phenotypes at later times postinfection . The slower decline in splenic latency for γHV68-IκBαM might be attributed to ( i ) a delay in virus ( or latently infected cell ) trafficking from the lung to the spleen , ( ii ) a contribution of virus ( or latently infected cells ) from nonsplenic reservoirs of latency that supplements infected cell turnover in the spleen , ( iii ) direct infection of a more stable , nonproliferating reservoir of cells ( e . g . , memory B cells ) , ( iv ) an accumulation of virus in one particular subset of cells that compensates for loss in another , and/or ( v ) altered trafficking of latently infected cells from the spleen to other sites ( egress ) . Cidofovir treatment of B-cell–deficient mice between 16 and 42 dpi has been demonstrated to reduce viral latency in PECs and splenocytes after intraperitoneal infection [46 , 47] . In contrast , utilizing the same treatment regimen in C57BL/6 mice infected with the IκBαM or IκBαM . MR virus , there was no change in the levels of viral latency in the spleens of mice following antiviral therapy from 16 to 39 dpi . These data are consistent with the interpretation that virus reactivation and seeding of the spleen from alternative reservoirs does not contribute to splenic latency at 6 wk postinfection . It is possible that the IκBαM-expressing γHV68 may directly infect a nonproliferative stable cell reservoir , such as memory B cells , that is detected at all time points . We have previously observed route-dependent phenotypes for a γHV68 mutant in which the M2 gene has been disrupted [44 , 51] . Similarly , when mice were infected via intraperitoneal inoculation , a distinctly different phenotype of the γHV68-IκBαM was observed . By this route , acute phase replication at 9 dpi was nearly undetectable compared to WT or MR , yet these altered kinetics did not substantially affect the establishment of latency in the PEC or splenocytes by the IκBαM virus at 16 dpi . Instead , the defect in the establishment of splenic latency was more modest than observed following intranasal inoculation ( compare Figures 4 and 9 ) . A disconnect between lytic replication and latency establishment following intraperitoneal inoculation has been reported for other γHV68 mutants ( e . g . , a γHV68 mutant lacking 9 . 5 kb at the left end of the genome [53] , a γHV68 mutant containing a stop mutation in the M2 ORF [51] , or a γHV68 mutant containing an insertion of a LacZ expression cassette in the M1 ORF [52] ) . Interestingly , mice lacking B cells do not maintain acute replication of WT γHV68 in the spleen at day 9 following intraperitoneal infection , yet the spleens do harbor significant levels of latently infected cells at later timepoints [22 , 43] . This indicates a role for B cells in prolonging the acute phase of lytic replication , perhaps via reactivation and seeding permissive cells in the spleen . However , it is notable that B cells appear dispensable for the establishment of latency in other cell types in the spleen ( e . g . , dendritic cells and macrophages ) [22] . Further characterization of the kinetics of mutant virus replication and the induction of the host immune response are required to understand the basis for the more rapid clearance of acute virus replication in the spleen for the IκBαM-expressing virus . Regardless of the mechanism for the shortened acute phase of replication , we can infer that the virus has accessed the latent reservoir of the spleen before 9 dpi after intraperitoneal inoculation . Following intraperitoneal inoculation , γHV68 did not require NF-κB activation for establishment of latency , or reactivation from latency , in PECs . Since macrophages are the major component of PECs [26] , this suggests that NF-κB activation is dispensable for establishment of γHV68 latency in macrophages . This is consistent with the observation that the frequency of viral genome–positive non–B cells in the lungs at day 16 was also largely unaffected by inhibition of NF-κB activation ( see Figure 7 ) . In contrast to PECs , there was a slight , but significant , decrease in latency establishment in the spleens of mice following intraperitoneal infection with IκBαM-γHV68 . This phenotype also was apparent in a frequency analysis of latency in sorted B-cell populations ( unpublished data ) . The mild phenotype upon intraperitoneal inoculation with IκBαM-γHV68 compared to the approximately 50-fold decrease in splenic latency upon intranasal infection indicates that there are critical mechanistic differences with respect to seeding splenic latency between these routes of inoculation . The analysis of splenic latency in IκBαM-γHV68–infected mice following intraperitoneal inoculation suggests the establishment of a form of B-cell latency that does not require an active role for the virus in driving activation and proliferation of these infected splenic B cells , similar to phenotypes uncovered for M2- and v-bcl2–null viruses [44 , 46 , 51 , 69] . Further investigation of splenic latency following intraperitoneal inoculation is required to fully understand the observed differences in establishment of splenic B-cell latency . Following intranasal inoculation , there is a significant impact of inhibiting NF-κB activation on establishment of latency in the spleen , but this phenotype is reduced substantially upon the more direct and permissive intraperitoneal route of inoculation ( compare Figures 4 and 7 ) . This raises the possibility that altered trafficking of latently infected cells to the spleen also contributes to the splenic latency phenotype observed at day 16 with the IκBαM-expressing virus following intranasal inoculation . Interestingly , although splenic latency is established in B-cell–deficient mice after intraperitoneal inoculation , there is a defect in the establishment of latency in the spleen following intranasal inoculation of these mice [21 , 24] . Adoptive transfer of B cells into B-cell–deficient mice infected with γHV68 restored latency in the spleen [23] . In addition , it has previously been reported that intranasal infection of mice with replication defective mutants ( ORF31- or ORF50-null viruses ) , resulted in virus infection of B cells in the lung , yet these viruses were not detected in the spleen [48 , 49] . Taken together , the seeding of the spleen may result from reactivation of latently infected B cells that traffick from the lungs to the spleen , thereby seeding acute virus replication in the spleen and the subsequent establishment of splenic latency . This led us to investigate NF-κB–mediated functions in the lung , the site of primary infection . As shown here , the IκBαM-expressing γHV68 was substantially impaired in the establishment of latency in lung B cells ( see Figure 7 ) , where a 90% reduction in latency in this particular cell subset was observed . It is also notable that increasing the inoculating dose of virus 100-fold did not ameliorate this defect in B-cell latency in the lungs . The defect in establishment of B-cell latency in the lungs ( Figure 7 ) , coupled with both a slight diminishment of acute replication in the spleen ( Figure 3E ) and a substantial defect in the establishment of latency in the spleen ( Figure 4 ) , suggests that the efficient seeding of γHV68 latency in the spleen following intranasal inoculation of virus requires the egress of latently infected B cells from the lung . The absence of heightened lytic replication or viral load in latency reservoirs in mice infected with IκBαM-expressing γHV68 demonstrates that the host is able to mount an effective immune response ( i . e . , this method of targeted NF-κB inhibition only in virus-infected cells does not appear to result in any gross immune dysfunction ) . However , the requirement of NF-κB activation for the efficient establishment of lung B-cell latency could reflect , in part , a role for this host cell factor in mediating inflammatory cytokine production by infected cells that could lead to alterations in the recruitment , activation , and/or subsequent trafficking of infected cells to the spleen . In an examination of the profile of lymphocytes in both the lungs and spleens at early and late times after infection , we failed to observed any significant differences in the numbers of B or T cells recruited to the lungs or present in the spleens of mice infected with the IκBαM-expressing γHV68 . The number of PNAhi and IgD− cells were slightly lower for IκBαM-infected mice . The frequency of proliferating cells at 16 dpi , as detected by the Ki67 nuclear proliferation antigen ( Table 3 ) , and confirmed by bromodeoxyuridine incorporation from 8 through 16 dpi ( unpublished data ) , was not dramatically reduced . This contrasts with the nearly 3-fold decrease in the activation of B cells in the lungs and spleens at 16 dpi as determined by the presence of the CD69 activation marker , which correlated with the reduction in viral load in the B cells of the lung and spleen . It is known that γHV68 infection results in polyclonal activation of B cells , characterized by an upregulation of the CD69 activation marker and an increase in nonspecific antibody production , a process that is dependent on CD4+ T cells [43 , 70 , 71] . A reduction in CD69 activation has been correlated to both a decrease in splenomegaly and splenic latency in mice infected with γHV68 mutants deficient in vbcl-2 [69] , containing a large insertion in the M3 gene [72] , or a modification linking the ORF73 product to a CD8–T-cell epitope [73] . The lack of an acute replication defect for the IκBαM-expressing γHV68 , as with these other mutants after intranasal infection , substantiates the conclusion that CD69 activation correlates with differences in the levels of splenic latency and extends that correlation to latency in B cells at the primary site of infection . It is very unlikely that each of these mutated ORFs has a direct role in CD69 upregulation , as acute titers were normal for each mutant . Considering the disconnect between the large numbers of B cells activated and the small fraction that is virus infected , there may be a factor secreted by latently infected B cells that activates neighboring B cells in a paracrine manner . Nonspecific B-cell activation might represent a mechanism by which the virus drives B-cell participation in germinal center reactions to increase the likelihood of gaining access to the long-lived memory B-cell reservoir . NF-κB activation via the classic signaling pathway drives the expression of a multitude of proinflammatory molecules , including the cytokines TNF , interleukin ( IL ) -1 , and IL-6; chemokines RANTES ( regulated on activation , normal T-cell expressed and secreted ) , macrophage inflammatory protein-1a , and membrane cofactor protein-1 , and the inflammatory mediators inducible nitric oxide synthase and cyclooxygenase 2 ( COX-2 ) ( reviewed in [74] ) . These inflammatory molecules are detected in lung lavage during acute γHV68 replication [75–77] and might be altered upon infection with the IκBαM-expressing virus . Inducible nitric oxide synthase−/− mice have been reported to exhibit no alteration in immune clearance or splenomegaly upon γHV68 infection [78] . NF-κB also drives the expression of cellular COX-2 , an enzyme that catalyzes the production of the inflammatory prostaglandin E2 . COX-2 is induced upon γHV68 infection in vitro , and the inhibition of COX-2 with the nonsteroidal anti-inflammatory drug NS-398 reduced viral protein production and virion production in vitro [79] . However , this group reported that the targeting of COX-2 did not translate into an inhibition of acute replication in the lungs of mice upon treatment with NS-398 [79] . IL-6 is detected in the lungs and lymph nodes of infected mice and is produced upon in vitro stimulation of lymphocytes from infected mice [80 , 81] . However , there was no difference in virus growth , in establishment of latency , or in the immune response to the virus in IL-6−/− mice [81 , 82] . The inflammatory response to γHV68 is multifaceted , involving the contribution of proinflammatory molecules that are produced by the heterogeneous population of infected cells and the multiple immune cell types that infiltrate and are activated at the site of infection [75] . Thus , the lack of a substantial phenotype upon the knockout of single NF-κB–regulated inflammatory components is not altogether unexpected . In future studies , the characterization of the cytokine and chemokine responses to infection by IκBαM-expressing γHV68 might reveal alterations of biological relevance . A number of gammaherpesvirus gene products have been shown to modulate NF-κB activity . EBV LMP-1 and the NF-κB modulatory proteins of KSHV , vIL-6 , K15 , K1 , and vFLIP are not encoded by γHV68 and , as such , are not candidates for mediating NF-κB activation during establishment of γHV68 latency . The γHV68 viral G protein–coupled receptor ( vGPCR; encoded by ORF 74 and a homolog of the KSHV vGPCR ) is a weak inducer of NF-κB activation in a ligand-dependent manner [83 , 84] . The expression of the chemokines RANTES and macrophage inflammatory protein , which mediate vGPCR activation in cell culture , are detected in the lungs of γHV68-infected mice [85] , raising the possibility that vGPCR mediated induction of NF-κB during the acute phase of virus replication in the lungs . However , the significance of such modulation on the establishment of latency is unclear , since a stop mutation in ORF74 manifests as a defect in reactivation , not latency establishment in vivo [83 , 86] . Given the importance , as shown here , of NF-κB activation for establishment of latency in vivo , it is possible that initial activation of NF-κB ( e . g . , through virus binding and entry ) may serve to promote establishment of a latent infection in the appropriate cellular context prior to expression of latency-associated viral genes that activate NF-κB . For example , in the case of EBV infection of primary B cells , NF-κB is activated upon the binding of the viral glycoprotein gp350 to the cell surface [87] . It is not known whether binding of the γHV68 homolog of EBV gp350 , gp150 , is involved in the initial activation of NF-κB during γHV68 infection of permissive cells . Regardless , the phenotype of the gp150-null virus is distinct from the IκBαM phenotype [88] , making it unlikely that gp150 plays a critical role in activating NF-κB during the establishment of viral latency . It seems very likely that γHV68 encodes other , as-yet-unidentified , gene products that modulate the activation of NF-κB to manipulate the cell and the host microenvironment and thereby optimize conditions for viral persistence . We have shown here that NF-κB signaling is critical for efficient establishment of viral latency following intranasal inoculation of γHV68 pathogenesis and , as such , likely represents a key host factor that is manipulated by the virus to establish latency in B cells . To our knowledge , this is the first report to identify a host transcription factor required for the efficient establishment of gammaherpesvirus latency in vivo . As NF-κB signaling is intimately involved in inflammation and B-cell biology , future studies aim to determine the role of specific NF-κB subunits and signaling pathways involved in the establishment of latency , cytokine production , trafficking , proliferation , and participation of B cells in germinal center reactions . Further characterization of the molecular mechanisms by which γHV68 modulates NF-κB activation is critical to understanding virus–host interactions that transpire , ultimately allowing the virus to gain access to the long-lived memory B-cell reservoir . We anticipate that γHV68 latency in B cells closely parallels EBV infection in humans , which involves virus-driven B-cell proliferation and differentiation . In the case of EBV infection , it is thought that the ability of the virus to manipulate B-cell differentiation places the host at greater risk for lymphoproliferative disorders and lymphoma . Furthermore , modulation of NF-κB activation by EBV is clearly critical to these processes . Exploiting γHV68 infection of mice as a model system may lead to critical insights into the role of gammaherpesviruses in lymphomagenesis .
γHV68 WUMS ( American Type Culture Collection VR1465 , http://www . atcc . org ) was the WT virus . Virus passage and titer were performed as previously described [22] . NIH 3T12 cells and MEFs were maintained in DMEM supplemented with 100 U of penicillin/ml , 100 mg of streptomycin/ml , 10% FCS , and 2 mM l-glutamate ( cMEM ) . Cells were maintained at 37 °C in a 5% CO2 environment . MEF cells were prepared from C57BL/6 as previously described [89] . Vero-Cre cells were a gift from David Leib and were passaged in cMEM supplemented with 300 μg of hygromycin B/ml ( Calbiochem , http://www . calbiochem . com ) . The γHV68 genome cloned as a BAC was a kind gift of Ulrich Kozinowski [41] . For the generation of the recombinant γHV68-IκBαM . 1 and . 2 viruses , the mutant superrepressor form of IκBα that contains serine-to-alanine substitutions at amino acids 32 and 36 ( IκBαM ) driven by the CMV immediate-early promoter was inserted into the ORF27–ORF29b intergenic region of γHV68 . The 1 . 8-kb AseI-MluI fragment of pIκBαM ( Stratagene , http://www . stratagene . com ) was cloned into the PmlI site ( corresponding to γHV68 genomic position 46347 ) of plasmid JE110 [32] that contains the γHV68 genomic region between nucleotide positions 45237 and 48347 . Insertion of the IκBαM expression construct into the PmlI site of pJE110 in the rightward and leftward orientation generated JE110-IκBαM . 1 and JE110-IκBαM . 2 , respectively . To generate the targeting construct for use in recombination , the BglII-EcoRI fragment of pJE110-IκBαM was cloned into the suicide donor plasmid pGS284 to generate pGS284-IκBαM . Allelic exchange with WT γHV68-BAC in GS500 E . coli ( RecA+ ) cells was performed as previously described [90] , and the insertion of IκBαM to generate recombinant γHV68-BAC-IκBαM . 1 and . 2 was screened by colony PCR using primers ACGGGCTGAAGAAGGAGCGGCTACT and TGCCCAGGTAGCCATGGATAGAGG with Taq DNA polymerase ( Promega , http://www . promega . com ) . For generation of the marker rescue virus , the BglII-EcoRI fragment of pJE110 was cloned into pGS284 to generate pGS284-27-29B . Allelic exchange of this targeting plasmid with γHV68-IκBαM . 1-BAC was performed to generate γHV68-IκBαM . MR . Loss of IκBαM was verified by diagnostic PCR as described above . Viral stocks were generated by Superfect ( Qiagen , http://www . qiagen . com ) transfection of BAC clones of γHV68-IκBαM . 1 and . 2 or γHV68-IκBαM . MR into Vero-Cre cells . Following the appearance of cytopathic effect , wells were harvested and used to infect Vero-Cre cells to generate high-titer viral stocks . Virus was isolated from the supernatants of infected cells when monolayers exhibited greater than 50% cytopathic effect . After two rounds of 20-min room-temperature spins at 2 , 000 × g , clarified supernatants were spun at 15 , 000 × g for 2 h at 4 °C . The pellet was rinsed with PBS , resuspended in 50 mM Tris ( pH 7 . 5 ) , 10 mM MgCl2 , and 25 U of DNase ( Invitrogen , http://www . invitrogen . com ) , and incubated for 35 min at 37 °C before being layered over a 20% sucrose cushion and centrifuged at 110 , 000 × g for 1 h at room temperature . The virion pellet was resuspended in 3 ml of 10 mM Tris ( pH 7 . 5 ) , 1 mM EDTA and mixed with 3 ml of a 2× lysis buffer containing 2% Sarkosyl , 0 . 5% SDS , 40 mM Tris ( pH 7 . 5 ) , 200 mM NaCl , 20 mM EDTA , and 333 μg/ml proteinase K ( GIBCO-BRL , http://www . gibco . com ) and then incubated at 37 °C overnight . Samples were subjected to a phenol/chloroform and chloroform extraction , and DNA was precipitated with 3 M Na aAcetate and 2 volumes of 100% ethanol . Southern analysis was performed with a HindIII restriction digest of the viral DNA and a 32P-labeled probe that detects ORF27 and the intergenic region between ORFs 27 and 29b ( bp 45237 to 46420 of γHV68 WUMS ) and with a BamHI restriction digest of the viral DNA and a 32P-labeled IκBαM probe ( BamHI-XhoI fragment of pIκBαM ) . Female C57BL/6J mice ( catalog No . 000664; The Jackson Laboratory , http://www . jax . org ) were housed at the Yerkes vivarium in accordance with federal and university guidelines . C57BL/6-Tg ( BCL2 ) 22Wehi/J ( catalog No . 002319; The Jackson Laboratory ) were maintained as heterozygous animals by breeding with C57BL6/6J mice under sterile breeding conditions at the Yerkes vivarium . Tg ( BCL2 ) mice contain a transgene construct consisting of the human BCL2 cDNA driven by the Emu immunoglobulin heavy chain enhancer and SV40 promoter; BCL2 expression is thereby restricted to the B-cell lineage [60] . All protocols for animal studies were approved by the Institutional Animal Care and Use Committee of Emory University . Mice between 8 and 12 wk of age were placed under isofluorane anesthesia prior to intranasal inoculation with 1 , 000 PFU or 1 × 105 PFU of virus in 20 μl of cMEM or intraperitoneal inoculation of 1 , 000 PFU of virus in 0 . 5 ml of cMEM . Spleens were harvested into cMEM , homogenized , and filtered through a 100-μm-pore nylon cell strainer ( Becton Dickinson , http://www . bd . com ) . Erythrocytes were removed with red blood cell lysis buffer ( Sigma , http://www . sigmaaldrich . com ) . Except where indicated , pooled splenocytes from three to 15 mice were used in all experiments . Lungs were harvested intact and briefly incubated in HEPES-buffered saline solution with 1 . 3 mM EDTA supplemented with 2% fetal calf serum ( HBSS++ ) prior to inflation with approximately 1 ml of 300 U/ml collagenase type I ( MOP4405; Worthington Biochemical , http://www . worthington-biochem . com ) prepared in the HBSS++ solution , minced , and then incubated for 1 h at 37 °C . Collagenase-disrupted lungs were homogenized , filtered through a 100-μm cell strainer , treated with blood cell lysis buffer , and resuspended in PBS supplemented with 0 . 5% or 2% FCS , depending on downstream application . Individual mice were weighed prior to the first injection and weekly thereafter . Cidofovir ( Vistide; Gilead Sciences , http://www . gilead . com ) was administered subcutaneously in the scruff of the neck at a dose of 25 mg/kg of body weight [47 , 48] every 3 d between 16 and 39 dpi with 1 , 000 PFU of the viruses as indicated in the legend to Figure 5 . Plaque assays were performed as previously described [53] , with minor modifications . NIH 3T12 cells were plated onto six-well pates 1 d prior to infection at 2 × 105 cells per well . Organs were subjected to four rounds of mechanical disruption of 1 min each using 1 . 0-mm zirconia/silica beads ( Biospec Products , http://www . biospec . com ) in a Mini-Beadbeater-8 ( Biospec Products ) . Serial 10-fold dilutions of organ homogenate were plated onto NIH 3T12 monolayers in a 200-μl volume . Infections were performed for 1 h at 37 °C with rocking every 15 min . Immediately after infection , plates were overlaid with 2% methylcellulose in cMEM . After 6 to 7 d , plates were stained with a neutral red overlay , and plaques were scored the next day . The limit of detection for this assay is 50 PFU per organ . NIH 3T12 cells ( 3 × 105 ) were seeded onto a six-well dish 1 d prior to treatment . Cells were infected at an MOI of 1 or 10 in 200 μl for 1 h at 37 °C with rocking every 15 min and then 2 ml of fresh medium was added . At 6 h after the addition of innoculum , a total of 2 μg of DNA per well was transfected by the lipofection-based method ( LipofectAMINE Plus; Invitrogen ) . pNF-κBLuc is an NF-κB–responsive reporter construct that expresses Photinus pyralis luciferase ( Stratagene ) . pMEKK expresses an activator of NF-κB ( Stratagene ) . pIκBαM expresses IκBα S32/36 , a dominant inhibitor of NF-κB ( Stratagene ) , and pBSIIKS+ ( Stratagene ) is an empty vector used to bring up DNA content to equivalent levels for each transfection . pEGFP+ ( Stratagene ) was used to monitor transfection efficiency by immunofluorescence microscopy ( Nikon , http://www . nikon . com ) . Cell extracts were harvested 24 h posttransfection and assayed using the luciferase reporter assay system ( Promega ) . Transfections were performed independently in triplicate at least two times . Cells were harvested , washed once in PBS , and resuspended into ice-cold hypotonic lysis buffer ( 10 mM HEPES [pH 7 . 9] , 10 mM KCl , 1 . 5 mM MgCl2 , 0 . 1 mM EDTA , 1 mM dithitothreitol , 0 . 5 mM phenylmethylsulfonyl fluoride , protease inhibitor cocktail; Roche , http://www . roche . com ) for 15 min prior to the addition of 1/20 volume of 10% Nonidet P-40 . Nuclei were spun down , washed in hypotonic lysis buffer , and then resuspended in high salt buffer ( 25% glycerol , 20 mM HEPES [pH 7 . 9] , 420 mM NaCl , 1 . 5 mM MgCl2 , 0 . 2 mM EDTA , 0 . 5 mM dithitothreitol , 0 . 5 mM phenylmethylsulfonyl fluoride , protease inhibitor cocktail; Roche ) with vigorous shaking at 4 °C . The supernatant collected after 10 , 000 × g centrifugation for 10 min at 4 °C was nuclear extract . Nuclear extracts were assayed for NF-κB activation by electrophoretic mobility shift assay . Five micrograms of nuclear extract were incubated with a 32P-labeled oligonucleotide containing the NF-κB consensus site , AGTTGAGGGGACTTTCCCAGGC , in a binding reaction containing 2 mM HEPES ( pH 7 . 9 ) , 1 mM EDTA , 5 mM dithiothreitol , 0 . 05% Triton X-100 , 5% glycerol , and 2 μg of poly d ( I-C ) ( Roche ) for 30 min at room temperature . Competition experiments were performed with 20- and 100-fold molar excess of unlabeled oligonucleotides containing the WT or mutated NF-κB consensus site ( AGTTGAGGCGACTTTCCCAGGC ) . Nucleoprotein complexes were subjected to electrophoresis in 5% native polyacrylamide gels at 180 V , dried under vacuum , and analyzed by PhosphorImager analysis ( Typhoon 9410; Amersham Biosciences , http://www . amersham . com ) . Cells resuspended in PBS supplemented with 2% FCS were stained for FACS using a combination of the following antibodies: phycoerythrin ( PE ) - or allophycocyanin-conjugated antibodies to CD19; fluorescein isothiocyanate ( FITC ) -conjugated antibodies to B220 , IgD , CD69 , PNA , Ki67 , and the IgG1 isotype control . When necessary , rat anti-mouse CD16/CD32 ( Fc block ) was used to block Fc receptors prior to staining . All reagents were obtained from BD Biosciences ( http://www . bdbiosciences . com ) except the FITC-conjugated Ki67 ( Novocastra Laboratories , http://www . ebiotrade . com/buyf/Novocastra/index . htm ) . For FACS , cells were resuspended at 2 × 107 cells/ml and incubated for 30 min on ice in PBS containing 1% FCS and Fc block . Cells were stained with PE-conjugated anti-CD19 at 5 μl per 1 × 107 cells by incubation for 30 min on ice in the dark . Cells were washed twice with PBS containing 1% FCS and resuspended at 1 × 108 cells/ml . Stained cell populations were acquired using a MoFlo fluorescence activated cell sorter ( DAKO , http://www . dako . com ) , FACSAria ( BD Biosciences ) , or FACSVantage ( BD Biosciences ) . Sorted and unsorted cell populations were resuspended in cMEM supplemented with 10% dimethyl sulfoxide and stored at −80 °C for limiting-dilution PCR analyses or in cMEM at 4 °C for limiting-dilution ex vivo reactivation analyses as described below . For flow cytometry analysis , cells were resuspended at 1 × 106 cells/ml in PBS containing 2% FCS for 20 min on ice in the dark and stained with 1:100 dilution of allophycocyanin-conjugated anti-CD19 and 1:100 to 1:200 dilution of FITC-conjugated antibodies . For intracellular staining for the detection of the Ki67 nuclear antigen , cells were surface stained prior to fixation in 2% paraformaldehyde and permeabilization with 0 . 1% Triton X-100 in PBS , followed by incubation with 1:100 dilution of the FITC-conjugated anti-Ki67 antibody or the isotype control . Data were collected on an FACSCaliber ( BD Biosciences ) and analyzed using FlowJo software ( TreeStar , http://www . treestar . com ) . Murine B cells were isolated by depletion of non–B cells using the B Cell Isolation Kit ( Miltenyi Biotec , http://www . miltenyibiotec . com ) per the manufacturer's recommendations . Briefly , cells were resuspended at 2 × 108 cells/ml in 1× PBS containing 0 . 5% FCS followed by staining with Fc block ( 0 . 125 μg/106 cells ) on ice for 15 min . Cells were labeled with biotin-antibody cocktail ( biotin-conjugated antibodies against CD43 , CD4 , and Ter-119 ) , 10 μl per 1 × 107 cells for 15 min on ice followed by staining with anti-biotin microbeads , 20 μl per 1 × 107 cells for 15 min on ice . Cells were washed twice with PBS containing 0 . 5% FCS and subjected to magnetic separation using the autoMACS ( Miltenyi Biotec ) . Following separation , stained cell populations were analyzed by flow cytometry as described above . Lung B cells were isolated by positive selection of B cells using a PE-conjugated antibody against CD19 and the PE isolation kit per the manufacturer's recommendations ( Miltenyi Biotec ) . Briefly , cells were resuspended at 2 × 108 cells/ml in 1× PBS containing 0 . 5% FCS followed by staining with Fc block ( 0 . 125 μg/106 cells ) on ice for 15 min . Cells were labeled with CD19 conjugated to PE , 5 μl per 1 × 107 cells for 30 min on ice in the dark , spun out , and resuspended in 80 μl of PBS containing 0 . 5% FCS per 1 × 107 cells followed by staining with anti-PE microbeads , 20 μl per 1 × 107 cells , for 15 min on ice in the dark . Cells were washed twice with PBS containing 0 . 5% FCS , filtered through 40-μm cell strainers , and subjected to magnetic separation using the autoMACS ( Miltenyi Biotec ) . Limiting-dilution analysis to determine the frequency of cells containing virus capable of reactivating from latency was performed as previously described [22 , 26] . Briefly , bulk splenocytes or sorted cell populations were resuspended in cMEM and plated in serial 2-fold dilutions ( starting with 105 cells ) onto MEF monolayers in 96-well tissue culture plates . Twelve dilutions were plated per sample , and 24 wells were plated per dilution . Wells were scored for cytopathic effect at 21 to 28 d postplating . To detect preformed infectious virus , parallel samples of mechanically disrupted cells were plated onto MEF monolayers . This process kills more than 99% of live cells , which allows preformed infectious virus to be discerned from virus reactivating from latently infected cells [22 , 26] . Unless indicated otherwise , significant levels of preformed virus were not detected in these assays . Limiting-dilution analysis to determine the frequency of cells harboring the viral genome was performed using a single-copy-sensitive nested PCR assay as previously described [24 , 26] . Briefly , frozen samples were thawed , counted , resuspended in isotonic buffer , and plated in serial 3-fold dilutions in a background of 104 uninfected NIH 3T12 cells in 96-well plates ( Eppendorf Scientific , http://www . eppendorf . com ) . Plates were covered with PCR foil ( Eppendorf Scientific ) , and cells were lysed with proteinase K for 6 h at 56 °C . Then , 10 μl of round 1 PCR mix was added to each well by foil puncture . Following first-round PCR , 10 μl of round 2 PCR mix was added to each well by foil puncture and samples were subjected to round 2 PCR . All cell lysis and PCR were performed on a PrimusHT thermal cycler ( MWG Biotech , http://www . mwg-biotech . com ) . Products were resolved by ethidium bromide staining on 2% agarose gels . Twelve PCRs were performed for each sample dilution , and a total of six dilutions were performed per sample . Every PCR plate contained control reactions ( uninfected cells and ten copies , one copy , 0 . 1 copy of plasmid DNA in a background of 104 cells ) . All of the assays demonstrated approximately single-copy sensitivity with no false positives . γHV68-IκBαM . R1 and . R2 viruses were recovered from splenocytes harvested 16 dpi . Cytopathic effect was observed in wells at 14 d postplating . Wells that exhibited viral cytopathic effect and contained the fewest number of input splenocytes were harvested and used to infect NIH 3T12 fibroblasts to generate high-titer viral stocks and viral DNA . All data were analyzed by using GraphPad Prism software ( GraphPad Software , http://www . graphpad . com ) . Titer data were statistically analyzed with the Mann-Whitney nonparametric two-tailed t-test . Based on the Poisson distribution , the frequencies of reactivation and viral genome–positive cells were obtained from the nonlinear regression fit of the data where the regression line intersected 63 . 2% . The frequencies of reactivation and genome-positive cells were statistically analyzed by unpaired two-tailed t-test of the log 63 . 2% effective concentration .
The GenBank ( http://www . ncbi . nlm . nih . gov/Genbank ) sequence for γHV68 WUMS is U97553 . | A central aspect of chronic infection of a host by herpesviruses is the ability of these viruses to establish a quiescent infection ( latent infection ) in some cell type ( s ) in which there is only intermittent production of progeny virus ( virus reactivation ) . The establishment of a latent infection in the antibody producing cells of the host immune system ( B lymphocytes ) is critical for life-long persistence of gammaherpesviruses , as well as the development of virus-associated lymphoproliferative diseases ( e . g . , B-cell lymphomas ) . Nuclear factor ( NF ) -κB transcription factors are a family of cellular proteins that play an important role regulating gene expression in B cells , and it has been shown that gammaherpesviruses have evolved multiple strategies for manipulating NF-κB activity . However , to date there has been no reported examination of the role of NF-κB in the establishment of chronic gammaherpesvirus infection in vivo . Murine gammaherpesvirus 68 ( γHV68 ) infects rodents and shares genetic and biologic properties with the human gammaherpesviruses , Epstein-Barr virus and Kaposi sarcoma–associated herpesvirus . To selectively block the function of NF-κB in infected cells , we engineered a transgenic virus that expresses a repressor of NF-κB activation ( IκBαM ) . Notably , this recombinant virus was defective in the establishment of latency in B cells in the lungs and spleen following intranasal inoculation . We also observed that the decrease in B-cell infection could not be rescued by forced expression of the cellular Bcl-2 protein , which is normally upregulated by NF-κB and serves to protect B cells from some forms of cell death . Thus , we conclude that NF-κB is an important host factor for the successful establishment of a chronic infection by gammaherpesviruses , and likely requires functions of NF-κB apart from its role in B-cell survival . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods",
"Supporting",
"Information"
] | [
"mus",
"(mouse)",
"viruses",
"virology"
] | 2007 | Inhibition of NF-κB Activation In Vivo Impairs Establishment of Gammaherpesvirus Latency |
Human T-lymphotropic virus type 1 ( HTLV-1 ) is a retrovirus that persists lifelong in the host . In ∼4% of infected people , HTLV-1 causes a chronic disabling neuroinflammatory disease known as HTLV-1-associated myelopathy/tropical spastic paraparesis ( HAM/TSP ) . The pathogenesis of HAM/TSP is unknown and treatment remains ineffective . We used gene expression microarrays followed by flow cytometric and functional assays to investigate global changes in blood transcriptional profiles of HTLV-1-infected and seronegative individuals . We found that perturbations of the p53 signaling pathway were a hallmark of HTLV-1 infection . In contrast , a subset of interferon ( IFN ) -stimulated genes was over-expressed in patients with HAM/TSP but not in asymptomatic HTLV-1 carriers or patients with the clinically similar disease multiple sclerosis . The IFN-inducible signature was present in all circulating leukocytes and its intensity correlated with the clinical severity of HAM/TSP . Leukocytes from patients with HAM/TSP were primed to respond strongly to stimulation with exogenous IFN . However , while type I IFN suppressed expression of the HTLV-1 structural protein Gag it failed to suppress the highly immunogenic viral transcriptional transactivator Tax . We conclude that over-expression of a subset of IFN-stimulated genes in chronic HTLV-1 infection does not constitute an efficient host response but instead contributes to the development of HAM/TSP .
HTLV-1 is an exogenous retrovirus that is widespread in the tropics and subtropics [1] , [2] . Chronic infection is characterised by a strong humoral and cellular immune response against the virus [3] . Whilst the majority of HTLV-1 carriers efficiently control the infection and remain clinically asymptomatic lifelong , some 1–4% develop a neurodegenerative inflammatory disorder known as HAM/TSP [1] . Here , mononuclear cell infiltrates in the central nervous system are associated with myelin and axonal destruction [1] , [4] . Clinically , patients present with multiple sclerosis-like symptoms including progressive spastic paraparesis , sensory loss and disturbances of bladder or bowel function [1] , [5] . The sequence and nature of the events that lead to the observed neuronal damage in HAM/TSP are not known and because of the lack of a suitable animal model a direct investigation of pathogenic mechanisms is not possible . Thus , clinical management of HAM/TSP is unsatisfactory and current treatment remains only palliative . To date , the strongest correlate of risk of HAM/TSP remains a high HTLV-1 proviral load [6] , i . e . the percentage of peripheral blood mononuclear cells that carry the provirus . The proviral load can differ among individuals by 1000-fold; the mean proviral load is ∼15-fold higher in patients with HAM/TSP than in asymptomatic HTLV-1 carriers ( ACs ) [6] . However , a significant proportion of infected individuals who present with high HTLV-1 proviral loads never develop the inflammatory disease , suggesting that additional factors contribute to the pathogenesis of HAM/TSP . Although other risk factors have been proposed , few differ systematically between ACs and patients with HAM/TSP after normalisation for proviral load: the frequency of certain lymphocyte subsets ( HTLV-1-specific CD4+ T cells [7]; FoxP3+CD4+ T cells [8]; natural killer ( NK ) cells [9] ) ; the level of expression of HTLV-1 genes [10] and the pattern of integration of the HTLV-1 provirus in the host cell genome [11] . However , the contribution of each of these factors to the pathology of HAM/TSP is not known . Blood transcriptional profiles have revealed previously unsuspected pathways of pathogenesis in autoimmune and infectious diseases [12] , [13] . Exposure to host or pathogen-derived immunogenic factors trigger changes in gene expression in peripheral blood leukocyte populations that reflect the host immune response mounted against pathogens as well as pathologic conditions of the immune system [14] . Thus , patient blood samples provide an easily accessible source of immunophenotypic information and are of particular value in HTLV-1 infection , in which CD4+ T cells constitute the major viral reservoir [15] . Previous studies on gene expression in HTLV-1 infection have focused on specific cell subpopulations ( HTLV-1-infected cells , CD4+ T cells , CD8+ T cells ) [16]–[21] . In contrast , whole blood studies integrate gene expression profiles of all leukocytes . Thus , the resulting transcriptional profiles provide a comprehensive overview of the status of the immune system and associated pathological changes . In this study , we aimed to identify the principal biological pathways and cell types that are deregulated in HTLV-1 infection and HAM/TSP by quantifying and analyzing differences in blood gene expression patterns of HTLV-1-infected individuals and seronegative controls . Unexpectedly , although the class I-restricted T cell response plays a pivotal role in controlling viral replication and risk of disease [22] , [23] , the outstanding differences in the transcriptional patterns lay in IFN-stimulated genes . We found that over-expression of a distinct subset of IFN-stimulated genes – a ‘transcriptional signature’ – was associated with presence of the inflammatory disease and positively correlated with clinical severity of HAM/TSP . This IFN-inducible transcriptional signature was absent in patients with multiple sclerosis , suggesting distinct pathogenetic pathways in the two diseases despite their clinical similarities . We conclude that IFNs do not efficiently control chronic HTLV-1 infection but may instead contribute to HAM/TSP pathogenesis . Our results identify IFN signaling as a key factor and a novel therapeutic target in HAM/TSP .
Gene expression profiles were generated from blood samples taken from patients with HAM/TSP , ACs and HTLV-1-seronegative individuals ( Table S1 A ) . To identify a transcriptional signature that reflects HTLV-1 infection per se irrespective of clinical phenotype , we compared gene expression profiles between uninfected and all HTLV-1-positive individuals ( Figure S1 ) . HTLV-1 infection resulted in altered blood expression levels of 542 genes ( Table S2 ) . The majority of identified gene products were involved in cell cycle control , cell proliferation and anti-viral immune responses: the most strongly associated cellular process was the p53 signaling pathway ( Figure 1 A ) . In accordance with previous in vitro findings [24] , [25] , mediators of cell cycle arrest and apoptosis were over-represented in patients with HAM/TSP while expression of molecules in the DNA damage response pathways was inhibited ( Figure 1 B–D ) . Differential gene expression in HTLV-1 infection is driven both directly , by the viral infection of T cells ( in proportion to the proviral load , i . e . the percentage of HTLV-1-infected PBMCs ) , and by the presence or absence of the inflammatory disease HAM/TSP . To identify genes associated with HAM/TSP whose expression varied independently of proviral load , we subdivided ACs into two groups: those with high proviral load ( ≥1% PBMCs ) or low proviral load ( <1% PBMCs ) . A distinct 80-gene blood transcriptional signature in patients with HAM/TSP was identified by non-parametric group comparison ( Table S3 ) . Hierarchical clustering analysis based on similarity in gene expression grouped individuals into two clusters associated with presence or absence of the inflammatory disease ( two-tailed Fisher's exact test: P<0 . 0001 ) but not gender , age and ethnicity ( Figure 2 A ) . To validate the HAM/TSP transcriptional signature identified in the first cohort ( training set ) , we applied the list of 80 genes to a second , independently collected and processed patient cohort ( test set; Figure 2 B; Table S1 B ) . Based on the expression levels of these 80 genes , k-nearest neighbor class prediction identified patients with HAM/TSP in the test set with a sensitivity of 70% and specificity of 92%; a support vector machine approach gave similar results ( Table S4 ) . Two ACs in the training set and three in the test set were misclassified as HAM/TSP , representing 10% and 17% of the respective asymptomatic cohorts . However , three of these five individuals had previously presented with urticaria and urinary urgency , which are associated with onset of HAM/TSP , and with episodes of uveitis , and other inflammatory conditions associated with HTLV-1 [26] . These observations also suggest that the transcriptional signature is specifically associated with HTLV-1-associated inflammatory disease , rather than viral infection alone . Transcriptional changes in the 80-gene signature were quantified using the ‘weighted molecular distance to health’ ( WMDH ) statistic [12] which measures the difference in expression of specified genes between patients and controls . The WMDH was significantly higher in patients with HAM/TSP than in ACs with high or low proviral load , confirming the association between the transcriptional signature and the inflammatory disease ( Figure 2 C ) . Two patients diagnosed with HAM/TSP clustered with uninfected and asymptomatic individuals in the test set ( Figure 2 B: filled triangle , empty square ) . To test whether this observation reflected heterogeneity in disease severity , we grouped patients according to their performance in the 10-meter timed walk , a test of the patient's mobility , on the day the blood sample was taken ( Figure 2 D ) . Indeed , the two outlier patients had only mild or moderate impairment of mobility . In the whole cohort , the WMDH was significantly greater in patients who took >30 s to walk 10 m or required wheelchairs compared with less disabled HAM/TSP patients ( healthy controls <8 s ) . Thus , the degree of molecular perturbation of the transcriptional signature reflected the clinical severity of HAM/TSP ( Figure 2 D ) . Clinically , HAM/TSP closely resembles certain forms of multiple sclerosis; both diseases show neurological abnormalities due to axonal lesions in the spinal cord and brain [1] , [5] . We therefore compared the present data with the recently reported whole blood gene-expression profiles from 99 patients with multiple sclerosis [27] . Employing the k-nearest neighbor class prediction approach , using the 80-gene list , none of the multiple sclerosis patients were classified as having HAM/TSP . Thus , the signature distinguished HAM/TSP from the clinically similar condition multiple sclerosis ( Figure 3 A ) . To explore the biological function of the observed transcriptional changes in HAM/TSP , we employed a modular analysis framework [28] that compares transcript abundance between patients and healthy control subjects in 28 groups ( “modules” ) of co-regulated genes , the majority of which have been shown to reflect specific cell populations or biological processes . The modular blood signature ( Figure 3 B ) revealed an over-expression of IFN-stimulated genes in HAM/TSP that was absent in asymptomatic virus carriers ( module M3 . 1 ) . Differentially expressed genes included key proteins of IFN signaling ( STAT1 , STAT2 ) , genes associated with antigen processing ( TAP-1 ) , cell migration ( CXCL10 ) and the innate antiviral response ( IFI35 ) . Transcriptional changes in the IFN-stimulated genes were reproduced in the test set and were independent of proviral load ( Figure S2 ) . A separate pathway-focused analysis of the 80-gene blood transcriptional signature , using Ingenuity Systems Pathway Analysis software , independently confirmed IFN signaling as the dominant pathway associated with HAM/TSP ( Figure 3 C ) . Here , IFN-stimulated transcripts constituted 29% ( 23 genes ) of the 80 genes , and included genes inducible by both type I ( IFN- α/β ) and type II ( IFN-γ ) IFNs . To test whether the expression of the distinct subset of IFN-stimulated genes in the 80-gene blood transcriptional signature reflects common immune responses observed in several diseases , we quantified expression of these genes in bacterial and other inflammatory diseases ( Figure S3 A ) . No significant transcriptional changes in IFN-stimulated genes were detected in patients with multiple sclerosis , Still's disease or individuals infected with Staphylococcus or group A Streptococcus . Expression levels in adults or children with systemic lupus erythematosus ( SLE ) were similar to those observed in patients with HAM/TSP; however , both adult and pediatric SLE differentially expressed 92–95% of IFN-stimulated genes comprised in the IFN module ( M3 . 1 ) , whilst the IFN response in patients with HAM/TSP was limited to a small subset ( Figure S3 B ) . We conclude that in contrast to the broad IFN-stimulated gene expression pattern in SLE , the IFN-mediated immune response in patients with HAM/TSP is restricted to a distinct subset of IFN-stimulated genes . To confirm that the observed HAM/TSP transcriptional signature resulted from differential transcriptional regulation rather than differences in cell composition , we quantified frequencies of peripheral leukocyte subsets in HTLV-1 carriers and control subjects by flow cytometry . No significant differences in the abundance of antigen-presenting or major effector cell populations were observed between the study groups ( Figure S4 ) . CD4+ T cells constitute the main reservoir for HTLV-1 in vivo . To test whether the observed IFN-inducible signature was restricted to HTLV-1-infected CD4+ T cells we selected genes with the highest association with HAM/TSP ( 11 genes; see Methods ) and quantified their expression levels in separated blood leukocyte populations by quantitative RT-PCR ( Figure 4 A ) . The IFN-inducible signature was present in neutrophils and monocytes ( Fisher's method of combining probabilities: P = 0 . 0271 and P = 0 . 0260 respectively ) but not in CD4+ or CD8+ T cells ( P = 0 . 3 and P = 0 . 1 , respectively , Figure 4 A ) . The mRNA levels of four representative IFN-stimulated genes ( STAT1 , CD64 , FAS and CXCL10 , Figure S5 ) correlated well with their protein levels , thus confirming the presence of the IFN-inducible signature at the protein level . Furthermore , increased protein levels of STAT1 , a key molecule in IFN signaling , were detected by flow cytometry in all peripheral blood effector cell populations from patients with HAM/TSP , but not in ACs or uninfected controls ( Figure 4 B ) . These results suggest that the observed IFN-inducible transcriptional signature in whole blood did not result from direct activation of IFN-mediated anti-viral pathways in infected CD4+ T cells but rather that all cell types had been systemically exposed to IFNs in vivo . To identify the molecular basis of the IFN-inducible signature , we tested for abnormal production of IFNs and abnormal responses to IFNs in patients with HAM/TSP . In agreement with previous findings [29] , [30] we found that protein levels of the IFN-γ-inducible chemokine CXCL10 were significantly increased in plasma from patients with HAM/TSP compared to ACs; however , IFN-α2 , IFN-β , IFN-λ and IFN-γ protein levels were not elevated ( Figure S6 ) . There were no differences in the production of IFN-α by stimulated plasmacytoid dendritic cells in short-term whole blood ex vivo assays . The frequency of CD8+ T cells producing IFN-γ in patients with HAM/TSP was greater than that in uninfected controls , but did not differ significantly from the frequency observed in ACs ( Figure S7 ) . Next we analyzed molecules of the IFN signaling pathway that are involved in IFN-responsiveness . Surface expression levels of IFN-α and IFN-γ receptors on peripheral leukocyte populations were similar in all study groups ( Figure S8 ) . We then measured phosphorylated STAT1 ( p-STAT1 ) by flow cytometry as a marker of activation of the IFN signaling pathway at the single-cell level . Stimulation with recombinant IFNs ( particularly IFN-γ ) led to an increase in p-STAT1 levels in all study groups ( Figure 5 A–B ) . Patients with HAM/TSP had higher STAT1 protein baseline levels which after IFN stimulation resulted in significantly higher p-STAT1 levels compared to ACs or uninfected controls . Expression of STAT1 itself is upregulated by IFNs [31]; it is likely that exposure to IFNs in vivo induces higher STAT1 protein and thus p-STAT1 levels which makes leukocytes from patients with HAM/TSP more sensitive to IFNs . IFNs activate several intracellular antiviral responses and are typically associated with a beneficial effect on the outcome of viral infection [32] . To examine the effect of IFNs on HTLV-1 protein expression , we quantified expression of HTLV-1 Tax and Gag after treatment with exogenous IFN-α and IFN-γ . Interestingly , whilst IFN-α treatment partially inhibited the expression of the HTLV-1 protein Gag , it could not suppress production of the pleiotropic viral transcriptional transactivator Tax ( Figure 5 C–D ) . In contrast , IFN-γ treatment did not alter the expression of either Gag or Tax proteins in infected CD4+ T cells ( Figure S9 A , B ) . Exogenous IFN-α and IFN-γ increased HBZ mRNA levels but this did not reach statistical significance ( Figure 5 E; Figure S9 C , D ) .
Despite the growing understanding of the mechanisms of HTLV-1 persistence , little is known about the pathogenesis of HAM/TSP and importantly , what distinguishes hosts who develop the disease from those who remain asymptomatic . The aim of this study was to identify biological pathways and molecules in whole blood gene expression profiles of asymptomatic HTLV-1 carriers and patients with HAM/TSP to generate new hypotheses on the mechanisms of viral persistence and the pathogenesis of HAM/TSP pathology . We found that the presence of HAM/TSP was associated with over-expression of a distinct subset of IFN-stimulated genes in circulating leukocytes . The IFN-inducible signature was absent in healthy HTLV-1 carriers and correlated positively with the clinical severity of the inflammatory disease . Gene expression signatures may reflect either differences in the frequencies of specific cell populations or a difference in gene expression at the single-cell level . We observed no significant differences in the frequencies of the blood cell populations ( Figure S4 ) : we conclude that the observed IFN-inducible signature resulted from a difference in the intensity of expression of IFN-stimulated genes at the single-cell level . Expression of IFN-stimulated genes was evident in neutrophils and monocytes , but not in T cells , which are the cellular reservoir of HTLV-1 in vivo [15] . It is therefore likely that the HAM/TSP-specific transcriptional profile originated from exposure to IFNs in vivo rather than direct activation of the IFN-response pathway in infected cells . The IFN-stimulated genes identified in HAM/TSP can be induced by both type I and type II IFNs [33] and we did not detect abnormal plasma levels of type I or type II IFNs in patients with HAM/TSP , perhaps because of their short half-lives in blood [34] . However , CD8+ T cells from patients with HAM/TSP produced high levels of IFN-γ in response to a non-specific stimulus ex vivo , which may reflect higher local production of IFN-γ in vivo . IFN-stimulated genes have been shown to mediate potent antiviral and neuroprotective effects [32] , [35] . However , several recent studies have implicated IFNs in the pathogenesis of autoimmune disorders such as rheumatoid arthritis [36] , systemic lupus erythematosus [37] , systemic sclerosis [38] and Sjögren's syndrome [39] . It has been hypothesized that IFNs trigger autoimmunity by promoting the activation and survival of autoreactive T and B cells in genetically predisposed individuals [40] . However , there is little evidence for antigen-specific autoreactivity in HAM/TSP [41]; instead , the risk of disease is linked to high frequencies of poor quality antiviral cytotoxic T cells [42]; i . e . low avidity of HTLV-1 antigen recognition and reduced lytic efficiency . IFNs may contribute to immunopathological processes in the central nervous system by stimulating and propagating a detrimental immune response to HTLV-1; however , the precise pathways responsible have yet to be determined . Despite similarities in clinical presentation the gene expression profiles of HAM/TSP and multiple sclerosis were distinct . The IFN-inducible signature present in HAM/TSP was absent in multiple sclerosis , suggesting that different pathogenetic mechanisms contribute to the axonal damage observed in both diseases . A recent study reported the expression of a subset of type I IFN-stimulated genes in 50% of patients with a particular clinical subtype of multiple sclerosis ( relapsing-remitting ) [43] . However , in agreement with our findings there was little overlap between the IFN-stimulated genes identified in HAM/TSP and the genes whose expression was altered in multiple sclerosis . Furthermore , we found that different subsets of IFN-stimulated genes are actively expressed in other diseases with IFN-inducible signatures ( Figure S3 ) . There are more than 300 known IFN-stimulated genes , the individual actions of many of which are largely uncharacterized . Thus , it is conceivable that there are cell-type- , pathogen- and disease-specific patterns of IFN-stimulated gene expression that contribute to a beneficial or detrimental outcome of immunological processes . Type I IFN therapy is effective in controlling certain persistent viral infections [44] , and there is evidence that it can slow down disease progression in multiple sclerosis [45] . However , clinical trials assessing the benefit of IFN-α and IFN-β therapy in the treatment of HAM/TSP have been less successful [46]; few patients showed any improvement and the overall clinical benefit was limited . Although a beneficial net effect of systemically administered IFN cannot be excluded , two findings in the present study support the clinical observation that IFNs do not effectively control HTLV-1 infection: ( i ) over-expression of IFN-stimulated genes was not observed in healthy HTLV-1 carriers who efficiently control the infection; ( ii ) while IFN-α reduced the expression of the structural viral protein Gag as reported previously [47] , we found that IFN treatment did not affect the expression of the viral transactivator protein Tax or HBZ mRNA . HTLV-1 Tax is essential to the viral life cycle as it has been shown to drive viral replication , promote host cell proliferation and viral cell-to-cell spread [48] , [49] . HBZ has been shown to down-modulate Tax function [50] , [51] whilst promoting infectivity and viral persistence [52] . A recent study has linked elevated HBZ mRNA levels to increased clinical severity of HAM/TSP [53] . We found that IFN treatment increased HBZ expression but this observation did not reach statistical significance . We conclude that although type I IFN treatment may decrease the production of viral structural complexes , i . e . capsids , it does not completely abrogate viral activity and proliferation of infected cells . In addition , HTLV-1 viral proteins have been reported to manipulate and evade the IFN response by suppressing interferon regulatory factors and up-regulating suppressors of IFN signaling [16] , [54] , [55] . Perturbations of the p53 signaling pathway were evident in all HTLV-1 infected individuals , irrespective of disease status . The p53 signaling pathway prevents cell proliferation of genomically unstable cells by promoting cell cycle arrest , senescence and apoptosis [56] . Altered p53 signaling could reflect a highly active immune response with rapid cell turnover and clonal expansion . However , abnormal p53 signaling has been implicated in HTLV-1-mediated leukemogenesis [49] . The p53 protein itself is usually intact in HTLV-1 leukemic cells , but there is evidence that the virus inhibits p53-induced anti-viral responses by targeting other components of the p53 signaling pathway [57] , [58] . This hypothesis is supported by our finding that HTLV-1-positivity was associated with low mRNA levels of molecules involved in the detection of DNA damage , i . e . the ataxia telangiectasia mutated protein ( ATM ) and the DNA-dependent protein kinase ( PRKDC ) . The protein products of these genes sense dsDNA breaks due to viral integration into the cell's genome and induce cell cycle arrest and/or apoptosis [59] . Decreased expression of ATM and PRKDC could therefore facilitate the survival of genomically unstable HTLV-1-infected cells and allow premature DNA replication in the presence of genomic lesions , thus contributing to the oncogenic potential of HTLV-1 . In summary , our study revealed that over-expression of a specific subset of IFN-stimulated genes distinguishes patients with HAM/TSP from asymptomatic HTLV-1 carriers , which indicates an unexpected role of IFN-stimulated genes in the pathogenesis of central nervous system inflammation in HTLV-1 infection . This finding opens new avenues for research on HAM/TSP pathogenesis and new areas of therapeutic intervention .
All study participants have given written informed consent . Ethical approval for this study was obtained as follows: IRB approval numbers: St Mary's NHS Trust Local Research Committee EC no . 02 . 31 ( 27 . 06 . 2002 ) ; National Research Ethics Service , Oxfordshire REC C no . 09/H0606/106 ( 16 . 11 . 2009 ) . Consenting participants attended the HTLV clinic at St Mary's Hospital , London , UK . Deterioration of mobility in patients with HAM/TSP was assessed by a 10-meter timed walk ( time taken to walk 10 m with or without a walking aid ) . The extent of deterioration was categorized as ‘mild’ ( 8–15 s; n = 5 ) , ‘moderate’ ( 16–30 s; n = 6 ) or ‘severe’ ( >30 s or use of a wheelchair; n = 6 ) . HTLV-1 proviral load was measured as described previously [6] and expressed as percentage of peripheral blood mononuclear cells ( PBMCs ) . Blood was collected in Tempus tubes ( Applied Biosystems , Foster City , CA , USA ) . Total RNA and cDNA were prepared as described previously [12] and analyzed on HumanHT-12 V3 or WG6 V3 expression BeadChip arrays ( Illumina , San Diego , CA , USA ) . Raw gene expression values were normalized per gene to the median gene expression value across all samples and supervised non-parametric analysis was carried out in GeneSpring GX7 . 3 . 1 ( Agilent Technologies , Edinburgh , UK ) . A series of filters was applied to identify genes differentially expressed between ACs and patients with HAM/TSP ( Figure S2 ) . To identify the genes most strongly associated with HAM/TSP , we selected genes that were >1 . 5-fold differentially expressed versus ACs in >50% of patients with HAM/TSP ( 11 genes ) . To visualize transcriptional patterns , an iterative agglomerative clustering method was applied to gene lists ( algorithm: Pearson correlation ) and samples ( algorithm: Spearman correlation ) . Microarray data was deposited in the Gene expression Omnibus database ( accession number GSE29312 ) . Illumina HT12 V3 microarray data relating to other diseases has been published previously and is publically available: multiple sclerosis [27] , Still's disease [12] , Staphylococcus or group A Streptococcus infection [12] , adult and pediatric systemic lupus erythematosus [12] . Class prediction was performed within GeneSpring using the k-nearest neighbor algorithm ( neighbors = 10; P-value ratio cut off = 0 . 65 ) or support vector machine ( Gaussian radial basis , diagonal scaling = 3 ) . Canonical pathway analysis was performed using the web-based Ingenuity Systems Pathway Analysis software ( Ingenuity Systems Inc . , Redwood City , CA , USA ) . Modular transcriptional fingerprints [28] for the HTLV-1-positive study groups were generated comparing BeadStudio raw gene expression values to baseline median expression values of uninfected controls for all genes present in a module ( Student t-test , P<0 . 05 ) . The weighted molecular distance to health ( WMDH ) was calculated as described previously [12] to quantify global transcriptional changes relative to a pre-determined baseline value . IFN-stimulated transcripts in the 80-gene blood transcriptional signature were identified using the Interferome database ( www . interferome . org ) [33] . Neutrophils ( CD15+ ) followed by CD8+ T cells or monocytes ( CD14+ ) followed by CD4+ T cells were isolated from whole blood using antibody-coupled magnetic whole blood beads ( Miltenyi Biotec , Bergisch Gladbach , Germany ) . RNA from blood and isolated cell populations was purified using the QIAamp RNA Blood Mini Kit ( QIAGEN , Hilden , Germany ) . Total cDNA was generated from 1 µg of RNA using the SuperScript III First-Strand Synthesis Supermix ( Invitrogen , Paisley , UK ) . Custom Taqman Low Density Array cards ( Applied Biosystems , Carlsbad , CA , USA ) comprised probes for the top 11 genes associated with HAM/TSP ( Table S5 ) and six reference genes for normalization ( B2M , ACTB , HMBS , GNB2L1 , RPLP0 , 18S ) . qRT-PCR was performed on a 7900HT fast real-time system ( Applied Biosystems ) . Ct values were extracted and analyzed using SDS2 . 3 software ( Applied Biosystems ) . Plasma concentrations of IFNs were measured by Luminex using a LEGENDplex assay kit according to the manufacturer's instructions ( Bioplex , Biolegend , San Diego , USA ) . Bioactivity of type I and III IFNs in plasma samples was assessed using the iLite Human Interferon Alpha Kit ( Neutekbio , Galway , Ireland ) . Fresh blood was stained with monoclonal antibodies for population specific markers ( neutrophils: CD45+CD14−CD16+ , monocytes: CD45+CD14+CD16− , B cells: CD45+CD19+CD20+ , T cells: CD45+CD3+CD4+ or CD8+ , NK cells: CD45+CD3−CD56+CD16+ ) and CD64 or FAS antibodies for 15 min at room temperature ( RT ) . Following lysis of erythrocytes with BD FACS Lysing solution ( BD Biosciences , Franklin Lakes , NJ , USA ) , samples were acquired on a CyAn ADP flow cytometer ( Beckman Coulter , Marseille , France ) and analyzed using FlowJo software ( TreeStar , Ashland , USA ) . All antibodies were purchased from eBioscience , Beckman Coulter , BD Pharmingen and Miltenyi Biotec . IFN-α secretion assay: blood samples were stimulated with 10 µM R-848 ( TLR-7/8 agonist; Insight Biotech , Wembley , UK ) for 3 h at 37°C . IFN-α secreted by plasmacytoid dendritic cells ( BDCA-2+CD123+ ) was measured using the Miltenyi Human IFN-α Secretion Assay Detection Kit . Intracellular IFN-γ production: blood samples were stimulated with 10 ng/ml PMA , 0 . 5 µg/ml calcimycin ( Sigma , Gillingham , UK ) in the presence of monensin ( eBioscience , San Diego , CA , USA ) for 4 h at 37°C and analyzed by flow cytometry . Fresh whole blood samples were incubated with IFNGR2-Fluorescein ( R&D Systems , Minneapolis , MN , USA ) , IFNGR1-PE ( eBioscience ) or IFNAR1-Fluorescein and IFNAR2-PE ( PBL Interferon Source , Piscataway , NJ , USA ) and immune cell population markers for 25 min at RT . Following lysis of erythrocytes with BD FACS Lysing solution ( BD Biosciences ) , samples were acquired on a CyAn ADP flow cytometer ( Beckman Coulter ) and analyzed using FlowJo software ( TreeStar ) . Heparinized whole blood was treated with 1000 IU/ml of recombinant IFN-α2a or IFN-γ ( PBL Interferon Source ) for 20 min at 37°C . Detection of STAT1 phosphorylation by flow cytometry was adapted for whole blood following the specifications refined by Chow and colleagues [60] . CD8+ T cells were depleted from PBMCs by positive selection using antibody-coated magnetic microbeads ( Miltenyi Biotec ) to prevent cytotoxic T cell-mediated killing of HTLV-1 expressing cells . Cells were incubated in the presence of 100 IU/ML IFN-α , 100 IU/ml IFN-γ or PBS in RPMI 1640 medium ( Sigma ) supplemented with 10% FCS , 2 mM L-Glutamine , 100 U/ml Penicillin and 100 µg/ml Streptomycin ( Life Technologies ) at 37°C in 5% CO2 . For HBZ mRNA expression cells were harvested after 0 , 2 , 4 , 8 , and 16 h and RNA extracted using the RNeasy kit ( QIAGEN , Hilden , Germany ) according to the manufacturer's instructions . The cDNA was produced using the QuantiTect Reverse Transcription kit ( QIAGEN ) and qRT-PCR was performed in 96-well plates on a 7900HT fast real-time system ( Applied Biosystems ) . Primer sequences detected spliced HBZ mRNA: forward primer 5′-GGACGCAGTTCAGGAGGCAC-3′ , reverse primer 5′-CCTCCAAGGATAATAGCCCG-3′ . Ct values were extracted and analyzed using SDS2 . 3 software ( Applied Biosystems ) . HBZ mRNA levels were normalised to endogenous GAPDH levels ( primers: forward 5′-CTTTGGTATCGTGGAAGGACTC-3′ , reverse 5′-GTAGAGGCAGGGATGATGTTCT-3′ . For Tax and Gag protein detection cells were fixed and permeabilized after 16 h using the FoxP3 staining buffer set from eBioscience and stained for CD3 , CD4 , CD8 , HTLV-1 Tax ( LT4 antibody ) [61] and HTLV-1 Gag ( Gin7 antibody ) [62] and analyzed by flow cytometry . Correlations were tested by Spearman's rank-order and differences between two study groups by Mann-Whitney U test . Pairwise comparisons were tested using the Wilcoxon signed-rank test . All tests were two-tailed . Box plots are depicted as median with error bars within the 1 . 5 interquartile range ( IQR ) ; bar graphs depict median with error bars representing the standard error of the mean ( SEM ) . Statistical programs used included SPSS version 17; Microsoft Excel 2007; GeneSpring GX7 . 3 . 1 and the DiagnosisMed package in R . | Infection with the Human T Lymphotropic virus is widespread in the tropics and subtropics , where it causes a chronic disabling disease of the nervous system abbreviated as HAM/TSP . There is no effective treatment available for HAM/TSP , because it is not understood how the virus causes the neuronal damage that results in the clinical symptoms of weakness and paralysis of the legs . Here , we compared the frequencies of cell populations and gene expression profiles from diseased and asymptomatic HTLV-1 carriers to identify abnormalities in biological pathways that cause HAM/TSP . We discovered a distinct group of genes that is over-expressed in white blood cells in patients with HAM/TSP , but not asymptomatic HTLV-1 carriers or patients with the clinically similar disease multiple sclerosis . The expression of these genes is induced by interferons , a group of anti-viral proteins that are usually beneficial to the host but can also cause inflammation . We also found that interferons did not efficiently suppress HTLV-1 protein expression in vitro . We conclude that interferons do not control chronic HTLV-1 infection but instead contribute to the development of HAM/TSP . Our study provides new insights into the development of HTLV-1-associated diseases and opens new areas of therapeutic intervention . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"medicine",
"infectious",
"diseases",
"clinical",
"immunology",
"virology",
"immunology",
"biology",
"microbiology"
] | 2012 | Systems Biology Approaches Reveal a Specific Interferon-Inducible Signature in HTLV-1 Associated Myelopathy |
The formation of gaps in the endothelium is a crucial process underlying both cancer and immune cell extravasation , contributing to the functioning of the immune system during infection , the unfavorable development of chronic inflammation and tumor metastasis . Here , we present a stochastic-mechanical multiscale model of an endothelial cell monolayer and show that the dynamic nature of the endothelium leads to spontaneous gap formation , even without intervention from the transmigrating cells . These gaps preferentially appear at the vertices between three endothelial cells , as opposed to the border between two cells . We quantify the frequency and lifetime of these gaps , and validate our predictions experimentally . Interestingly , we find experimentally that cancer cells also preferentially extravasate at vertices , even when they first arrest on borders . This suggests that extravasating cells , rather than initially signaling to the endothelium , might exploit the autonomously forming gaps in the endothelium to initiate transmigration .
We present a novel model of an endothelial cell ( EC ) monolayer that incorporates different intracellular mechanical structures and dynamical cell-cell adhesions . The intracellular mechanical state is determined by radial contractile actin stress fibers and the cell membrane together with the actin cortex . For simplicity , we combined membrane and cortex into single viscoelastic elements , composed of an elastic spring and a viscous damper , that we refer to , from now on , as membrane elements . The radial stress fibers are also modeled by viscoelastic elements with different mechanical properties from the membrane , similar to a model of epithelial cells [28] ( see Fig 1A ) . Neighboring cells may form cell-cell adhesions at adjacent nodes , and the resulting adhesion bond is modeled through a spring . The passive mechanical properties of the monolayer are thus modeled through a network of connected elastic and viscoelastic elements , similar to models of epithelial sheets [29 , 28] . Since we are interested in studying the opening dynamics of gaps in the endothelial barrier , we explicitly simulate the dynamical binding of adhesion complexes . Contractions represent myosin motor activity that is known to exhibit randomness [33] , so we employ Monte-Carlo simulations to estimate the occurrence of such forces as well as that of protrusive forces due to actin polymerization . The forces are then redistributed across the network of connected viscoelastic elements . Cell-cell adhesion complexes that mechanically link neighboring cells can dynamically bind and unbind in a force-dependent manner . The adhesion complexes in the model provide an effective description of both bonds of cell-cell adhesion molecules ( such as VE-cadherin ) and bonds of these adhesion molecules to the cytoskeleton . Cadherins and adhesion-cytoskeleton bonds are known to increase their binding strength in response to smaller forces , before they ultimately rupture [34] . This catch-bond type behavior is included in our model , and unbinding is thus simulated through a force-dependent Monte-Carlo simulation . Moreover , the number of VE-cadherins in an adhesion complex is modeled through a force-dependent adhesion clustering mechanism , as described in [18 , 23 , 35 , 36 , 37] . A more detailed description of the mathematical model and its numerical implementation is given in S1 Text . We employ our endothelial monolayer model to explore the dynamics of endothelial cell junctions . We predict the frequency , size and duration of gaps , as well as the preferred geometrical locations of the gap formation , and compare the predictions with our experimental measurements . The parameters used in the simulations are detailed in S1 Table . After comparing our predictions with the experimental results , we perform sensitivity analyses to investigate how cell mechanical properties , cell-cell adhesion characteristics and myosin generated forces regulate the formation , lifetime and size of gaps in the endothelium . Here we present a summary of the major parameters of the model that had a significant impact on our model behavior , and were consequently thoroughly investigated through sensitivity analysis in the remainder of this paper . Table 1 lists all these parameters , and for a complete list and discussion see the Supporting Information . The main parameters investigated are related to cell mechanical properties , adhesion properties or myosin force generated processes . Cell mechanical properties are dictated by stress fiber stiffness ( Ksf ) , membrane stiffness ( Kmemb ) and bending stiffness ( incorporated through a rotational spring constant , Kbend ) . Stress fiber stiffness controls the rigidity of the interior of the cell , whereas membrane stiffness controls the rigidity of the membrane and the adjacent actin cortex . Bending stiffness acts on the membrane nodes depending on the relative orientation between the edges connecting at a given node . Adhesion properties are controlled by the mechanical properties of the adhesion complexes and their binding and unbinding rates . Adhesion complex mechanics are modeled by linear springs , controlled by their stiffness constant , K a d h 0 . The binding rate depends on distance and can be controlled by the adhesion complex density , ρadh . We then model the reinforcement of a bond that is already formed by the additional recruitment of adhesive proteins into the bond . Reinforcement is force dependent and can be controlled by the binding rate constant for adhesion reinforcement , k r e i n f 0 . Unbinding follows a catch bond behavior . The catch bond unbinding curve can be modified through two rate coefficients: k s 0 , which represents a slip bond , and k c 0 , which is the additional parameter characterizing the initial increase in the bond lifetime with force ( see S12 Fig ) . Then , the model includes contractile forces due to myosin motor activity , and protrusive forces that may arise due to actin polymerization . These forces can be directed radially ( following the stress fibers direction ) or in a tangential direction ( following membrane segments ) . In the sensitivity analysis we have varied the magnitude of contraction forces in the radial direction ( FRadial ) and in the tangential direction ( Fcortex ) .
Fig 1B and 1C and S1 Movie show typical simulations of the monolayer dynamics of the computational model . We observe that gaps open preferentially at vertices , i . e . the intersections of three or more cells , as opposed to the border between two cells . We have quantified this by counting the total number of gaps formed as well as their lifetime at borders and vertices of the cell in the center of the monolayer , and showed that our model predictions are in line with the experimental observations ( Fig 1G and 1H ) . These experiments were performed by seeding HUVEC cells on glass , where they formed a continuous monolayer . The gaps were experimentally quantified through inspection of visible gaps within the VE-cadherin-GFP signal in the monolayer ( arrows in Fig 1E and 1F ) . Controls simultaneously showing VE-cadherin-GFP and CD31 staining show that the VE-cadherin gaps are also visible in the CD31 staining , indicating that the VE-cadherin gaps correspond to real physical gaps between two or more cells S7 Fig ( see Methods for further details of the experimental setup and quantification ) . Vertices are points where more than two cells exert forces and where tangential force components naturally propagate to . Therefore , it is expected that stress concentrates at the three cell vertex rather than at the two cells borders , and the simulations confirm this hypothesis ( Supplementary S9 Fig and S3 Movie ) . The forces on adhesion clusters at the vertices are thus more likely to exceed the corresponding force of maximal lifetime of the bonds , as will be discussed in more detail below . We study how variations in the mechanical properties of the cells , the cell-cell adhesion complexes or force variations affect the rate of gap formation . Fig 2A and 2B show how passive mechanical properties of the cell affect both the frequency ( Fig 2A ) and the location of the gap openings ( Fig 2B ) . Increasing stiffness of either the membrane or the stress fibers provokes a decrement of the gap generation frequency ( Fig 2A and S4 and S5 Movies ) . This is intuitive , since increasing stiffness stabilizes the movements of cells and makes the monolayer less dynamic . On the other hand , the location of the gap openings ( i . e . whether they occur at a vertex or border ) is critically affected by membrane stiffness at low values , until it stabilizes for intermediate and high membrane stiffness . In contrast , stress fiber stiffness affects gap location for very high stiffness , where gaps are almost fully prevented from opening at the borders ( Fig 2B ) . Interestingly , increasing bending stiffness first increases gap generation up to a maximum point , before it leads to a decrease in gap opening frequency ( Fig 2A ) . For small to intermediate bending stiffness , the frequency of gap openings increases , since bending stiffness is critical for effective force propagation between neighboring adhesion sites at vertices . When a single adhesion complex ruptures , bending stiffness leads to increased forces on neighboring adhesion complexes . After a peak in gap opening frequency at intermediate bending stiffness , a drop in the gap formation is observed for higher bending stiffness . This is caused by the resulting stabilization of the existing gaps at vertices . This high bending stiffness opposes sharp corners of the membrane at vertices and thus favors stable gaps that are permanently open , implying no new gaps are formed ( S7 Movie ) . On the other hand , at cell borders , a high bending stiffness implies that if a single adhesion cluster is ruptured , the forces on it are redistributed across many neighboring adhesion sites and this stabilizes the borders ( Fig 2B ) . Turning to the role of cell-cell adhesion complex properties , our model shows that as the junctions become more stable , gaps open less frequently ( Fig 2D ) . To increase cell-cell junction stability , we increase the mechanical stiffness of individual adhesion bonds , or the density of adhesion molecules . These results are in line with previous experimental work [14] , which reported that more stable cell-cell junctions result in fewer transmigrating cells . While the total number of gaps at either vertex or border decreases with increasing cell-cell adhesion complex stiffness or cell-cell adhesion density available for binding , we see that there are no significant differences between gaps generated at the vertex and gaps generated at the borders ( Fig 2E ) . Fig 2G and 2H show the impact of changing the cortical and radial forces , where the total force is kept constant ( when the radial force decreases , the cortical force is increased by the same magnitude ) . This is biologically relevant since cells are known to shift their cytoskeletal compartments in a context dependent manner [38] . In fact , cell monolayers subjected to shear flow have been reported to increase cortical actin while decreasing stress fibers [14] . Endothelial cells in particular , are known to exhibit both radial and tangential stress fibers with a different effect on gap opening dynamics [39] . As the force shifts from radial to cortical forces , total gap formation fluctuates with a slight increase as cortical forces increase ( Fig 2G ) . For high cortical forces , the gaps also clearly tend to localize more at the vertices ( Fig 2H ) . This is because contractions parallel to the membrane result in force concentrations at the vertices . For very high cortex forces , the typical stresses on adhesion clusters at the vertices may thus be higher than the force where the lifetime of catch bonds peaks ( Supplementary S12 Fig ) , explaining the small increase in the number of gaps formed ( Fig 2G ) . On the other hand , we will later show that these gaps formed at high cortical forces are typically small and have a short lifetime , limiting their potential for extravasation ( see Fig 3I and 3J ) . To take into account that molecular or physical perturbations may simultaneously affect multiple parameters , we now study how variations of pairs of these parameters at the same time may influence the monolayer integrity and the localization of the gap formation . Although , we have previously seen in Fig 2A that membrane and stress fiber stiffness have a similar effect on the gap opening frequency , in Fig 2C we can observe how the effect of varying stress fiber stiffness is clearly predominant over the effect of varying membrane stiffness . Fig 2F shows the impact of varying cell-cell adhesion stiffness and cell-cell adhesion complex density available for binding . Interestingly , there is a synergy between both parameters on regulating gap opening frequency , as evident through the curved shape of the levels of equal gap opening frequency ( Fig 2F ) . In Fig 2I we show the combined role of cortex and radial forces , thus not keeping total force fixed as in Fig 2G and 2H . This confirms that total force is the main driver of gap opening frequency , as opposed to a redistribution of forces between cortex and stress fibers ( Fig 2I ) . The lifetime and size of a gap are physical parameters that may also limit a cancer or immune cell’s potential to extravasate through the monolayer . Here , we show how the lifetime and size of a gap are influenced by cell mechanical and junction properties , without the presence of extravasating cells ( Fig 3 ) . We observe that membrane stiffness has a marginal influence on the life time of the gap , whereas increasing stress fiber stiffness clearly reduces the time that a gap is open and the gap size ( Fig 3A and 3B ) . Indeed , higher stress fiber stiffness will result in mechanical resistance to an opening gap and thus inhibit the propagation of the defect in the cell-cell junctions , leading to a stabilization of the monolayer ( see S4 and S5 Movies ) . The dominance of stress fiber stiffness over membrane stiffness in regulating lifetime and size remains valid in a broad range of parameter values ( Fig 3C and 3D ) . Interestingly , increasing bending stiffness to high values may increase gap lifetime ( Fig 3A ) . This is because higher bending stiffness will resist deviations from straight membranes . Thus , at straight borders , higher bending stiffness will resist gap openings whereas at vertices with high curvature , cells are more likely to adapt their shape resisting high curvature , thus favoring opened gaps . The dynamics of the monolayer for low bending stiffness is shown in S6 Movie . Fig 3E and 3G show that adhesion complex stiffness and density at low values do not have a big impact on lifetime , however as they increase , lifetime starts to decrease . Both stiffness and density have a similar effect , since the total stiffness of an adhesion complex depends on both density and single bond stiffness ( Eq . S10 ) . Higher stiffness of the adhesion complex leads to more passive mechanical resistance to gap openings , and this effect dominates for high stiffness . The level of noise due to repeats of our MC simulations is higher for these adhesion parameters than for the parameters determining cell mechanics . Likewise , for the gap size , the stabilizing effect of both adhesion complex stiffness and density dominates and leads to a reduction in gap size ( Fig 3F and 3H ) . However , the effect of increasing the density is slightly stronger than that of increasing single bond stiffness . This is because the density affects not only adhesion complex stiffness ( Eq . S10 ) , but also the rate of forming new adhesion complexes ( Eq . S9 ) and the rate of reinforcing existing bonds ( Eq . S11 ) . These effects together thus synergize to stabilize gaps and prevent them from growing too large . Earlier , we have shown that a shift in the force ( from radial to cortical ) produces an increment in gap formation ( Fig 2G ) . Fig 3I and 3J show that this shift in the force reduces gap lifetime and size . This indicates that , although the frequency of opening is increased , these gaps are smaller and last shorter in time which may reduce paracellular extravasation , as suggested in previous experimental work [14] . Combined changes of cortical and radial force show that although both kinds of forces are needed to increases gap size and lifetime , the impact of radial forces is clearly predominant over the impact of cortex forces ( Fig 3K and 3L ) . This is intuitive , since radial forces clearly separate cell borders generating bigger gaps and make them harder to close , whereas cortical forces distribute forces to vertex regions . This does not provoke large cell deformations , which is reflected in the low impact on the gap size and lifetime observed . In Fig 4A–4C we show the impact of varying the catch-bond unbinding parameter k c 0 that shifts the location of the peak of maximal lifetime of a single catch bond , while we maintain the actual maximum value through simultaneously shifting the slip-bond unbinding parameter k s 0 ( Eq . S12 and S12 Fig ) . We observe that for a pure slip bond ( corresponding to k c 0 = 0 ) , gaps occur at a higher rates than for small nonzero values of k c 0 . Increasing k c 0 further leads to a minimum in gap opening frequency , from which the frequency increases again . This minimum corresponds to a maximum of stability , where forces on the adhesion complexes are similar in magnitude to the peak of stability of the catch bond . Consequently , shifting the location of that peak even further towards higher forces ( by increasing k c 0 even further ) means we destabilize the catch bonds again . Note that the gap lifetime and size of gaps are much less influenced by the location of the catch bond maximum than the gap opening frequency . In Supplementary S11 Fig , we show histograms of the forces on adhesions comparing the number of bound clutches , the number of unbinding events , and the ratio of unbound to total bonds for slip bonds ( k c 0 = 0 ) to the catch bond with reference values ( k c 0 = 0 . 27 s - 1 ) . We see that adhesions in the catch bond case bear and disengage at higher forces than for the slip bond case , confirming that the typical forces on bonds are of such magnitude that the catch bond nature stabilizes the junctions . In Fig 4D–4F we modify the reinforcement binding rate k r e i n f 0 to check the influence of the reinforcement . This is different from the previous analysis where the adhesion complex density available for binding was changed , since now the binding probability based on distance is not affected ( Eq . S9 ) . However , we see the same trend of increasing stability with increasing k r e i n f 0 ( Fig 4D ) , in line with the result obtained from varying cadherin density ( Fig 2D ) , suggesting that binding is mainly regulated by this reinforcement process . Similar to the catch bond , we see that adhesion reinforcement is less important in determining gap size or lifetime ( Fig 4E and 4F ) than in regulating gap opening frequency . We have shown that both the magnitude of forces and the cytoskeletal compartment that generates the forces ( stress fibers or cortex ) affect gap opening frequency , size and and lifetime . Besides these broad compartments , many other biological and physical parameters affect how forces ultimately act on cell-cell junctions: Forces may act in a directed manner due to larger parallel actin bundles and synchronous myosin activation , e . g . initiated through waves of activators [15] , or may act more randomly [33] . We test variations in force applications through parameters that affect the transition time when forces change ( t T r a n s i t i o n F o r c e ) , through spatial force distributions and through the velocity at which forces are modified . In Fig 4G we observe how increasing the force transition time t T r a n s i t i o n F o r c e slowly reduces the gap opening . This is due to the fact that a slower , persistent application of forces leads to a redistribution of the forces through rearrangement and remodeling of the cell . It is consistent with experimental works that showed that force fluctuations influence gap opening dynamics [15] . Then , distributing the same radial forces over several adjacent stress fibers reduces gap opening frequency ( Fig 4H ) . More spatially distributed forces are less capable of damaging cell-cell junctions than localized peak forces , since such high peak forces are required to overcome the catch bond maximal lifetime . Likewise , high peak forces lead to longer lifetime and larger size of the resulting gaps ( Supplementary S13C and S13D Fig ) . Next we observe the effect of force persistence in time . We vary the force recalculation time parameter ( equally for all forces ) in Fig 4I . Results show that the time that forces are applied does not have a big influence on gap formation . This suggests that cells are able to adapt to forces in longer time scales and therefore it is not the time that forces are applied what regulates gap formation , but the transitions of force fluctuations and their spatial distribution . To demonstrate that the geometry of the gap opening dynamics is physiologically relevant , we quantified the characteristics of extravasating cancer cells through monolayers of HUVECs , as shown in S9 Movie . Here , a tumor cell is seen transmigrating through an endothelial monolayer at a tricellular junction as delineated by VE-cadherin GFP , followed by gap-closure after the tumor cell has completely cleared the barrier . Fig 5A shows the ratio of tumor cells that extravasated at vertices , relative to borders . We see that tumor cells preferentially extravasate at the vertices , in line with the previously observed increased frequency of gaps opening there ( Fig 1G ) and similar observations of extravasating neutrophils [32] . Moreover , even if cancer cells initially arrest at the border between two endothelial cells , they are much more likely to extravasate at a vertex at later points in time , rather than at the border where they initially attached to , perhaps first through migration on the surface of the endothelium and subsequent preferential attachment to points of exposed basement membrane as a result of inherent EC junctional dynamics ( Fig 5B ) . This could suggest that in addition to the possibility of cancer cells actively signaling to open gaps in the endothelium , endothelial barrier dynamics itself can also present the cancer cells with opportunities to begin the transmigration process .
The computational model presented in this paper allowed us to study how gaps in an endothelial monolayer initially open , grow , stabilize and finally close , and we identified which physical properties dominantly regulate each stage . The model simulates a cell monolayer in two dimensions . Adhesion between cells is simulated through binding or unbinding of adhesion complexes located on adjacent cells . These adhesion complexes are dynamically engaging and disengaging as the myosin generated forces cause cell deformations . Because of cell-cell adhesion rupture , gaps between the cells are formed . To perform our simulations , the model is based on a number of assumptions that simplified the model . First of all , due to the typically small height ( about 3μm ) of endothelial cells [40] , we neglected the third dimension perpendicular to the monolayer . However , disruptions of the spatio-temporal dynamics of adhesion molecules and cytoskeletal organization in the third dimensions are likely to impact gap formation . Incorporating such effects into our model would consequently require a 3D model of a cell with more detailed descriptions of the subcellular mechanics . However , the purpose of our model was to demonstrate the broad impact of subcellular mechanical structures on gap formation . For this reason , we modeled the cells in two dimensions and included only radial stress fibers and contractile actin fibers parallel to the membrane . This was motivated by experiments that indicated different roles of these actin structures on gap formation [23] . Each discrete adhesion complex is simulated as one cluster to simulate recruitment of proteins such as vinculin or talin , without increasing the total number of components in the simulation . Myosin generated forces included in the model are assumed to occur only in the direction of the stress fibers or membrane . Supplementary S14 Fig summarizes some of our key conclusions: By comparing our reference case with extreme variations of very low stress fiber or membrane stiffness , we see that both passive mechanical properties and adhesion complex properties are important in controlling gap opening frequency ( Supplementary S14A Fig ) . On the other hand , the lifetime and especially size of gaps increases significantly with lower stress fiber or membrane stiffness , since the softer cells are more likely to deform and adapt in response to the opened gap ( Supplementary S14B and S14C Fig ) . We also verify that stress fiber stiffness influence is stronger than membrane stiffness influence . In contrast , properties of the cell-cell adhesions strongly affect the frequency of the gap openings , but less so their lifetime or size . Indeed , decreasing the density of adhesion bonds or the adhesion stiffness strongly increase the frequency of forming gaps ( Supplementary S14A Fig ) , while only marginally affecting the size or lifetime of the gaps ( Supplementary S14B and S14C Fig ) . These data thus summarizes our biological model where adhesion properties control the initial formation of gaps , while cell mechanical properties are critical in limiting the size and duration of opened gaps . Our results that gaps open more frequently at vertices than at borders were true over wide ranges of parameters ( Fig 2B , 2E and 2H ) . Only extremely small bending stiffness led to similar frequencies of gaps at vertices and at borders ( Fig 2B ) . These results also show that earlier experimental observations , where neutrophils were found to extravasate preferentially at endothelial cell vertices , [32] , can be explained through the mechanical dynamics of the endothelial monolayer alone . Consequently , this may be a general mechanism for extravasating cells , and we found a similar behavior with extravasating cancer cells ( Fig 5 ) . This finding is complementary to the extensive literature that suggests that chemical or mechanical signaling of extravasating immune or cancer cells to the endothelium facilitates extravasation [1 , 8 , 31] . There are many potential hypotheses why both the autonomous dynamics of the endothelial monolayer and the bidirectional signaling with the extravasating cells may play a role during extravasation: It may be that initial autonomously forming gaps are important for extravasating cells to sense a gap and they consequently signal to widen the gap or to keep it open . The gap sizes that the model predicts are of the order of magnitude of a few microns , which is enough for extravasating cells to protrude through the gap . Our previous study indicates that tumor cells can squeeze significantly when transmigrating through artificial gaps [16] , so the autonomous gaps may be of sufficient size for complete transmigration . Nevertheless , endothelial gaps may widen during transmigration , so crosstalk between the transmigrating cell and the endothelium likely remains an important factor contributing to the likelihood and speed of extravasation . Then , whether bidirectional signaling or autonomous gap formation dominates the process may be cell type specific . For instance , it is still a major research question why certain cancer cells preferentially metastasize to certain organs [41] . We may speculate that not only the signaling of the specific primary tumor cells with an organ-specific type of endothelial cells influences the likelihood of extravasation [8] . Also , the mechanical properties of the endothelium of the target organs will likely play a major role . Our flexible modeling framework was tested with a HUVECs monolayer , yet , by changing the physical parameters of the model , it may be quickly adapted to other endothelia . Besides testing our model with different endothelial cells , some other important steps towards validating our model conclusions in vivo will be to test our model with more realistic three dimensional microvasculature with blood flow , embedded in extracellular matrix and surrounded by supporting cells such as pericytes , fibroblasts or , for brain , astrocytes [41 , 1] . Such real , in vivo microvasculature consists of vessels that are curved and exposed to shear stresses due to the flow . That , in turn , may be affected by extravasating cells that may obstruct blood flow . Similarly , matrix stiffness was shown to affect endothelial monolayer integrity [42 , 43] . Some complications in validating our results in vivo involve the lack of available in vitro cultures that are required to provide high throughput , microscopy resolution and level of experimental control that is lacking in vivo , making direct comparison of computational models to in vivo experiments unfeasible . However , the recent rapid progress in developing more complex and organ specific in vitro assays of 3D microvasculature will make such validations feasible in the near future [44 , 45 , 46] . Our model is based on a number of simplifications . We do not consider the effect of extracellular matrix and substrate stiffness properties on monolayer integrity , despite the known effect of these properties on cell mechanics . It is important to remark that cells on glass may behave very differently than in vivo endothelial vessels . Our model also does not include the effect of fluid pressure or tangential stress due to fluid flow . Pressure and blood flow would induce additional forces over the monolayer that could affect gap generation processes . For example , it was observed [14] that tangential flow could induce the strengthening of cell-cell junctions , therefore reducing paracellular extravasation . Modeling such complex environments presents a great challenge to both in vitro and in silico models . It is therefore essential to justify assumptions that can reduce this complexity and make the model development feasible . Here , we have assumed that the mechanics of the inside of a cell is determined by a fixed number of stress fibers , although it is known that inside the cell there are different polymer structures such as microtubules and intermediate filaments . Moreover , actin filaments are not fixed in time but appear and disappear depending on their stability and polymerization rates . To simulate all of this with high accuracy would require a completely different model in which the computational cost that would exceed current capabilities . For the purpose of this project , we focused on incorporating essential cell mechanical structures that have been implicated in the regulation of gap formation , and modeled a fixed number of stress fiber similar to other works [28 , 29] . Similarly , we have simulated adhesion complexes as discrete elements that can bind two membrane points of neighboring cells . In real cells , adhesion complexes between cells are formed by a great variety of proteins such as VE-cadherins , α-catenin , talin or vinculin . While the spatio-temporal dynamics of each of these adhesion molecules likely influences gap formation , no computational model can currently explain their precise organization in adhesion complexes and their resulting effect on gap formation . Consequently , our model included an effective term that describes the force dependent recruitment of adhesions , as observed in different experimental studies [23 , 20] . Moreover , there are also challenges to the mathematical modeling of complex 3D microvasculature . Modeling of epithelial sheets in 3D has proved challenging , with some recent interesting progress after decades of mainly focusing on epithelial monolayers in 2D [47 , 48 , 49] . These models are based on frameworks such as vertex models , where the dynamics of each cell is incorporated into the dynamics of vertices between cells . There are many other modeling frameworks that can capture different aspects of the complex cell behavior , such as cell based models [50] , immersed boundary models [51] or subcellular element models [52 , 53] . These modeling frameworks are , however , not directly suitable to predict the formation of gaps at either vertices or borders . Given these challenges , is was paramount to establish a 2D mathematical model of an endothelial monolayer that was validated with novel experiments and that was able to lead to insights into the mechanisms of endothelial gap formation .
Human umbilical chord vein cells ( HUVECs ) were transduced with VE-cadherin-GFP using methods described previously [45] . HUVECs at P7-10 were seeded onto 35 mm glass bottom Mattek dishes ( at 3 × 105 cells/dish ) , which had been plasma treated for 30 seconds previously . Cells were allowed to grow to confluence ( beyond 100% ) in EGM-2MV ( Lonza ) for 3 days before imaging . Dishes were imaged on an Olympus FV1000 confocal microscope with magnifications of 30X ( oil immersion ) , under an environmental chamber set at 37C and 5% CO2 . The chamber was equilibrated for ∼ 30 min prior to the start of image acquisition . For time-lapse videos of junctional dynamics , z-stacks of 40μm ( 4μm steps ) were taken at intervals of 3 minutes . Time-lapse images were appended and analyzed manually on ImageJ . A single unique junctional disruption is defined as a vertex or border with an observed gap of greater or equal than 3μm , and are preceded and proceeded at some point in time with a closure ( no visible gap in fluorescence greater than 0 . 6μm ) . The number of junctional disruption events was counted for each border and vertex of an image over a total time period of 2 hours . Vertices and borders belonging to the same cells were still considered to be unique . Tumor cells were suspended in EGM-2MV ( Lonza ) and a concentration of 15 , 000 cells/mL , and 1mL of the suspension was gently added to each HUVEC monolayer . Cells were allowed to settle first for ∼10 minutes before acquisition of t = 0 images . For quantification of extravasation , z-stacks were taken at 3μm steps at an endpoint of 6 hours to image the entirety of the tumor cell and endothelial monolayer . Any tumor cell that has breached the endothelial layer as evidenced by protrusion extension across and beneath the endothelial layer was considered as “extravasated” . Delineation of the endothelial barrier is visualized via CD31 staining ( Biolegend , Cat # 303103 ) for 30 min in EGM-2MV at 37C and 5% CO2 prior to imaging . | Transmigration of immune cells into and out of the blood vessels is a crucial process for the functioning of the immune system during infections and acute inflammations , and aberrant transmigration may contribute to chronic inflammations . Likewise , cancer metastasis critically depends on intra-and extravasation of cancer cells through the endothelium . While much research investigated the role of immune or cancer cells in signaling to the endothelium , facilitating effective transmigration , and some work uncovered a role of passive mechanical properties such as stiffness during transmigration , little is known about the active role the endothelium itself plays during such processes . Our computational model , together with new data , highlights the dynamic nature of endothelial cells , leading to gap formations through mechanical processes within the endothelium , without influence of cancer or immune cells . Thus , our results highlight the need to take the active mechanics of the endothelium into account when devising strategies to overcome the adverse effects of endothelial gap formation during inflammation or cancer . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [] | 2019 | Balance of mechanical forces drives endothelial gap formation and may facilitate cancer and immune-cell extravasation |
anzctr . org . au ACTRN12612000814875; anzctr . org . au ACTRN12613000565741; anzctr . org . au ACTRN12613001040752; ClinicalTrials . gov NCT02281344; anzctr . org . au ACTRN12612001096842; anzctr . org . au ACTRN12613001008718
Malaria vaccine research efforts have been directed predominantly at P . falciparum , since globally it is the major cause of malaria-related mortality [1] . However , it is now recognized that P . vivax is poised to become the dominant species in areas where it is endemic [2] and can be associated with severe pathology [2 , 3] . Yet , compared to what is known about responses to P . falciparum , little is known about immune responses to P . vivax infection . This lack in knowledge is due in part to confounders that are present in samples from naturally-infected individuals living in malaria-endemic regions where parasitic co-infections and cross-species immunity are present; and technical difficulties associated with experimental infection of humans due to a lack of a method for the continuous in vitro culture of P . vivax [4] . It has been generally assumed that P . vivax would elicit similar immune responses compared to P . falciparum . However , the two parasites display very different features in terms of life cycle , invasion mechanism and immunopathology [2 , 3 , 5] and thus may generate distinct host specific immune responses . A few studies have compared global frequencies of circulating lymphocyte populations during P . falciparum or P . vivax infection in naturally infected humans [6 , 7] , but have not investigated their activated or effector phenotype . The recent establishment of different models of Controlled Human Malaria Infection ( CHMI ) provides the opportunity to obtain samples from malaria-naive healthy volunteers following first exposure to Plasmodium blood-stage parasites , thereby greatly enhancing our understanding of the host-parasite immune response [8 , 9] . Until recently , such experimental infection studies could be done only with P . falciparum due to the lack of a continuous in vitro culture system of P . vivax as a source of parasitized red blood cells [8] . Recently , however , a cell bank of cryopreserved P . vivax infected erythrocytes was successfully derived from a naturally-infected individual and used to experimentally infect malaria-naive healthy adult volunteers , establishing for the first time a CHMI model with P . vivax [10] . Here , we have taken advantage of this novel resource to compare cellular immune responses generated following experimental blood-stage infection of naive volunteers with P . vivax or P . falciparum . Overall , we found marked differences in the immune profiles generated following infection with the two species . Specifically , P . vivax but not P . falciparum infection led to the expansion of a specific subset of CD8+ T cells which were associated with an activated phenotype and cytotoxic potential . This study enhances our understanding of P . vivax associated immunity and Plasmodium species-specific immunity , identifying for the first time components of the immune response to blood-stage infection that are species-specific .
Experimental infection of malaria-naive healthy adult volunteers was undertaken at QPharm Pty Ltd ( Brisbane , Australia ) ; all clinical studies were registered on the Australian and New Zealand Clinical Trials Registry ( ANZCTR ) : P . falciparum clinical trial ID numbers ACTRN12612000814875 , ACTRN12613000565741 , ACTRN12613001040752 and NCT02281344; and P . vivax clinical trial numbers ACTRN12612001096842 and ACTRN12613001008718 , with written informed consent and approval of the QIMR Berghofer Medical Research Institute Human Research Ethics Committee ( QIMRB-HREC ) and the Western Institutional Review Board ( ethics board for the trial sponsor , Program for Appropriate Technology in Health , PATH ) . Inoculum preparation , volunteer recruitment , infection , monitoring and treatment were performed as described previously for P . falciparum [11] or P . vivax [10] . In brief , healthy malaria-naive individuals were intravenously inoculated with freshly thawed P . falciparum 3D7 or P . vivax parasitized erythrocytes and treated with anti-malarial drugs when the parasitemia exceeded the approximate threshold of 10 , 000 parasites/mL , at day 7–8 post-infection or day 14 post-infection for P . falciparum or P . vivax , respectively . The infecting dose for P . falciparum was 1 , 800 viable parasitized red blood cells . Parasite growth modeling using in silico analysis estimated that the infecting dose for P . vivax was 15 fold lower compared to P . falciparum ( Khoury D & McCarthy JS , in preparation ) . Blood samples were collected prior to infection , at day 7 post-infection for P . falciparum infected volunteers , and day 14 post-infection for P . vivax infected volunteers . Peripheral blood collected in Lithium Heparin Vacutainers ( Becton Dickinson ) was either used directly for flow cytometry analysis , or peripheral blood mononuclear cells ( PBMC ) isolated using standard Ficoll density gradient centrifugation . Parasitemia was determined using a consensus P . falciparum or P . vivax species-specific quantitative PCR assay as previously described [12] . Parasite levels were assessed once daily until day four post-infection and then twice daily until treatment . All samples were batch tested in triplicate together after each study completion . Limit of detection was 64 parasites/ml [12] . Exponential growth equation fitting parasitemia kinetics for P . falciparum or P . vivax infected volunteers was calculated with GraphPad Prism ( version 6 . 0 ) . Staining buffer was PBS supplemented with 0 . 5% FCS and 4 mM EDTA . Whole blood collected in Lithium Heparin vacutainers was lysed and fixed with BD FACS lysing solution ( Becton Dickinson ) and lymphocytes permeabilized with BD FACS permeabilising solution 2 ( Becton Dickinson ) according to the manufacturer’s instructions . Cells were then resuspended in 50 μl of staining buffer containing anti-human CD4-BV510 ( Becton Dickinson , 1:200 dilution ) , anti-human CD8-APC-H7 ( Becton Dickinson , 1:400 dilution ) , anti-human CD19-PE-Cy7 ( Biolegend , 1:200 dilution ) , anti-human CD38-APC ( Biolegend , 1:400 dilution ) , anti-human Perforin-PE ( Biolegend , 1:400 dilution ) , anti-human Granzyme B-Pacific Blue ( Biolegend , 1:400 dilution ) and 1 μl of human Fc receptor blocking solution ( Human TruStain FcX , Biolegend ) for 30 minutes at room temperature , washed and resuspended in staining buffer before acquisition on LSR Fortessa 4 ( Becton Dickinson ) with Diva software . FlowJo software version 6 . 0 was used for gating . CD38+ CD8+ T cells , CD38- CD8+ T cells as well as CD8- cells were sorted from freshly isolated PBMC . Approximately 10x106 cells were resuspended in 50 μl of staining buffer containing anti-human CD4-BV510 ( Biolegend , 1:200 dilution ) , anti-human CD8-APC-H7 ( Becton Dickinson , 1:400 dilution ) and anti-human CD38-PerCpCy5 . 5 ( Biolegend , 1:400 dilution ) for 20 minutes at 4°C , washed and resuspended in staining buffer . Just before the sorting , 1 μg/mL of propidium iodide ( Sigma-Aldrich ) was added to allow for assessment of viability . Pi-CD8+CD4-CD38+ , Pi-CD8+CD4-CD38- , and Pi-CD8- cells were sorted using a BD Aria III cell sorter ( Becton Dickinson ) directly in staining buffer and kept on ice until further use for in vitro assays . Sorted CD38+ and CD38- CD8+ T cells were plated at 50 , 000 cells/well in RPMI 1640 containing 25 mM Hepes , 2 mM L-glutamine ( Invitrogen ) , and supplemented with 10 units/mL of Penicillin ( Life Technologies ) , 10 μg/mL of Streptomycin ( Life Technologies ) and 10% fetal bovine serum ( Life Technologies ) in a 96-well plate pre-coated overnight with 10 μg/mL of anti-human CD3 OKT3 antibody ( Biolegend ) together with 0 . 75x106 cells/mL autologous CD8- cells , anti-human CD107a-FITC ( Biolegend , 1:200 dilution ) and 1 μg/mL of co-stimulatory antibodies anti-human CD28 and anti-human CD49d ( Becton Dickinson ) for 5 hours at 37°C in an atmosphere of 5% C02 . Following stimulation , cells were resuspended in 20 μl of staining buffer containing anti-human CD4-BV510 ( Biolegend , 1:200 dilution ) , anti-human CD8-APC-H7 ( Becton Dickinson , 1:400 dilution ) for 20 minutes at 4°C , washed and resuspended in staining buffer before acquisition on LSR Fortessa 4 ( Becton Dickinson ) with Diva software . FlowJo version 6 . 0 was used for gating .
P . falciparum and P . vivax experimental infection of malaria-naive volunteers was performed under similar procedures [10 , 11] . However , due to logistical reasons associated with parasite density in the inoculum stock , and a lack of a continuous in vitro culture system for P . vivax [4] , the infecting dose for P . vivax was estimated to be 15-fold lower than that used in the P . falciparum studies ( Khoury D & McCarthy JS , in preparation ) . Demographics of P . falciparum and P . vivax infected volunteers were comparable in term of age , gender , BMI and ethnicity ( S1 Table ) . Parasitemia growth curves determined by quantitative PCR ( qPCR ) from P . falciparum and P . vivax experimental infection studies were similar for both parasites ( Fig 1 and Table 1 ) except for the delayed onset of detectable blood-stage parasitemia with P . vivax . Specifically , P . falciparum parasites were detected as early as day 4 of infection whereas P . vivax parasites were detected at day 8 of infection , consistent with the differences in the size of the starting inocula . Interestingly , all individuals infected with P . vivax developed symptoms of malaria before the time of treatment while more than 40% of P . falciparum infected individuals were asymptomatic until anti-malarial drug administration ( S2 Table ) . In order to compare cellular immune responses to infection between P . falciparum and P . vivax infected volunteers , we determined the phenotype of lymphocytes ex vivo from whole blood samples obtained prior to infection and during infection once the parasitemia exceeded 10 , 000 parasites/mL ( which corresponded to day 7 and day 14 for P . falciparum and P . vivax infected volunteers , respectively ) . CD38 is a surface glycoprotein that modulates cell adhesion , signal transduction and intracellular Ca2+ levels , and is specifically upregulated on lymphocytes following activation [13] . We have recently shown that the frequency of CD38+ T cells and B cells circulating in the peripheral blood of test volunteers was dynamically regulated during experimental blood-stage infection with P . falciparum and that the expansion of CD38+ CD4+ T cells following infection was inversely correlated with parasite burden [14] . Thus , we compared the frequencies of CD38+ T cells and B cells circulating before and after infection in P . falciparum or P . vivax infected volunteers . There was a higher frequency of CD38+ CD4+ T cells circulating following infection in P . falciparum but not P . vivax infected volunteers ( Fig 2A ) . Conversely , P . vivax infection but not P . falciparum elicited a higher frequency of circulating CD38+ CD8+ T cells ( Fig 2B ) . No significant differences were observed in the frequency of CD38+ B cells circulating following infection with P . falciparum or P . vivax ( Fig 2C ) . There was no correlation between the expansion of CD38+ CD8+ T cells following infection and parasite burden ( S1 Fig ) . Overall , these data suggest that the quality of the immune response generated following primary blood-stage infection in humans is Plasmodium species-dependent . Since very little information on P . vivax protective immune responses is available , we aimed to further understand the contribution of CD38+ CD8+ T cells to P . vivax blood-stage immunity . Effector CD8+ T cells perform classical cytotoxic functions by killing infected cells through perforin-mediated dependent mechanisms . To determine the cytotoxic potential of CD38+ CD8+ T cells generated during P . vivax infection , we measured their intracellular expression of granzyme B and perforin by flow cytometry , in an independent cohort ( n = 2 because of logistics associated with vivax experimental infection ) . We found similar expression of granzyme B and perforin in CD38+ and CD38- CD8+ T cells before infection ( Fig 3A and 3B ) . However , post-infection , CD38+ CD8+ T cells had a greater expression of granzyme B and perforin compared to CD38- CD8+ T cells ( Fig 3A and 3B ) . Additionally , CD38+ CD8+ T cells circulating during infection contained a higher amount of granzyme B and perforin in P . vivax infected volunteers compared to P . falciparum infected volunteers ( Fig 3C and 3D ) whereas no differences were observed prior to infection . In order to further investigate the cytotoxic potential of CD38+ CD8+ T cells and CD38- CD8+ T cells circulating during P . vivax infection , we tested their ability to degranulate in vitro following TCR stimulation . Prior to infection , CD38+ and CD38- CD8+ T cells had the same ability to degranulate , whereas post-infection CD38+ CD8+ T cells had a higher degranulation compared to CD38- CD8+ T cells ( Fig 4 ) . Overall , these findings suggest that CD38+ CD8+ T cells have a greater cytotoxic potential compared to CD38- CD8+ T cells and that this cytotoxic function is specifically activated in the CD38+ CD8+ T cells circulating during P . vivax infection but not P . falciparum infection .
Herein we report on the first study to compare the quality of cellular immune responses elicited in humans by experimental blood-stage infection with P . vivax or P . falciparum . In this study we utilized a model of controlled infection of malaria-naive human volunteers , thereby avoiding potential confounders of pre-existing immunity and cross-species immunity . We found marked differences between responses to the two Plasmodium species in terms of the phenotype of T cells that expanded during infection . P . falciparum infection elicited a significant expansion of CD38+ CD4+ T cells whereas P . vivax infection led to the expansion of CD38+ CD8+ T cells . We have recently shown that the frequency of circulating CD38+ CD4+ T cells was significantly increased following experimental infection with P . falciparum and inversely correlated with parasite levels [14] . Here we show that P . vivax blood-stage infection elicited a substantially different type of immune response compared to P . falciparum , with significant changes in the CD8+ T cell compartment rather than in the CD4+ T cell compartment . There was no significant association between the expansion of CD38+ CD8+ T cells and parasite burden in P . vivax infected volunteers , suggesting that the expansion of CD38+ CD4+ T cells in P . falciparum infection and the expansion of CD38+ CD8+ T cells in P . vivax infection might have distinct contributions to the immune response to blood-stage infection . A possible explanation for the qualitative differences observed between P . falciparum and P . vivax associated immune responses may relate to their distinct tropism during blood-stage infection . P . vivax merozoites preferentially invade reticulocytes [15] which , although they are enucleated , still express MHC-I molecules which remain from the nucleated reticuloblast stage . This is in contrast to mature erythrocytes that have completely lost expression of MHC molecules . Previous work using a genetically engineered mouse model of murine malaria has shown that CD8+ T cells can be activated by parasitized erythroblasts but not parasitized mature red blood cells through MHC-I dependent mechanisms [16] . Thus , we hypothesize that the higher proportion of infected reticulocytes in P . vivax infection leads to the activation of a higher proportion of CD38+ CD8+ T cells in an MHC-I dependent manner in comparison to P . falciparum infection . While it is established that CD4+ T cells and parasite-specific antibodies are critical for protective immune responses to blood-stage malaria [17] , the contribution of CD8+ T cells is less clear . Studies in mice have shown that CD8+ T cells were activated and associated with protective function in lethal [18] or chronic [19] blood-stage malaria . However , no association between CD8+ T cells and protective immunity to primary blood-stage infection in humans has been reported yet . Our data suggest that the CD38+ CD8+ T cells that specifically expand during P . vivax infection display increased cytotoxic function compared to other CD8+ T cells . Hence their function might be to kill parasitized reticulocytes through MHC-I dependent and perforin-mediated mechanisms . This proposal is further supported by the enhanced expression of cytotoxic molecules observed in CD38+ CD8+ T cells circulating during P . vivax compared to P . falciparum infection . In contrast , we could not identify an expansion of cytotoxic CD8+ T cell population following P . falciparum infection of malaria-naïve volunteers . These findings have important implications in regard to P . vivax vaccine development . Indeed , most efforts so far have been directed toward the direct translation of the findings associated with P . falciparum vaccine development to P . vivax [20] ( e . g . the use of ortholog antigens that were shown to be protective against P . falciparum ) . Here we show that P . falciparum and P . vivax elicit qualitatively different immune responses and are likely to require distinct vaccine-induced immune responses for protection . Thus , the immunization strategy may need to be adapted for each Plasmodium species to mount optimal protective immune responses , and the design of a universal vaccine conferring protection against multiple Plasmodium species might be conceptually more challenging than initially thought . In conclusion , in this study we report that primary exposure of humans to different Plasmodium species elicited qualitatively distinct immune responses: P . falciparum infection generated changes in CD4+ T cells whereas P . vivax preferentially activated a subset of CD8+ T cells expressing the activation marker CD38 and associated with a cytotoxic function . The specific expansion of CD38+ CD8+ T cells following P . vivax infection might be due to the preference for P . vivax merozoites to infect reticulocytes which can activate CD8+ T cells through MHC-I dependent mechanisms . Overall our data are consistent with the proposal that protective immune responses to Plasmodium are species-dependent . These findings have important implications for malaria vaccine development strategies . | The specific immune responses that contribute to protective immunity in humans following Plasmodium infection are yet to be fully characterized . The species P . vivax and P . falciparum account for most human infections , yet little is known about P . vivax specific immune responses and whether they are similar to or distinct from P . falciparum . Here , we establish that P . vivax and P . falciparum elicit distinct cellular immune responses following primary infection , with the expansion of a subset of CD38+ CD8+ T cells with a cytotoxic potential in P . vivax but not in P . falciparum infection . This study provides the first evidence for the activation of CD8+ T cells in P . vivax blood-stage infection and demonstrates the existence of species-dependent host immune responses to malaria . These findings have important implications for P . vivax vaccine development , and suggest that future malaria vaccine studies should be adapted according to the target Plasmodium spp . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"blood",
"cells",
"parasite",
"groups",
"medicine",
"and",
"health",
"sciences",
"immune",
"cells",
"plasmodium",
"immunology",
"plasmodium",
"falciparum",
"parasitic",
"diseases",
"parasitic",
"protozoans",
"parasitology",
"apicomplexa",
"protozoans",
"red",
"blood",
"... | 2016 | Plasmodium vivax but Not Plasmodium falciparum Blood-Stage Infection in Humans Is Associated with the Expansion of a CD8+ T Cell Population with Cytotoxic Potential |
Identification of regulatory elements within the genome is crucial for understanding the mechanisms that govern cell type–specific gene expression . We generated genome-wide maps of open chromatin sites in 3T3-L1 adipocytes ( on day 0 and day 8 of differentiation ) and NIH-3T3 fibroblasts using formaldehyde-assisted isolation of regulatory elements coupled with high-throughput sequencing ( FAIRE-seq ) . FAIRE peaks at the promoter were associated with active transcription and histone modifications of H3K4me3 and H3K27ac . Non-promoter FAIRE peaks were characterized by H3K4me1+/me3- , the signature of enhancers , and were largely located in distal regions . The non-promoter FAIRE peaks showed dynamic change during differentiation , while the promoter FAIRE peaks were relatively constant . Functionally , the adipocyte- and preadipocyte-specific non-promoter FAIRE peaks were , respectively , associated with genes up-regulated and down-regulated by differentiation . Genes highly up-regulated during differentiation were associated with multiple clustered adipocyte-specific FAIRE peaks . Among the adipocyte-specific FAIRE peaks , 45 . 3% and 11 . 7% overlapped binding sites for , respectively , PPARγ and C/EBPα , the master regulators of adipocyte differentiation . Computational motif analyses of the adipocyte-specific FAIRE peaks revealed enrichment of a binding motif for nuclear family I ( NFI ) transcription factors . Indeed , ChIP assay showed that NFI occupy the adipocyte-specific FAIRE peaks and/or the PPARγ binding sites near PPARγ , C/EBPα , and aP2 genes . Overexpression of NFIA in 3T3-L1 cells resulted in robust induction of these genes and lipid droplet formation without differentiation stimulus . Overexpression of dominant-negative NFIA or siRNA–mediated knockdown of NFIA or NFIB significantly suppressed both induction of genes and lipid accumulation during differentiation , suggesting a physiological function of these factors in the adipogenic program . Together , our study demonstrates the utility of FAIRE-seq in providing a global view of cell type–specific regulatory elements in the genome and in identifying transcriptional regulators of adipocyte differentiation .
Sequencing allowed identification and mapping of the human genome [1] . Transcriptional regulation of genes is essential for manifesting cellular phenotypes and complex biological processes . Coordinated actions of transcription factors and cofactors on regulatory DNA sequences produce transcriptional activation of the eukaryotic gene . Therefore , identification and mapping of the genome's regulatory elements is critical for understanding how cell-type-selective regulation of genes in the genome is achieved . Traditionally , regulatory elements have been identified by DNase I hypersensitivity assay combined with Southern blot analysis [2] . That assay coupled with microarray or high-throughput sequencing ( DNase-Chip or DNase-seq ) were effectively applied in genome-wide identification of open chromatin regions [3] , [4] , [5] , [6] . Lieb and his colleagues recently developed formaldehyde-assisted isolation of regulatory elements ( FAIRE ) as a simple procedure to isolate nucleosome-depleted DNA from chromatin [7] , [8] . FAIRE detects open chromatin structure much the way the DNase I hypersensitivity assay does [8] , [9]—but with advantages , like obviating the need for clean nuclei preparation and laborious enzyme titrations [7] , [8] . Coupled with high-throughput sequencing ( FAIRE-seq ) , FAIRE allows unbiased identification of potential regulatory elements without requiring prior knowledge of ( or about ) binding factors . FAIRE-seq's genome-wide detection of open chromatin genomic regions in human pancreatic islets was successfully used to determine a causal single nucleotide polymorphism in loci associated with type 2 diabetes development in genome-wide association studies [10] . The adipocyte is central in controlling energy balance and whole-body glucose and lipid homeostasis [11] . Advances in adipocyte research have shown that adipose tissue stores excess energy and secretes hormones and metabolites to communicate with other organs , maintaining systemic metabolic homeostasis [12] . Peroxisome proliferator-activated receptor gamma ( PPARγ; NR1C3 ) is both necessary [13] , [14] , [15] and sufficient [16] for adipocyte differentiation . Necessary for both development and maintenance of mature adipocytes , PPARγ is crucial in systemic glucose and lipid homeostasis [13] , [14] , [15] , [17] , and , importantly , is the molecular target of thiazolidinediones , widely prescribed for obese diabetics [18] . C/EBPα-β-δ act with PPARγ , forming the adipogenic transcription cascade [19] . C/EBPβ and δ are induced by adipogenic stimulus , inducing PPARγ , which activates expression of C/EBPα , which binds and further activates expression of PPARγ , providing a positive regulatory loop [11] , [20] . Genome-wide approaches now dissect the transcriptional mechanisms of adipocyte differentiation . ChIP-chip or ChIP-seq studies of adipogenic regulators [21] , [22] , [23] , [24] , [25] , [26] , [27] , [28] , [29] have provided valuable mechanistic insights into adipogenic transcription never before gained by conventional experiments: New concepts include co-localization of PPARγ and cell type–specific transcription factors [27] , low conservation rate of PPARγ binding sites between murine and human adipocytes [28] and the role of C/EBPβ as a pioneer factor that establishes “hot spots” where multiple adipogenic regulators cooperatively work in the very early stage of differentiation [6] . Our study took an unbiased approach to mapping adipocyte-specific regulatory elements in the genome by using FAIRE in 3T3-L1 adipocytes ( on day 0 and day 8 of differentiation ) and NIH-3T3 fibroblasts . We show that the FAIRE peaks contain regulatory elements such as promoters , enhancers and insulators , and that adipocyte-specific non-promoter FAIRE peaks are functionally linked to genes regulated during differentiation—about half these peaks being overlapped by PPARγ . We show that highly regulated genes in adipocyte differentiation are associated with clusters of multiple adipocyte-specific non-promoter FAIRE peaks . Furthermore , because FAIRE does not require a prioi knowledge of bound transcription factors , we could employ computational motif analyses of DNA sequences from the adipocyte-specific FAIRE peaks in an unbiased manner and identify a motif for nuclear family I ( NFI ) transcription factors in addition to motifs for PPAR and C/EBPs . We show the functional role of NFIA and NFIB in adipocyte differentiation . We demonstrate the utility of FAIRE-seq both in providing a global view of cell type–specific cis-regulatory elements in the genome and identifying transcriptional regulators of adipocyte differentiation .
Regulatory elements in the genome are characterized by open chromatin structures accessible to regulatory factors [30] . To explore genome-wide changes in open chromatin conformation during adipocyte differentiation , we used FAIRE—a method of isolating genomic regions depleted of nucleosomes [7]—combined with high-throughput sequencing ( FAIRE-seq ) to identify open chromatin sites in the adipogenic cell line 3T3-L1 before ( day 0 ) and after ( day 8 ) differentiation and in NIH-3T3 fibroblasts , which cannot differentiate into adipocytes . This approach identified in the genome 37 , 781 FAIRE peaks in 3T3-L1 on day 0 and 26 , 611 on day 8 , plus 36 , 111 in NIH-3T3 cells—all , with a false discovery rate of <10−4 . By using ChIP-seq analyses , we also generated genome-wide maps of binding sites for PPARγ , the master regulator of adipocyte differentiation , for RXRα , its heterodimer partner , for histone H3 lysine 4 trimethylation ( H3K4me3 ) , and for CCCTC-binding factor ( CTCF ) [31] . Figure 1 shows a representative map of results generated near Klf15 and Pparg , both transcription factors up-regulated by differentiation , and both important in adipocyte differentiation [16] , [32] . Consistent with previous observations [10] , 28% of the FAIRE peaks were detected near the transcription start sites ( TSSs ±500 bp ) of RefSeq genes [33] and are referred to as promoter FAIRE peaks ( Figure S1A ) , while 72% were located outside known TSSs , and are referred to as non-promoter FAIRE peaks . Notably , only 8% of the non-promoter FAIRE peaks were located in a −5 kb proximal promoter region while the majority of non-promoter FAIRE peaks were located in introns and distal regions ( Figure S1A ) . Average profiling revealed that a FAIRE signal , H3K4me3 and histone H3 lysine 27 acetylation ( H3K27ac ) were observed at TSSs of actively transcribed genes ( Figure S1B and S1D ) . On the other hand , non-promoter FAIRE peaks were accompanied by monomodal enrichment of H3K4me1 and were devoid of H3K4me3 enrichment , a condition described as the signature of enhancers [34] , [35] ( Figure S1D ) . CTCF binding sites are important in insulator function and high-order chromatin structure [31] . The CTCF binding sites in our study ( day 0 or day 8 ) were largely overlapped by those in a study by Mikkelsen ( day 0 or day 7 ) [28] ( 86 . 3% and 88 . 5% , respectively ) . CTCF binding accounted for about one fifth of either the promoter or non-promoter FAIRE peaks ( Figure 1 and Figure S1C ) . Collectively , these data suggest that the open chromatin sites identified by FAIRE-seq show characteristics of regulatory elements such as promoter , enhancer and insulator . We next compared the FAIRE peaks in 3T3-L1 cells on day 0 and day 8 and in NIH-3T3 cells . The promoter FAIRE peaks were relatively constant among the three groups . Over 70% of those peaks on day 0 and day 8 3T3-L1 cells and in NIH-3T3 cells were shared by all three groups ( Figure 2A ) . In contrast , non-promoter FAIRE peaks showed dynamic change . The three groups shared only 25% , 45% , and 26% of non-promoter FAIRE peaks in , respectively , day 0 and day 8 3T3-L1 cells and NIH-3T3 cells . This contrasts with an invariable biding pattern of CTCF in the non-promoter regions; in 3T3-L1 cells , 89 . 5% of the non-promoter CTCF binding sites on day 0 overlapped those on day 8 . What's more , a significant proportion of the non-promoter FAIRE peaks were cell type–specific ( Figure 2A ) , implying the role of non-promoter regulatory elements in cell type–specific transcriptional regulation . We divided the non-promoter FAIRE peaks in day 0 and day 8 3T3-L1 cells into tertiles by FAIRE signal intensity , and defined adipocyte- or preadipocyte-specific FAIRE peaks as indicated by red or green boxes in the 4-by-4 table in Figure 2B . By this definition , we judged each non-promoter FAIRE peak as adipocyte-specific , preadipocyte-specific or invariant ( Figure 2B ) . Figure 1 , Figures S2 and S3 show examples of adipocyte-specific non-promoter FAIRE peaks ( indicated by asterisks ) in loci near Klf15 , Pparg , Cebpa [16] , [20] , Mgll [36] , Srebf1 and cidec [37]—all of which are abundantly expressed in adipose tissue and induced during adipocyte differentiation ( data not shown ) . Remarkably , multiple adipocyte-specific FAIRE peaks existed in the vicinity of these genes and included introns and downstream regions ( Figure 1 , Figures S2 and S3 ) . To determine whether non-promoter FAIRE peaks were functionally associated with cell type–specific gene expression , we analyzed the relationship between the presence of the adipocyte- or preadipocyte-specific non-promoter FAIRE peaks and the change in gene expression during adipocyte differentiation . Those FAIRE peaks were enriched in the vicinity of genes , expression levels of which were highly induced or suppressed during adipocyte differentiation ( Figure 2C ) . Importantly , as the number of the adipocyte-specific FAIRE peaks associated with a gene increased , the fraction of up- or down-regulated genes increased or decreased , respectively ( Figure 2D , red lines ) , while as the number of associated preadipocyte-specific FAIRE peaks increased , the fraction of up- or down-regulated genes decreased or increased , respectively ( Figure 2D , green lines ) . Conversely , the more robust the induction of the expression level of a gene during adipocyte differentiation , the greater the numbers of adipocyte-specific FAIRE peaks associated with the gene ( Figure 2E , red line ) . In contrast , the more robust the reduction of the expression levels of a gene during adipocyte differentiation , the greater the numbers of associated preadipocyte-specific FAIRE peaks that were associated ( Figure 2E , green line ) . Invariant FAIRE peaks were associated specifically with neither up- nor down-regulated genes ( Figure 2E , blue line ) . We next employed a gene ontology ( GO ) analysis tool ( DAVID ) [38] to determine what kind of biological processes were associated with genes bound by the adipocyte-specific FAIRE peaks . We found that biological processes ( e . g . , adipocyte differentiation ) were significantly enriched compared with the genomic background ( Figure 2F ) . It was of interest that embryonic placenta development—for which PPARγ is critical [13] , [14] , [15]—was enriched ( Figure 2F ) . Together , these data highlight the role of the cell type–specific non-promoter open chromatin sites detected by FAIRE-seq in differentiation-dependent transcriptional regulation . PPARγ is key regulator of adipocyte development [16] , [20] . To elucidate the contribution of PPARγ to adipocyte-specific transcriptional regulation , we conducted ChIP-seq analyses using antibodies specific for either PPARγ or RXRα [24] in 3T3-L1 adipocytes at 36 hours and day 8 after induction of differentiation . The number of PPARγ binding sites increased during differentiation while that of RXRα binding sites remained virtually constant ( Figure 2G ) . Significant overlap between the PPARγ and RXRα binding sites was consistent with the heterodimer formation of PPARγ and RXRα [21] , [39] ( Figure 2G ) . Microarray and GO analysis revealed that the PPARγ binding sites were enriched in the vicinity of genes up-regulated by adipocyte differentiation ( Figure S4B ) and the bound genes were associated with adipocyte differentiation and lipid metabolism ( Figure S4C ) . Using MEME [40] , we performed de novo motif analysis of genomic regions bound by PPARγ , and found that the AGGTCA-n-AGGTCA ( called DR-1 ) shown was the most over-represented one ( E-value 1 . 3×10−055 ) ( Figure S4A ) . An extension AGT 5′ outside of DR-1 appeared to correspond to the direct interaction between the DNA and the hinge region between the DNA-binding domain and the ligand-binding domain [41] . As shown in genomic loci ( Figure 1 , Figures S2 and S3 ) , a significant proportion of adipocyte-specific non-promoter FAIRE peaks overlapped the PPARγ/RXRα binding sites . To determine the contribution of PPARγ to the adipocyte-specific open chromatin regions , we calculated percent fractions of the FAIRE peaks that were stratified by FAIRE signal in 3T3-L1 on day 0 and day 8 ( Figure 2B ) —and that overlapped either the PPARγ binding sites ( Figure 2G ) or C/EBPα binding sites in 3T3-L1 reported by Schmidt et al . [42] . Both PPARγ and C/EBPα binding sites were enriched in the fractions of adipocyte-specific FAIRE peaks ( Figure 2H and 2I ) , and they respectively accounted for 45 . 3% and 11 . 7% of the adipocyte-specific FAIRE peaks ( averages of the red bars in Figure 2H and 2I ) . These data support the role of PPARγ and C/EBPα as primary transcription factors that drive adipocyte-specific gene expression—although they may not explain all of it . Genes that were highly induced by adipocyte differentiation often harbored multiple adipocyte-specific FAIRE peaks and/or PPARγ binding sites in their vicinity , as suggested by the linear correlation between the number of the associated adipocyte-specific FAIRE peaks and the robustness of the induction of the gene by adipocyte differentiation ( Figure 2D and 2E ) . ( See Figure 1 , Figures S2 and S3 for representative genes . ) To determine whether these multiple regions have functional regulatory elements , we selected AdipoR2 [43] , [44] . Although AdipoR2 was regulated by PPARγ and PPARα ( [45] and data not shown ) , conventional −2 kb promoter studies failed to identify the response element [46] . Our ChIP-seq analysis revealed a cluster of multiple PPARγ/RXRα binding sites in the intron 1 , downstream of the TSS of AdipoR2 ( Figure S2B , arrow heads ) . We identified putative DR-1 motifs in these biding sites ( Figure 3A ) and tested them by gel-shift assay and luciferase assay . These motifs were indeed bound by the PPARγ/RXRα heterodimer , and were functional in the luciferase assay ( Figure 3B and 3C ) , suggesting the functionality of these elements and the advantage of a genome-wide approach over the conventional “promoter-bashing” approach to identifying such response elements . Recent genome-wide studies revealed clustering of open chromatin regions detected by Dnase I hypersensitivity assay or by FAIRE in the genomes of CD4+ T cells [47] , pancreatic islet cells [10] , [48] and binding sites for certain transcription factors [49] ) —certainly the PPARγ binding sites and adipocyte-specific FAIRE peaks in our analyses tended to form clusters as indicated by an additional peak in distribution histograms of the distance to the nearest peak among the PPARγ binding sites or the adipocyte-specific FAIRE peaks ( Figure 4A ) . We calculated the total number of PPARγ binding site clusters for different window sizes and compared them with a random data set comprised of the same number of sites ( Figure 4B ) . The PPARγ binding sites had a significantly higher number of clusters in a window size raging from 800 bp to ∼30 kb . Similar results were obtained for the adipocyte-specific FAIRE peaks ( Figure 4A and 4B ) . On the other hand , multiple genes involved in adipocyte function [55] , [56] , [57] were often co-regulated in certain genomic regions that harbor clusters of adipocyte-specific regulatory elements ( see Figure S2C , Figure 4C , and Figure S5 ) . We therefore statistically tested—method in reference [50]—to see if neighboring genes tended to be co-regulated during adipocyte differentiation , and found that neighbors of highly induced genes ( >10 fold ) were indeed more likely to be up-regulated over three fold ( 18% , or 112 of 618 neighbors ) than the 2 , 012 of 21 , 343 total genes ( 9% ) that were up-regulated over three fold ( P = 1 . 26×10−12 , one-sided Fisher test ) . Neighbors of randomly selected genes were not significantly up-regulated ( p = −0 . 67 , average of 1 , 000 trials , Figure 4D ) . Together , these data suggest that the transcriptional regulation of genes during adipocyte differentiation involves multiple adipocyte-specific regulatory elements—which tend to form clusters—and that co-regulation of neighboring genes often occurs during adipocyte differentiation . Next , we performed enrichment analyses of known motifs using AME in the MEME suite and the TRANSFAC [51] and JASPER [52] motif databases to identify motifs enriched in either adipocyte- or preadipocyte-specific FAIRE peaks compared with the background ( statistical values shown as corrected p-value in Figure 5 ) . We also determined the enrichment ratio ( Ad/pAd ) by calculating the ratio of occurrence of a motif in the adipocyte-specific FAIRE peaks and in the preadipocyte-specific FAIRE peaks as described in reference [28] . Using both parameters , we obtained motifs that had been significantly enriched in either kind of FAIRE peak and that occurred in significantly different number . Figure 5 shows the top of the list of TRANSFAC motifs enriched in the adipocyte- and preadipocyte-specific FAIRE peaks . The motifs for PPARγ ( and other DR1 motifs ) and C/EBPs were among the list , consistent with their critical roles in adipogenic transcription . Motif analyses using the JASPER motif database showed enrichment of the motifs for PPARγ , C/EBPs and the motif for Zfp423 , a recently identified adipogenic regulator [53] ( Figure S6 ) . Motif analyses of the preadipocyte-specific FAIRE peaks showed significant enrichment of a motif for AP-1 , a downstream transcription factor complex of the growth factor/MAP kinase signaling pathways , which include epidermal growth factor and c-Jun N-terminal kinases , known inhibitors of adipogenesis [54] , [55] ( Figure 5 and Figure S6 ) . We also performed de novo motif analysis ( MEME ) [40] of the adipocyte-specific FAIRE peaks , and observed significant enrichment of motifs that corresponded to those for PPARγ and C/EBPs ( Figure S7 ) . Together , these instances of enrichment of known regulators indicate the validity of this approach . There were several other motifs for transcription factors , their functions not previously linked to adipocyte differentiation ( Figure 5 , Figures S6 and S7 ) . We focused on a motif for the NFI family transcription factors . The murine NFI family consists of NFIA , NFIB , NFIC and NFIX , and was identified as a site-specific DNA-binding protein that bound to the adenovirus origin of replication [56] . It forms a dimer to bind to the symmetric consensus sequence TTGGC ( N5 ) GCCAA [57] . We first examined the expression change of these factors in in vitro adipocyte differentiation and found that the expression of NFIA and NFIB were significantly induced during differentiation of 3T3-L1 and of another adipogenic cell line , 3T3-F442A ( Figure 6A and 6C ) . Consistent with this pattern , both NFIA and NFIB were highly expressed in a variety of adipose tissue depots in addition to the brain ( Figure 6B ) . We next examined the effect of siRNA knockdown of NFIA and NFIB on adipogenic gene regulation and adipocyte differentiation ( Figure 6C ) . Interestingly , induction of the expression of the adipogenic transcription factors PPARγ and C/EBPα and of downstream genes was significantly suppressed by siRNA knockdown of either NFIA or NFIB ( Figure 6C ) . Consistent with the gene expression change , we observed significant reduction of lipid accumulation as judged by oil red O staining , suggesting physiological roles for NFIA and NFIB in adipocyte differentiation ( Figure 6D ) . We confirmed the effect of NFIA and NFIB knockdown on adipogenesis by using independent pooled siRNA ( Figure S8 ) . We next asked whether overexpression of these factors influence adipocyte differentiation . We amplified NFIA and NFIB coding sequences from cDNA prepared from adipocytes , and cloned them into retroviral pMXs-puro vectors . We also made a dominant negative NFIA that lacks the C-terminal transactivation/repression domain ( NFIA-DN ) [58] . Overexpression of NFIA—but not NFIA-DN or NFIB—resulted in robust induction of PPARγ , C/EBPα and aP2 ( Figure 7A ) at a basal state . Surprisingly , the induction of these factors was robust enough to make the cells to form lipid droplets visible and stainable by oil red O even before initiation of differentiation by the DMI ( dexamethasone , IBMX and insulin ) treatment ( Figure 7B and 7C ) . However , after the DMI treatment , NFIA-expressing cells were overtaken by control cells , and on day 7 , NFIA and NFIB overexpressing cells showed attenuated differentiation ( Figure 7D and 7E ) . We speculate that this was caused by secondary effects of overly strong overexpression levels ( >30 fold , Figure 7A ) . Almost complete suppression of adipogenesis by NFIA-DN overexpression was consistent with the results of knockdown experiments ( Figure 6 , Figure 7D and 7E ) . Nevertheless , the robust induction of PPARγ , C/EBPα and aP2 by NFIA overexpression at the basal state implies direct action of NFIA on transcriptional control of these factors . To dissect the mechanism by which NFIs regulate PPARγ , C/EBPα and aP2 , we examined DNA sequences of the adipocyte-specific FAIRE peaks and/or the PPARγ binding sites in the vicinity of these factors and found that some of them have NFI binding motifs as listed in Figure 8A . ChIP analysis using an anti-NFI antibody confirmed actual binding of NFI to these sites ( Figure 8B and 8C ) . We extended this experiment by counting NFI motifs in the FAIRE peaks on a genome-wide scale . Interestingly , percent fractions of genes harboring NFI binding motifs in the FAIRE peaks were higher when the genes were bound by PPARγ and induced during differentiation ( Figure 8D ) , indicating a significant degree of specificity for the NFI's action on the adipogenic transcriptional program . Collectively , we demonstrated that the combination of FAIRE-seq and computational motif analyses is useful in identifying novel regulators of adipocyte differentiation . The 3T3-L1 adipogenic cell line was established by isolating clonal sublines of mouse fibroblast line 3T3 [59] . Lastly , we compared FAIRE peaks between ‘undifferentiated’ 3T3-L1 and NIH-3T3 cells . As shown in Figure 2A , a substantial proportion of FAIRE peaks was unique to either 3T3-L1 or NIH-3T3 cells . We defined non-promoter FAIRE peaks as specific to 3T3-L1and NIH-3T3—as we did for the adipocyte- or preadipocyte-specific FAIRE peaks in Figure 2B . The 3T3-L1- or NIH-3T3-specific FAIRE peaks were enriched in the vicinity of genes whose expression levels were higher in 3T3-L1 or NIH-3T3 , respectively ( Figure S9A ) . Motif analysis of the 3T3-L1-specific FAIRE peaks showed that the binding motif for EBF ( Figure S9B ) had the highest enrichment ratio ( 1 . 81 ) and a statistically significant p-value of 3 . 9E-3 . Although the p-value of the motif for PPARγ/RXR did not reach statistical significance , that motif had an enrichment ratio of 1 . 84 . These two factors were among the handful that were proven to transform NIH-3T3 cells into adipocytes when ectopically introduced [16] , [60] .
We demonstrated that genome-wide mapping of open chromatin regions by FAIRE-seq is a simple , accurate method that allows a snapshot view of regulatory elements in the genome . Although open chromatin regions detected by FAIRE-seq include promoters of transcribed genes , enhancers and insulators , open chromatin regions that vary in two different conditions likely contain regulatory elements that play roles in the specific biological process . By comparing open chromatin regions in preadipocytes and adipocytes , we identified the adipocyte- and preadipocyte-specific FAIRE peaks in the genome . Functionally , we demonstrated that the adipocyte-specific FAIRE peaks were associated with genes up- regulated by adipogenesis while the preadipocyte-specific FAIRE peaks were associated with genes down-regulated by adipogenesis ( Figure 2C , 2D and 2E ) . Adipocyte gene expression appears mediated through multiple regulatory elements distal to transcription start sites ( TSSs ) : greater induction of gene expression by differentiation means greater likelihood that more adipocyte-specific FAIRE peaks are associated with the gene ( Figure 2D and 2E ) . This implies that optimal gene transcriptional regulation may require coordinated actions of multiple regulatory elements . Therefore , although valuable and informative , the proximal promoter assay may not always be sufficient ( e . g . , AdipoR2 , see Figure S2B and Figure 3 ) . Nevertheless , the importance of proximal promoter regions is obvious given the fact that many proximal promoter regions are successfully used to generate tissue-specific transgenic lines . Recently , Mikkelsen et al . demonstrated in adipocytes that many cis-regulatory elements are often not conserved between human and murine adipocytes even though the expression pattern of genes is conserved [28] . They observed that such motifs were located within linage-specific transposon insertions . Existence of multiple regulatory elements around biologically important genes could be a mechanism by which cells maintain key gene regulations against genomic changes during evolution . Clustering of regulatory elements could also result from an accumulative effect of such evolutional genomic changes . Computational motif analysis is used to discover new transcription-factor binding motifs in sequences inferred from genome-wide studies such as ChIP-seq [61] . In genome-wide ChIP analysis of transcription factors , motif analysis is used to obtain their accurate binding motifs and discover unknown DNA binding factors that co-localize with the transcription factors of interest , for example , see [27] , [62] , [63] . The analyses , however , relied on prior knowledge about transcription factors and the regions to be analyzed are limited to their biding sites . In contrast , the combination of motif analyses and mapping of regulatory elements by FAIRE-seq does not require such prior knowledge , hence offers a distinct advantage in unbiased screening for novel transcription factors important in given biological processes . In our study , we retrieved the motifs for PPARγ and C/EBPs and for known regulators that top the list of the motifs identified in the adipocyte- or preadipocyte-specific FAIRE peaks ( Figure 5 , Figures S6 and S7 ) . Furthermore , we demonstrated that NFIA and NFIB were functionally required for proper adipocyte differentiation ( Figure 6 ) . These results demonstrated that motif analyses of cell type–specific FAIRE peaks are useful in identifying regulators of a biological process in an unbiased manner . To our knowledge , few studies have employed motif analysis and our unbiased approaches in investigating enhancer-like DNA regions . Mikkelsen et al . recently employed ChIP-seq for H3K27ac to define enhancer regions specific for adipocyte differentiation . Both studies similarly detected the motifs for PPAR , C/EBPs and AP-1 in the most enriched motifs . There are , however , differences . Mikkelsen discovered PLZF and SRF as novel negative regulators [28] and we found NFIA and NFIB as regulators of adipocyte differentiation—perhaps due to differences in methods . First , we directly compared FAIRE peaks and H3K27ac peaks detected in the Mikkelsen study and found considerable , but not complete , overlap especially in the non-promoter regions: 94% of 10 , 461 promoter FAIRE peaks and 45% of 27 , 320 non-promoter FAIRE peaks overlapped H3K27ac in 3T3-L1 on day 0 . There may be different classes of enhancer elements that prefer either H3K27ac or open chromatin . Also , we used two parameters to sort motifs: the statistical significance of enrichment ( p-value ) in either kind of cell type–specific FAIRE peaks; and the motif enrichment ratio between the adipocyte- and preadipocyte-specific FAIRE peaks ( see [28] ) . The combination guarantees significant enrichment of the peaks' motifs and the difference in their number depending on whether they are adipocyte- or preadiocyle-specific . The motifs for PLZF and SRF were not on the top of our list since the p-values were not significant—probably due to relatively lower occurrence , although we also found a significant enrichment ratio of 0 . 37 and 0 . 50 , respectively . We calculated p-values and the enrichment ratios of the top motifs in the Mikkelsen's study by using our adipocyte- and preadipocyte-specific FAIRE peaks and found general similarity ( Figure S10 ) . Overall , both studies notably demonstrate the utility of the combining computational motif analysis and unbiased mapping of regulatory elements in identifying new regulators of adipocyte differentiation . Siersbæk et al . recently employed DNase-seq to investigate genome-wide change in open chromatin structure at various time points during 3T3-L1 differentiation [6] . They reported dramatic increase in the number of open chromatin sites in the first hours of differentiation . Such regions included what they termed “hot spots” that were bound by multiple adipogenic regulators , facilitating binding of PPARγ and C/EBPα during the late stage of differentiation . We found that the DNaseI hypersensitive sites in 3T3-L1 cells on day 0 or day 6 in the Siersbaek study [6] significantly overlapped the FAIRE peaks on day 0 or day 8 in our study ( 78 . 8% and 80 . 9% , respectively ) ( Figure S11 ) , suggesting that both methods detect similar open chromatin regions . Although limited amount of motif analyses of the DNase I sites was conducted in their study , we think a combination of motif analysis and DNase-seq should work in a similar way . The NFI family was identified as site-specific DNA-binding protein that bound to the adenovirus origin of replication [56] , [57] . Although defects in development of organs such as brain , lung , tooth , bone and skeletal muscle in Nfia , Nifb , Nifc and Nfix-deficient mice were documented [64] , [65] , [66] , [67] , [68] , [69] , no publication has reported direct evidence that NFI family transcription factors are involved in adipogenesis , but it is a reasonable supposition since bone , muscle and adipocytes have a common mesenchymal precursor [70] . Interestingly , Graves et al . demonstrated that NFI was bound to the adipogenic −5 . 4 kb enhancer region in the aP2 promoter [71] , which is the original adipogenic enhancer region where the PPARγ/RXR heterodimer was found to bind and act [72] . The NFI binding motif they examined by gel shift assay [72] was close to the best-characterized PPARγ binding sites in the region , and was also in site 9 ( Figure 8A , right panel , site 9 ) , which was indeed bound by NFI in ChIP assay ( Figure 8C ) . Forced overexpression of NFIA in 3T3-L1 cells dramatically induced expression of PPARγ , C/EBPα and aP2 and caused lipid droplet formation before initiation of differentiation . Our ChIP data suggest that activation of these genes by NFIA is through direct binding of NFI to regulatory elements near these genes . In overexpression experiments , NFIB did not activate the adipogenic genes ( Figure 7 ) . NFI factors are known to undergo extensive alternative splicing [57] . We speculate that this could be due to truncation of the C-terminus caused by lack of exons 10 and 11 in the NFIB cDNA that we cloned ( NM_001113209 . 1 ) while the NFIA clone completely matched NM_010905 . 3 . NFI was also implicated in functions of other nuclear receptors such as the androgen receptor ( AR ) , estrogen receptor ( ER ) and glucocorticoid receptor [4] , [73] , [74] . Further studies are necessary to elucidate the mode of action of NFIs and positioning of NFIs in the adipogenic regulatory network .
3T3-L1 , NIH-3T3 , 3T3-F442A and HEK293T cells were maintained in DMEM , supplemented with 10% FBS . For adipocyte differentiation , two days after confluence , 3T3-L1 cells were treated with dexamethasone ( 1 µM ) , IBMX ( 0 . 5 mM ) , and insulin ( 5 µg/ml ) ( DMI ) for 48 hours , followed by treatment with insulin alone , with medium replacement every two days thereafter . For differentiation of 3T3-F442A , cells were treated with insulin ( 5 µg/ml ) after confluence , with medium replacement every two days . All animal works have been conducted according to the institutional guidelines . Generation of characterization of antibodies for human PPARγ and human RXRα was described previously [24] . Rabbit polyclonal anti-histone H3 trimethyl K4 ( ab8580 ) was from Abcam . Antibodies against CTCF were from Upstate ( #07–729 ) . Anti-NFI antibody ( H-300 ) was from Santa Cruz ( sc-5567 ) . FAIRE experiments were performed based on a protocol published by Giresi et al . [7] . Briefly , cells were fixed with 1% formaldehyde for five minutes at room temperature , the fixation stopped by adding 2 . 5 M glycine ( final 125 mM ) . Fixed cells were scraped and collected in 15 ml tubes ( 4×10∧6 cells/tube ) and washed twice with cold PBS , then 8×106 cells were re-suspended in 800 µl of MC lysis buffer ( 10 mM Tris-HCl pH 7 . 5 , 10 mM NaCl , 3 mM MgCl2 , 0 . 5% NP-40 ) and incubated on ice for ten minutes . After spinning for four minutes at 8000 rpm , the pellet was re-suspended in 400 µl SDS lysis buffer ( 1% SDS , 10 mM EDTA , 50 mM Tris-HCl pH 8 . 0 , proteinase inhibitor cocktail ) and incubated on ice for ten minutes . Glass beads ( size , 200 mg ) ( Polysciences Inc . #05483-250 ) were added and the DNA was sheared by sonicator . Next , we added 200 µl cold ChIP dilution buffer ( 0 . 01% SDS , 1 . 1% Triton X-100 , 1 . 2 mM EDTA , 16 . 7 mM Tris-HCl pH 8 . 0 , 167 mM NaCl ) , and after spinning for one minute at 8 , 000 rpm , supernatant was transferred to a new 1 . 5 ml tube . Aliquote was taken , de-crosslinked , purified by phenol/chloroform extraction , and run on a gel to ensure average fragment sizes of 300 bp . Remaining samples were processed three times by phenol/chloroform extraction to recover DNA not bound by nucleosome in the water phase . The samples were de-crosslinked by overnight incubation at 65°C and purified by ethanol precipitation . They were subsequently treated with RNase A ( final 50 ug/ml ) , purified by QIAquick PCR purification kit ( Qiagen ) and used for subsequent analyses . ChIP was performed as descried previously [24] , [75] . For ChIP using anti-PPARγ , RXRα and CTCF antibodies , 3T3-L1 cells were cross-linked with 1% formaldehyde for ten minutes at room temperature and were prepared for ChIP as described previously . For ChIP using anti-H3K4me3 antibody , the nuclei of 3T3-L1 cells were prepared by centrifugation through a sucrose gradient and were digested with MNase ( TaKaRa ) . After centrifugation , the supernatant was used for ChIP . Sequences of primers used for qPCR were listed in Table S1 . High-throughput sequencing was performed by using the Genome Analyzer System ( GA II ) ( Illumina ) as described elsewhere [76] . In short , we repaired ends of DNA samples , created 3′-dA overhang , ligated Illumina adaptors , size-fractioned the samples by gel extraction and amplified them with 8 cycles of PCR according to the manufacturer's instructions . We then purified the DNA and performed cluster generation and 36 cycles of sequencing on an Illumina cluster station and 1G analyzer following the manufacturer's instructions . Sequences were mapped to the reference murine genome , NCBI build 37 ( mm9 ) . Peak detection was performed using Findpeaks 3 . 1 . 9 . 2 [77] with a false discovery rate ( FDR ) cut-off of 1×10−4 . Operations such as intersections , unions , and subtractions of genome regions were performed with a web-based GALAXY genome analysis tool [78] , [79] . Average profiling of FAIRE and histone modifications near transcription start sites or FAIRE peaks were generated using “sitepro” in the CEAS package [80] . For definition , we first ranked peaks based on signal intensity , which were detected in 3T3-L1 on either day 0 or day 8 with a FDR of 10−4 . We then classified each peak into tertiles ( high , mid , low ) for either day the peak that had the higher percentile ( see also the 4-by-4 table in Figure 3B ) . Gene ontology annotation analysis was performed using DAVID ( ver . 6 . 7 ) [38] . The top 2 , 000 genes were used , sorted by the number and maximum height of the adipocyte-specific FAIRE peaks within a region ±25 kb from TSS . For genes bound by PPARγ , we used the top 931 genes with more than three PPARγ binding sites within a region ±25 kb from TSS . To detect enrichment of specific—rather than general—terms , following the instructions of DAVID's developer , we used GOTERM_BP_4 and GOTERM_BP_5 , and sorted result lists by using both fold enrichment and Benjaini p-value [38] , [81] . Statistical clustering analyses of the PPARγ binding sites and the adipocyte-specific FAIRE peaks were performed as described in references [47] , [48] . Enrichment analyses of known motifs were performed with AME ver . 4 . 6 . 0 in the MEME suite [82] . After removing repeat regions with RepeatMasker [83] , DNA sequences from the center 150 bp regions of the top 2 , 000 cell type–specific FAIRE peaks were analyzed with a fixing partition of 2 , 000 , dinucleotide randomization and p-value threshold of 10−4 and p-value report threshold of 0 . 05 . We used the licensed version of TRANSFAC database ( Release 2010 . 4 ) [51] and the JASPAR CORE database [52] . Motif enrichment ratios ( adipocyte-/preadipocytes-specific FAIRE ) for motifs in the TRANSFAC or JASPAR CORE database were determined by a method described in reference [28] . Instances of motifs were enumerated in the adipocyte- or preadipocytes-specific FAIRE peaks by using FIMO ver . 4 . 6 . 0 in the MEME suite , with a p-value threshold of 10−4 , normalized by total nucleotide length . Motif enrichment ratios were determined by dividing the normalized adipocyte enrichment values by preadipocyte values . MEME ver . 4 . 3 . 0 [40] was used to identify de novo motifs over-represented in the adipocyte- or preadipocyte-specific FAIRE peaks and the PPARγ binding sites . After removing repeat regions with RepeatMasker [83] , DNA sequences from the center 150 bp regions of the top 800 cell type–specific FAIRE peaks with higher signals were used for the analyses . Identified enriched de novo motifs were next analyzed by TOMTOM in the MEME suite for comparison against a database of known motifs . The Gel shift assay and luciferase reporter assay were performed as previously described [84] , [85] . For the luciferase assay , putative PPRE motifs were cloned in tandem ( 3× or 6× ) into pGL3 basic reporter plasmid ( Promega ) together with the tk minimal promoter . The −5 . 4 kb aP2 promoter luciferase construct is described in reference [84] . The 3T3-L1 cells were transfected with either control siRNA or siRNA for murine NFIA and NFIB ( Santa Cruz Biotechnology , sc-37007 , sc-36045 and sc-43566 , Sigma MISSION siRNA , SASI_Mm02_00309629 , 00309630 , 00307243 , 00307244 ) by using Lipofectamine RNAiMAX ( Invitrogen ) just before they reached confluence . Induction of differentiation ( the DMI treatment ) was started two days after confluence , as described in a method for differentiation of 3T3-L1 cells . The 3T3-L1 adipocytes were washed with PBS , fixed with formalin for 30 minutes at room temperature , rinsed with 60% isopropanol and stained with oil red O solution—freshly made by mixing 0 . 5% oil red O in isopropyl alcohol and water ( 3∶2 ) —and left to sit for one hour; the cells were then washed with water and dried . Total RNA was isolated using TRIzol reagent ( Invitrogen ) , then 0 . 5 µg of the total RNA was reverse transcribed using high-capacity cDNA reverse transcription kits ( Applied Biosystems #4375222 ) and random hexamers . Real-time quantitative PCR ( SYBR green ) analysis was performed on a 7900HT Fast Real-Time PCR System ( Applied Biosystems ) . Primer sequences are listed in Table S1 . Expression was normalized to 36B4 . Transcriptome analysis of 3T3-L1 during differentiation by using a GeneChip Mouse Genome 430 2 . 0 array ( Affimetrix ) was described previously [24] . Heat maps were generated by using GENOMICA , developed by Yaniv Lubling and Eran Segal at the Weizmann Institute of Science . Microarray data of 3T3-L1 and NIH-3T3 cells used in Figure S11 was obtained from GEO ( accession number GSE10246 ) . We amplified NFIA and NFIB coding sequences from cDNA prepared from adipocytes using primers listed in Table S1 , and cloned them into retroviral pMXs-puro vectors . We also made a dominant negative NFIA that lacks the C-terminal transactivation/repression domain ( NFIA-DN ) [58] . Plat E cells were transfected with pMXs-puro plasmids using Lipofectamine 2000 ( Invitrogen ) . Culture medium containing viruses after two day incubation was centrifuged at 2 , 000 rpm for 5 min and supernatant was collected and supplemented with 10 µg/ml polybrene . Conditioned medium with viruses was used to infect 3T3-L1 cells and then selection was started by adding 2 µg/ml puromycin and incubated for 2 days . FAIRE-seq and ChIP-seq raw data are deposited into the DNA data bank of Japan ( DDBJ accession number: DRA000378 ) . | Humans consist of a few hundred types of specialized-function cells . Spatial and temporal transcriptional regulation of genes is essential for manifestation of cellular phenotypes . Identification of regulatory regions in the genome is central to understanding the mechanism of cell type–specific gene regulation . Recently developed high-throughput sequencing technology and computational analyses allow genome-wide investigation of the genome's chromatin structure . Using the FAIRE-seq technique , we identified the genome's open chromatin regions , which harbor regulatory elements in adipocytes . Open chromatin regions distal to genes' transcription start sites significantly differ among cell types . Multiple cell type–specific open chromatin regions exist near genes regulated during adipocyte differentiation . Computational motif analysis of adipocyte-specific open chromatin regions revealed enrichment of a binding motif for the NFI transcription factor family . These factors bind to the regulatory elements near adipogenic PPARγ , C/EBPα , and aP2 genes and regulate their expression . Overexpression of NFIA in 3T3-L1 cells resulted in robust induction of these genes and lipid droplet formation without differentiation stimulus and knockdown of NFIA or NFIB significantly suppressed both induction of genes and lipid accumulation during differentiation . Our study demonstrates the utility of FAIRE-seq in providing a global view of regulatory elements and in identifying transcriptional regulators of cellular functions . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"medicine",
"carbohydrate",
"metabolism",
"diabetic",
"endocrinology",
"cell",
"differentiation",
"hormones",
"gene",
"function",
"dna",
"transcription",
"histone",
"modification",
"developmental",
"biology",
"nutrition",
"obesity",
"epigenetics",
"chromatin",
"lipid",
"met... | 2011 | Global Mapping of Cell Type–Specific Open Chromatin by FAIRE-seq Reveals the Regulatory Role of the NFI Family in Adipocyte Differentiation |
One quadrillion synapses are laid in the first two years of postnatal construction of the human brain , which are then pruned until age 10 to 500 trillion synapses composing the final network . Genetic epilepsies are the most common neurological diseases with onset during pruning , affecting 0 . 5% of 2–10-year-old children , and these epilepsies are often characterized by spontaneous remission . We previously described a remitting epilepsy in the Lagotto romagnolo canine breed . Here , we identify the gene defect and affected neurochemical pathway . We reconstructed a large Lagotto pedigree of around 34 affected animals . Using genome-wide association in 11 discordant sib-pairs from this pedigree , we mapped the disease locus to a 1 . 7 Mb region of homozygosity in chromosome 3 where we identified a protein-truncating mutation in the Lgi2 gene , a homologue of the human epilepsy gene LGI1 . We show that LGI2 , like LGI1 , is neuronally secreted and acts on metalloproteinase-lacking members of the ADAM family of neuronal receptors , which function in synapse remodeling , and that LGI2 truncation , like LGI1 truncations , prevents secretion and ADAM interaction . The resulting epilepsy onsets at around seven weeks ( equivalent to human two years ) , and remits by four months ( human eight years ) , versus onset after age eight in the majority of human patients with LGI1 mutations . Finally , we show that Lgi2 is expressed highly in the immediate post-natal period until halfway through pruning , unlike Lgi1 , which is expressed in the latter part of pruning and beyond . LGI2 acts at least in part through the same ADAM receptors as LGI1 , but earlier , ensuring electrical stability ( absence of epilepsy ) during pruning years , preceding this same function performed by LGI1 in later years . LGI2 should be considered a candidate gene for common remitting childhood epilepsies , and LGI2-to-LGI1 transition for mechanisms of childhood epilepsy remission .
Postnatal mammalian brain development proceeds in three phases the first of which is construction of the primary neural network ( ages zero to two years in humans , zero to one week in mice , and estimated zero to one to two months in dogs ) . In humans , this phase generates a network of approximately one quadrillion synapses . The second phase , pruning ( ages two to 10 years in humans , seven to 17 days in mice , and estimated two to four months in dogs ) , is chiefly characterized by massive removal of unneeded or otherwise inappropriate synapses , almost half the original synapses . The third and final phase is the remainder of life , during which synapse numbers remain stable [1]–[3] . Epilepsies are by far the most common neurological diseases in children two to 10 years of age , the three most common of which are Rolandic Epilepsy , Panayiotopoulos syndrome , and Childhood Absence Epilepsy ( CAE ) . The first two of these three syndromes are focal-onset epilepsies where seizures start from defined brain regions , while CAE is a generalized epilepsy where seizures appear to start simultaneously from all brain regions . All three syndromes share a remarkable feature of remission after age 10 , i . e . after network pruning is complete [4] . All three are genetically complex syndromes , and paucity of gene information has impeded their understanding , including how and why they remit . To date , a few ion channel mutations ( e . g . in GABRG2 , CACNA1H ) have been found in CAE , accounting for far less than 1% of patients with this syndrome [5] . While the above three syndromes begin and end during the pruning phase of neurodevelopment in the vast majority of cases , other genetic epilepsies begin near or after the end of this phase , i . e . after age eight in most cases . These include the generalized Juvenile Myoclonic Epilepsy ( JME ) ( to date with mutations in the EFHC1 or GABRA1 genes; penetrance ∼50% ) [6] , [7] and the focal-onset Autosomal Dominant Lateral Temporal Lobe Epilepsy ( ADLTE ) ( also called Autosomal Dominant Partial Epilepsy with Auditory Features ) with mutations in the LGI1 gene [8] ( penetrance 67% ) [9] . JME is generally a non-remitting lifelong epilepsy [10] . Remission rate in ADLTE has not been determined , although the literature indicates that most cases remain on seizure medications , unlike Rolandic epilepsy , e . g . , where the vast majority do not [10] , [11] . In the present work , we show that mutation of the Lgi2 gene , a gene closely related to Lgi1 , causes remitting focal-onset epilepsy in dogs between ages one and four months , which is equivalent to human two to eight years . LGI2 belongs to a family of four closely related neuronal proteins including the well-studied LGI1 . We report functional and expression studies of LGI2 , which , combined with previous LGI1 studies , suggest a novel concept of the basis of remission common in childhood epilepsy .
The Lagotto Romagnolo is an ancient curly-haired water dog ( water dogs , or water spaniels , originally served to retrieve game falling in water ) , which was selected in Italy to become an excellent truffle hunter . The popularity of the breed fluctuated with the truffle industry and in the early 1970s underwent a strong genetic bottleneck to near extinction , when a group of dog lovers decided to save it . The breed has since gained popularity for reasons unrelated to truffle or water hunting , and its numbers are in the thousands spread across most developed countries ( http://www . lagottoromagnolo . org/ ) . The breed is affected by an epilepsy , Benign Familial Juvenile Epilepsy ( BFJE ) , described in detail in reference [12] . Onset is at five to nine weeks of age , and the epilepsy invariably completely remits by four months of age . Remission is so reliable that the epilepsy is considered by many breeders as an unfortunate particularity of the breed and often disregarded . The seizures consist of whole-body tremors sometimes associated with alteration of consciousness . Electroencephalography ( EEG ) reveals unilateral epileptic discharges in central-parietal and occipital lobes , and magnetic resonance imaging ( MRI ) is normal . During the months with epilepsy the animals are often ataxic , but this resolves completely as the seizures disappear [12] . Towards the goal of mapping and identifying the BFJE gene we first reconstructed a large multinational Lagotto pedigree from which an example with 212 Finnish dogs including 34 cases is shown in Figure S1 . The dogs live in homes as private pets , often in different countries in Europe , or were still with their breeders . Disease segregation suggested autosomal recessive inheritance ( Figure S1 ) . Next , we performed a single nucleotide polymorphism ( SNP ) genome-wide association study with DNA from 11 of the affected dogs and 11 unaffected littermates ( discordant sib-pairs ) ( Figure S1 ) and found very strong association in a region of chromosome 3 ( CFA3 ) , peaking at the marker at base-pair 89159216 ( Praw 0 . 000035; Pgenomewide 0 . 08 ) ( Figure 1A and 1B ) . There was no significant association at any other genomic locus , the next best association being over 100-fold less significant ( Figure 1A ) . Genotype analysis around the 89159216 SNP revealed a 1 . 7 Mb block of homozygous SNPs between markers at 87 . 3 Mb and 89 . 0 Mb in the 11 cases and none of the controls ( Figure 1B ) . This region contains nine genes , including Lgi2 . Sequencing Lgi2 revealed an exonic homozygous protein-truncating sequence change , c . 1552A>T ( p . K518X ) , in all 11 affected and none of the 11 unaffected animals ( Figure 1C ) . Genotyping a cohort of 140 dogs for the 89159216 SNP , for the Lgi2 c . 1552 sequence change , and for three additional SNPs from the homozygous region revealed extremely high associations including Praw 4 . 47×10−16 at 89159216 and Praw 1 . 05×10−23 ( the highest association ) at Lgi2 c . 1552 ( Table 1 ) . These results strongly suggested that Lgi2 c . 1552A>T ( p . K518X ) is the BFJE mutation . Next we studied segregation of the sequence change in the pedigree . Of the 28 affected dogs from which we had samples , 26 ( 93% ) were homozygous Lgi2 c . 1552T ( p . 518X ) ( i . e . homozygous for the nonsense codon ) , two were heterozygous ( 7% ) , and none was homozygous for the wild-type ( wt ) A nucleotide . The two affected dogs that were heterozygous were also heterozygous for the 13 SNP haplotype around the Lgi2 locus , and we found no evidence for compound heterozygosity as all other variants in the gene were synonymous ( Table S1 ) . These results suggested that if the Lgi2 c . 1552A>T ( p . K518X ) change is the BFJE mutation , it can , in a minority of cases , cause the epilepsy heterozygously . To explore this further , we screened an independent set of 36 sporadic Lagottos and found three homozygous for c . 1552T , 14 heterozygous ( 39% ) , and 19 wild-type ( wt ) . All three dogs homozygous for c . 1552T had the syndrome , as did one of the carriers ( 7% ) , appearing to confirm the 7% rate of disease through heterozygosity , assuming that Lgi2 c . 1552A>T ( p . K518X ) is causative . Among 112 unaffecteds of the genotyped 140 dogs , 69 were homozygous for the wt A nucleotide , 41 were heterozygous , and two , 1 . 8% , were homozygous for c . 1552T ( OR = 532 , 95%CI: 95 . 0-5747 . 1 and p = 1 . 05×10−23 ) . The latter two may be mis-specified as unaffected - clinical information on many of the dogs in the pedigree was obtained through retrospective questionnaires , and it is possible that a breeder missed seizures , as the epilepsy in some cases is mild and short-lived [2] . Alternatively , these two cases may represent incomplete penetrance , assuming , again , that the sequence change we identified is causative . Similarly , other recent recessive gene discoveries indicate incomplete penetrance including canine lens luxation [13] , degenerative myoelopathy [14] and a form of neuronal ceroid lipofuscinosis [15] . At this point , there were two possibilities . Either Lgi2 c . 1552A>T ( p . K518X ) is the BFJE mutation with an incomplete penetrance in a minority of cases , or it is not the causative variant . To gather more data we proceeded with functional studies of the consequences of the truncating sequence change on the LGI2 protein . We first determined whether the c . 1552A>T sequence change prevents Lgi2 mRNA expression , e . g . through mRNA instability . RT-PCR experiments showed no mRNA reduction ( Figure 2 ) . LGI2 belongs to a family of neuronally secreted proteins ( LGI1 to LGI4 ) conserved across mammals and composed of N-terminal leucine-rich repeats ( LRR ) and C-termini containing seven EPTP repeats [3] . K518X truncates LGI2 within the seventh EPTP repeat ( exon 8 of the gene ) ( Figure S2 ) . Similar mutations truncating LGI1 in the EPTP repeats in humans , including in the seventh repeat , cause ADLTE , the human epilepsy with most commonly onset after age eight and persistence through adulthood . Where studied , the vast majority of LGI1 mutations , truncating or otherwise , prevent secretion of the protein encoded by the mutant allele , and ADLTE is therefore usually a disease due to lack of neuronal secretion of half the required amount of LGI [11] , [16] , [17] . We asked whether the LGI2 K518X truncation prevents LGI2 secretion . We performed western blot experiments with V5-tagged wt and mutant LGI2 transfected in HEK293 cells and found that while both proteins were present in cell lysates only wt LGI2 was found in the culture medium ( Figure 3 ) , indicating that the truncation prevents secretion . Following secretion , LGI1 interacts with a subfamily of the ADAM ( a-disintegrin-and-metalloproteinase ) family of neuronal membrane proteins [18] , [19] . Members of this subfamily , ADAM11 , ADAM22 ( post-synaptic ) , and ADAM23 ( pre-synaptic ) , lack the metalloproteinase domain that other ADAMs use to convey extracellular signals intracellularly [19] . To determine whether LGI2 also binds ADAM22 , ADAM23 and ADAM11 following secretion , we performed immunofluorescent cell surface-binding assays [18] in permeabilized and non-permeabilized cells by co-expressing wt or truncated LGI2 with different ADAMs . Wt LGI2 was secreted and then bound ADAM22 , ADAM23 and ADAM11 expressed on the cell surface ( ADAM11 result not shown ) . Truncated LGI2 was not secreted and did not bind the ADAMs ( Figure 4A–4B ) . We also performed co-immunoprecipitation in rat brain and found that both Adam22 and Adam23 antibodies co-precipitated Lgi2 ( Figure 5 ) . In summary , wt LGI2 binds the same ADAM substrates of LGI1 following secretion , and the Lagotto K518X mutation prevents secretion and ADAM interaction , in the same fashion as the well-characterized truncating LGI1 epilepsy mutations . Summarizing the results to this point , the genome-wide association study revealed a highly significant association in the vicinity of Lgi2 , an extremely strong association ( p = 1 . 05×10−23 ) with the protein-truncating c . 1552A>T ( p . K518X ) sequence change in the gene , and no significant association with any other locus . Lgi2 is a close homologue of the epilepsy ( ADLTE ) gene LGI1 , and the Lgi2 truncating mutation is closely similar to the most common type of epilepsy-causing mutations in LGI1 . The consequence of the truncation on Lgi2 is identical to the consequence of truncation on LGI1 , prevention of neuronal secretion and binding to ADAM receptors , which is presently the most favored mechanism of epileptogenesis in ADLTE . Finally , LGI1 mutations , including truncation mutations , are non-penetrant in 33% of individuals , compared to 1 . 8% non-penetrance in the case of the canine Lgi2 truncation . Considering all the above , we believe the data meet the burden of proof that Lgi2 c . 1552A>T ( p . K518X ) is the BFJE mutation . BFJE is transmitted in imperfect Mendelian fashion . In the vast majority of cases , 93% , homozygous mutation is required for the disease to manifest . In a minority , 7% , heterozygosity suffices . Conversely , 1 . 8% of dogs may be resistant to seizing despite homozygous mutation . Finally , we found no dog with homozygous wt genotype at Lgi2 c . 1552 that has BFJE . Mouse studies show that LGI1 starts being expressed midway through the synapse pruning phase of brain development ( after postnatal day 13 ( p13 ) ) , and gradually increases to reach high and stable adult levels by the end of this phase ( after p17 ) [20] , [21] . Not surprisingly , the mice lacking LGI1 develop seizures after mid-phase pruning [22] and the great majority of human patients with ADLTE have onset of their epilepsy after age 10 years , the end of the pruning phase in humans , with the remaining few having initial seizures in the latter half of this phase [11] , [16] , [17] . However , it is worth noting that the very first seizures in ADLTE are often auditory seizures , which might not initially be appreciated to be seizures and might have occurred earlier than what is currently documented in the literature . Because BFJE occurs only in ages equivalent to human two to eight years , we sought to determine whether expression of LGI2 differs from that of LGI1 . We examined the expression profile of all four LGI genes in adult tissues using the human GeneSapiens expression database [4] and found that the amount of LGI2 in adult brain is much lower than that of the other three ( Figure S3 ) . We next chased Lgi2 expression levels in mouse forebrain and cerebellum by performing quantitative RT-PCR every other day from birth till 27 days . Lgi2 expression in the cerebellum did not change appreciably over this time ( Figure 6A ) . Expression in the forebrain , on the other hand , was highest at birth and through phase one of postnatal development ( neural network construction phase ) , and declined to half the original amount by midway through the pruning phase ( Figure 6B ) . Considering that BFJE occurs only during the pruning phase , these results suggest that LGI2's main functions take place in the developmental phase preceding the phase in which the epilepsy occurs .
Epilepsy is a common symptom of insult to the brain from various causes including tumors , trauma , stroke , and neurodegenerative disease . For example , the most common cause of epilepsy in the elderly is stroke and in neonates hypoxic ischemic encephalopathy . However , epilepsy can also be a disease onto itself , where seizures are the only or preponderant neurological symptom , i . e . the brain is normal except for its propensity to seize . Resolving the basic mechanisms of this type of ‘pure’ epilepsy is expected to provide the clearest insights into epileptogenesis . As mentioned , these pure epilepsy syndromes ( sometimes called idiopathic , Greek for ‘disease onto itself’ ) are the commonest neurological diseases with onset in two to 10 year-old children , and in this age group most are genetic , commonly polygenic , and often characterized by remission in adolescence [5] . Genetic-idiopathic epilepsy syndromes are the most common neurological diseases of dogs , in some breeds 10 times more common than in humans [23] . In the present work we identify the first of the canine idiopathic epilepsy genes , in a remitting syndrome with onset and offset equivalent to human childhood two to 10 years . This epilepsy can now be eliminated from the Lagotto through selective breeding . Carrier frequency of the mutation is very high . We tested 576 Lagottos from three different countries and found a carrier rate of 32% ( Table S2 ) . On the other hand , the mutation appears restricted to this particular epilepsy in this particular breed . We tested 121 epileptic dogs from 40 different breeds , including Barbets , a Lagotto-related French water spaniel breed afflicted with a separate epilepsy , and none carries the BFJE mutation ( Table S3 ) . Genetic epilepsies of various types , as simple or complex traits , are highly enriched in various canine breeds , including Miniature Wirehaired Dacshunds [24] , Finnish Spitzs [25] and Belgian Shepherds [23] due to pure-breeding . Each of these traits is in genetic isolation within its corresponding breed . This vastly improves signal to noise ratio in genetic studies compared to human populations [26] , which should facilitate mapping epilepsy genes . The Lagotto themselves segregate a second epilepsy with onset in adulthood completely distinct from BFJE [12] . We have established that this second epilepsy is not associated with the BFJE mutation ( Table S1 ) , and are presently mapping its gene ( s ) . Five out of the six adult-onset cases with persistent seizures in our pedigree were genotyped and only one of them was homozygous for the BFJE mutation . However , the puppyhood history of this case is unknown and it was impossible to confirm retrospectively whether this case has also had BFJE . This case has an affected littermate with classical BFJE who is homozygous for the mutation . On the other hand , all the other genotyped adult-onset dogs were wildtypes strongly suggesting that this form of epilepsy has its own genetic cause , and that this single homozygous case may have suffered from both BFJE and the adult-onset form of epilepsy . The BFJE gene is a homolog of the human epilepsy gene LGI1 . LGI1 is neuronally secreted and binds three metalloproteinase-lacking ADAM receptors . Significant progress has started to be made in elucidating LGI1's functions at these receptors . LGI1 interaction with post-synaptic ADAM22 strengthens and stabilizes ADAM22-containing synapses [18] , [21] , [22] . Interaction with pre-synaptic ADAM23 enhances neurite outgrowth from ADAM23-containing axons [27] . Through its seven-bladed β-propeller structure ( encoded by the EPTP repeats ) , LGI1 simultaneously binds ADAM23 and ADAM22 , pulling pre and post-synaptic membranes together , physically stabilizing synapses containing these two proteins and strengthening neurotransmission in these synapses [22] . Importantly , LGI1 regulates neuronal terminal pruning and maturation , again through a combined pre and post-synaptic action [21] . Humans not expressing or secreting LGI1 from one allele develop epilepsy starting in the vast majority of cases after age eight and seeming to persist in adulthood in many cases . Mice completely lacking LGI1 are normal until midway through the pruning phase of brain development ( ∼P13 ) , when LGI1 would normally have started being expressed , after which they develop seizures that progressively worsen as LGI1's amounts would normally have progressively increased , and die of violent convulsions by four weeks of life [22] , [28] , [29] . These results show that LGI1 is a vital protein , vital specifically in protecting the brain against seizures . In humans its partial loss results in epilepsy , and only epilepsy , and in the mice its complete loss leads to death from epilepsy prior to the presence of any other neurological symptom . This vital anti-epileptic role is mediated at least in part through the above three ADAM receptors suggesting that the LGI1-ADAM complexes and their related pathways are essential components of neural network electrical stability in the maturing and mature brain . We show in our study that lack of secretion of LGI2 is also associated with epilepsy , at an earlier stage of development , and secreted LGI2 interacts with the same ADAM receptors as LGI1 , suggesting that LGI2 participates in protecting the brain against seizures during the pruning phase of neurodevelopment at least in part through the same system utilized by LGI1 in the subsequent phase . Importantly , LGI2 expression is highest in the phase preceding pruning and epilepsy . This suggests that the LGI2 anti-epileptic activity anticipates the pruning phase , i . e . LGI2 acts during the network construction phase to help prepare a network that will not seize during the pruning phase . To date , there has been no compelling evidence-based theory of why so many epilepsies of childhood begin and end with the start and end of the pruning phase . Our results , combined with the body of LGI1 work , suggest the following . Construction of the initial network includes mechanisms , in which LGI2 participates , that ensure that the network will not seize during the pruning phase . Defects in these anticipatory anti-epileptic processes result in epilepsy as the massive changes of the pruning phase commence . The pruning phase itself encompasses mechanisms , in which LGI1 participates , that ensure that the pruned and remodeled network to serve the rest of the animal's life is electrically stable . These mechanisms are able to correct or compensate for earlier instabilities , e . g . those introduced by LGI2 deficiency , resulting in the remission that characterizes so many childhood idiopathic epilepsies . Of the remaining two LGI proteins , LGI4 appears not to have a major role in the central nervous system . Instead , its chief function appears to be in regulating neuron-Schwann cell interaction , its secretion defect resulting in inability of Schwann cells to correctly myelinate peripheral nerves , resulting in peripheral nervous system hypomyelination and the murine claw-paw phenotype [30] . LGI3's function , on the other hand , appears to be similar to that of LGI1 , as there is evidence that like LGI1 it regulates neurite outgrowth [31] . LGI3 starts being expressed at p7 in mouse , i . e . at the very start of the pruning phase , has steady and high expression throughout the brain in adulthood , and interacts with ADAM22 and ADAM23 , as well as presynaptic SNARE complexes [31]–[34] . However , LGI3 does not rescue the LGI1 knockout mouse epilepsy [22] , and therefore the two proteins are at least not interchangeable . Four phenotypes have been associated with LGI1 . The first is normalcy in the up to 33% of patients with heterozygous mutations in ADLTE families . Second is ADLTE in the remaining patients with heterozygous LGI1 mutations . The third is the recent realization that the acquired autoimmune epilepsy syndrome Limbic Encephalitis , long thought to be due to auto-antibodies against a potassium channel , is in fact due to an auto-antibody against LGI1 . As expected , all patients with this condition are over 10 years of age [35] . The final phenotype is the intractable and fatal mid-pruning phase-onset murine epilepsy caused by complete LGI1 deficiency . In the clinic , we not infrequently encounter previously normal children with onset of explosive catastrophic epilepsy . Homozygous mutations in LGI1 ( and possibly LGI3 ) should be considered as a possible cause of this presentation . The phenotype associated with LGI2 in the present study is of a remitting epilepsy with focal onset and centrotemporal and occipital spikes on EEG , occurring within the age range equivalent to human two to 10 years . Two of the most common human epilepsy syndromes occur in this age range , Rolandic Epilepsy , which is focal in onset with centrotemporal spikes on EEG , and Panayiotopoulos Syndrome , again focal-onset , with both centrotemporal and occipital spikes on EEG . LGI2 should be considered a candidate gene in these common epilepsies .
We have collected blood samples from privately owned pets for our genetic studies and have a valid ethical permission for the proposed blood sampling in the study , ESLH-2009-07827/Ym-23 ( Oct 2009–Oct 2012 ) from the Animal Ethic Committee , The State Provincial Office of Southern Finland , P . O . B150 , 13101 Hämeenlinna . Mapping of the benign focal juvenile epilepsy ( BFJE ) locus in Lagotto Romagnolo was based on clinically studied litters from Finland [12] , German and Switzerland including 25 epileptic puppies , 17 healthy littermates and 12 parents . Furthermore , based on the retrospective questionnaire-based phenotype information , we expanded our study cohort to a total of 112 healthy LR dogs and 28 BFJE cases and collected also 36 sporadic dogs . Additionally , the study population included also five adult-onset epilepsy LR cases [12] and our clinically diagnosed juvenile epilepsy cases from other breeds including Barbets , Collies and German Shepherds ( Table S1 ) . Population-based allele and genotype frequencies were estimated from a population of 576 Lagotto samples from three different countries ( Table S2 ) . EDTA-blood samples were collected and genomic DNA was extracted using a commercially available kit ( Puregene , Gentra Systems , Minneapolis , MN ) . The Finnish Kennel Club's breeding database , Koiranet , was utilized for pedigrees . Altogether 22 dogs including seven discordant full sibs and four half-sibs were selected for GWAS . Genotyping was performed with Affymetrix's Canine SNP Array version 1 containing 26 , 578 markers ( Affymetrix , Santa Clara , CA ) . The SNP association analysis was performed with PLINK software [36] with the criteria of MAF <0 . 05 , call rate >75% and <25% of missing genotypes in individual dogs . After applying these filters , 17 , 273 SNPs remained in the analysis for all dogs . Genome-wide significance was ascertained through 10 000 random permutations of epilepsy phenotype . Exons and splice junctions were amplified by PCR with primers listed in . The PCR products were purified with ExoSAP-IT kit ( USB Corporation , Cleveland , Ohio ) and sequenced with an ABI Prism 3730xl DNA analyzer ( Applied Biosystems , Foster City , CA ) . To confirm that the Lgi2 is the causative gene in the associated 1 . 7 Mb region we sequenced four SNPs around the associated homozygozity region together with the mutation in 112 healthy and 28 epileptic Lagottos . Odds ratio was calculated using conditional maximum likelihood estimation and corresponding 95% CI was calculated from Fisher exact test . The calculations were done with R statistical software package . Absence of the mutation in other breeds was studied by sequencing epileptic cases from altogether 40 different breeds ( Table S3 ) . To study the effect of the nonsense mutation on the stability of Lgi2 transcript , total RNA was isolated from an affected and a healthy Lagotto dog from peripheral blood using PAXgene Blood RNA Kit ( PreAnalytix , Hombrechtikon , Switzerland ) . Total RNA isolated from the cerebellum of a Saluki puppy euthanized due to hernia diaphragmatica was used as an amplification control . cDNA synthesis was performed using the High Capacity RNA-to-cDNA kit ( Applied Biosystems , Foster City , CA ) , and exon 4- and exon 8-specific primers ( Table S4 ) were used to amplify Lgi2 by PCR . Transcriptional profiling of the LGI2 mRNA expression levels across a large number of human tissues was retrieved from the public GeneSapiens ( PMID: 18803840 ) database containing data from a meta-analysis of 9873 samples analyzed using the Affymetrix gene expression microarrays [37] . To study the effect of the mutation on the expression and secretion of LGI2 in cell culture , we obtained the human LGI2 clone from GeneScript Corporation ( Piscataway , NJ ) . The mutant LGI2 clone including the premature stop codon ( p . K534X corresponding to canine p . K518X ) was prepared from the wt clone and both were cloned into the pcDNA3 . 1D/V5-His vector in frame with the C-terminal V5-tag using pcDNA3 . 1 Directional TOPO Expression Kit ( Invitrogen , Carlsbad , CA ) . The recombinant constructs were confirmed by sequencing . HEK293 cells were grown in DMEM-GLUTAMAX medium ( Gibco Laboratories , North Andover , MA ) supplemented with 10% fetal calf serum ( FCS ) , 100 IU/ml penicillin , 100 µg/ml streptomycin and 1 mM Sodium Pyruvate and transiently transfected with the FuGENE 6 reagent ( Roche Diagnostics , Indianapolis , IN ) according to the manufacturer's instructions . Expression of the wt and mutant LGI2 were analyzed 48 hours after transfection by immunostaining on Western blots with anti-V5 antibodies from cell lysates and culture media . Media samples were concentrated 100-fold with Sentricon-10k concentrator ( Millipore , Billerica , MA ) before loading onto gel . GADPH was used as internal loading control using anti-GADPH antibodies . Proteins were visualized using the enhanced chemiluminescence method . COS7 cells were co-transfected using human V5-tagged wt LGI2B or LGI2B p . K534X mutant or rat FLAG-tagged wt LGI1 with mouse HA-tagged Adam22 or Adam23 . LGI1 and ADAM clones were described previously [18] . 24 hours after transfection , cells were fixed with 2% paraformaldehyde at RT for 10 min , blocked with PBS containing 10 mg/ml BSA and stained with anti-Flag or anti-V5 antibodies followed by Cy3-conjugated secondary antibody without permeabilization to visualize only the cell-surface bound LGIs . Then , the cells were permeabilized with 0 . 1% Triton X-100 for 10 min , blocked with PBS containing 10 mg/ml BSA , and stained with anti-HA polyclonal antibody , followed by Alexa488-conjugated secondary antibody . Fluorescent images were taken with a confocal laser microscopy system ( Carl Zeiss LSM 510; Carl Zeis , Oberkochen , Germany ) . To study the developmental expression of the Lgi2 transcript a colony of C57/BL6 mouse was established for tissue and RNA harvesting . Every other day after birth ( except days 5 and 17 ) one mouse was sacrificed and the forebrain of the cerebrum and the cerebellum were harvested and deep-frozen in liquid nitrogen before total RNA isolation by the QIAGEN RNeasy mini kit . The isolated RNA was DNase I-treated before RT-PCR by the SuperScript First-Strand Synthesis System ( Invitrogen , Carlsbad , CA ) . Quantitative PCR was performed using a SYBR Green method with MxPro-3005P multiplex Quantitative PCR systems . Lgi2-specific forward , atgtgtacgtggccatcgctca , and reverse , caaacttggtccagctctcgtcgta , primers were used for amplification . | Major remodeling of the neuronal synaptic network occurs during childhood . The quadrillion synapses formed till the end of age two are trimmed to 500 trillion by age 10 through a selective process of strengthening of ideal connections , removal of redundant ones , and formation of new contacts . Very little is known about the basic mechanisms that direct this massive reorganization that leads to the adult brain . The most common epilepsies of humans occur in childhood and are characterized by remission prior to adulthood . Not much is known about their genetics and basic remission mechanisms . We describe here a canine equivalent disease and identify the defective gene , Lgi2 . We show that the gene product is a secreted protein and interacts with neuronal ADAM receptors known to be involved in the regulation of synaptic remodeling in the developing brain . Our work sheds important light on the basic mechanisms of the most common neurological disease of children and discloses processes of epilepsy remission . The identification of the first focal epilepsy gene in dogs has also enabled the development of a genetic test to identify carriers for breeding purposes . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"genome-wide",
"association",
"studies",
"neurobiology",
"of",
"disease",
"and",
"regeneration",
"animal",
"genetics",
"genetic",
"mutation",
"genetics",
"molecular",
"genetics",
"biology",
"genetics",
"of",
"disease",
"neuroscience",
"genetics",
"and",
"genomics"
] | 2011 | LGI2 Truncation Causes a Remitting Focal Epilepsy in Dogs |
Obg proteins are a family of P-loop GTPases , conserved from bacteria to human . The Obg protein in Escherichia coli ( ObgE ) has been implicated in many diverse cellular functions , with proposed molecular roles in two global processes , ribosome assembly and stringent response . Here , using pre-steady state fast kinetics we demonstrate that ObgE is an anti-association factor , which prevents ribosomal subunit association and downstream steps in translation by binding to the 50S subunit . ObgE is a ribosome dependent GTPase; however , upon binding to guanosine tetraphosphate ( ppGpp ) , the global regulator of stringent response , ObgE exhibits an enhanced interaction with the 50S subunit , resulting in increased equilibrium dissociation of the 70S ribosome into subunits . Furthermore , our cryo-electron microscopy ( cryo-EM ) structure of the 50S·ObgE·GMPPNP complex indicates that the evolutionarily conserved N-terminal domain ( NTD ) of ObgE is a tRNA structural mimic , with specific interactions with peptidyl-transferase center , displaying a marked resemblance to Class I release factors . These structural data might define ObgE as a specialized translation factor related to stress responses , and provide a framework towards future elucidation of functional interplay between ObgE and ribosome-associated ( p ) ppGpp regulators . Together with published data , our results suggest that ObgE might act as a checkpoint in final stages of the 50S subunit assembly under normal growth conditions . And more importantly , ObgE , as a ( p ) ppGpp effector , might also have a regulatory role in the production of the 50S subunit and its participation in translation under certain stressed conditions . Thus , our findings might have uncovered an under-recognized mechanism of translation control by environmental cues .
Families of small P-loop GTPases are universally employed as molecular switches in all domains of life [1] . In E . coli there are ∼20 known GTPases , and most of them are involved in cellular processes related to ribosome functions . Among them , ObgE ( Obg in E . coli , also known as CgtA or YhbZ ) belongs to the highly conserved Obg GTPase family ( spo0B-associated GTP-binding protein ) . Obg proteins are essential for cell growth in all tested bacterial species ( [2] and references therein ) . The ObgE homologs are also widely present in eukaryotic organelles , including both chloroplasts and mitochondria . Despite extensive genetic studies of Obg proteins in several bacterial species , the exact molecular role of this protein family is still unclear . On the cellular level , it has been implicated in a variety of regulatory events in different species , including cell cycle control , DNA replication , stress response , sporulation , morphological development , and ribosome assembly ( reviewed in [2] , [3] ) . The pleiotropic phenotypes of Obg proteins indicate that this protein family might participate in certain essential processes that are central to various cellular functions . In line with this view , ObgE and its homolog in Vibrio cholera have been implicated in the ( p ) ppGpp-mediated stringent response [4]–[6] . The ppGpp ( guanosine tetraphosphate ) and pppGpp ( guanosine pentaphosphate ) , collectively called as ( p ) ppGpp , are important secondary messengers in bacteria and plants , regulating many cellular activities in response to various environmental stresses ( reviewed in [7] , [8] ) by targeting transcription factors , GTPases , as well as proteins in nucleotide and lipid metabolism [9] . The first evidence implicating Obg in the ( p ) ppGpp pathway was from the crystal structure of the Bacillus subtilis Obg protein , in which a ppGpp molecule was found in the active site of its GTPase domain ( GD ) [10] . Later , it was shown that ObgE binds to ppGpp with a physiological affinity in vitro [4] , and perturbation of ObgE function by different non-lethal mutations affects the total level of ( p ) ppGpp or the relative ratio of pppGpp/ppGpp during different growth phases [4] , [5] . Furthermore , ObgE and its V . cholera homolog were found to co-purify and interact with SpoT [5] , [6] , [11] , the cellular enzyme responsible for hydrolyzing ( p ) ppGpp to GTP or GDP [12] , postulating a possible role of ObgE on ( p ) ppGpp degradation by SpoT . These observations have established a link between Obg proteins and the regulation of ( p ) ppGpp , and suggested that many phenotypes associated with mutant Obg proteins might originate from impaired temporal control of the cellular ( p ) ppGpp level . In the meantime , converging biochemical evidences show that Obg proteins also physically interact with the ribosome or ribosomal subunits; the observations include bacteria Vibrio harveyi , E . coli , Caulobacter crescentus , Salmonella typhimurium , B . ubtilis , Chlamydia abortus , and Mycobacterium tuberculosis ( reviewed in [3] ) . It was also revealed that mutations in ObgE lead to cellular defects in the 50S subunit maturation [13] , [14] . Moreover , Obg proteins from E . coli and S . typhimurium were shown to have physical [15] or genetic interaction [16] with 23S rRNA modification enzymes . On the basis of these data , a primary role of ObgE in the 50S subunit maturation was proposed [13] , [14] . Given the absolute requirement of ribosome function for all cellular activities , this provides an alternative explanation of pleiotropic phenotypes of Obg proteins in multiple disparate cellular events . In the present work , we demonstrate that ppGpp enhances the binding of ObgE to the 50S subunit and promotes dissociation of the 70S ribosome into subunits . Interestingly , we find that ObgE plays an important role as a 50S based anti-association factor , which inhibits the formation of 70S ribosomes from the naked subunits as well as from an mRNA programmed 30S-preinitiation complex ( 30S-preIC ) . The inhibition in subunit association also leads to a slower dipeptide formation when ObgE is bound to the 50S subunit . Importantly , a C-terminus deleted construct of ObgE ( ObgE-NG ) , containing the N-terminal domain ( NTD ) and the central GD , is sufficient for its anti-association activity , suggesting that the NTDs constitute the main activity center of ObgE . Next , we solved the cryo-EM structure of the 50S subunit bound with ObgE . Structural data reveal that the NTD of ObgE is a structural mimic of the A-site tRNA , which exhibits specific interactions with the ribosomal peptidyl-transferase center . Together with previously published data , our results suggest that ObgE might be an important player in the ( p ) ppGpp regulatory circuit , regulating the assembly of the 50S subunit , and blocking the subunit association and further downstream events in protein translation in response to signal of the nutrient availability .
Previous studies showed that endogenous ObgE co-fractionates mainly with the 50S fraction [5] , [11] , [13] , [14] , In addition , GTP or GMPPNP was proven to stimulate the binding of ObgE to the 50S subunit [13] or to the 23S rRNA [14] . However , it is not clear whether ( p ) ppGpp could also modulate the binding of ObgE to the 50S subunit . To test this possibility , we performed co-sedimentation assay to examine the binding of ObgE to the 50S subunit in the presence of different nucleotides . Although ObgE in the apo state showed weak binding to the 50S subunit , addition of guanine nucleotides significantly enhanced its binding to the 50S subunit ( Figure 1A ) . Especially , ppGpp increased the occupancy of ObgE on the 50S subunit by over 5-fold , compared to the apo state . In addition , similar to other ribosome-interacting GTPases , ObgE showed a higher affinity to the 50S subunit in the presence of GTP or GMPPNP than GDP ( Figure 1A ) . These observations indicate that ppGpp binding , similar to GDP and GTP , indeed influences the interaction between ObgE and the 50S subunit . The marked effect of ppGpp , as well as the subtle difference with other nucleotides , on the 50S binding ability of ObgE , suggests that ObgE might sense changes in the nucleotide pool during different growth phases and adjust its behavior accordingly . Previously , it was shown that overexpression of the plasmid-encoded ObgE leads to the co-fractionation of the protein with the 70S , as well as polysome fractions [5] . We then sought to examine whether ObgE could bind to the 70S ribosome in vitro . However , unexpectedly , ObgE was found incompatible with the 70S ribosome when added in excess , and the latter was disassembled into subunits ( Figure 1B and 1C ) . Upon subunit dissociation , ObgE remained associated with the separated 50S subunits ( Figure 1D ) . Importantly , this dissociation of 70S ribosomes did not seem to require energy input from GTP-hydrolysis , because both GDP and ppGpp also enabled this 70S-spliting activity ( Figure 1C ) . As illustrated in Figure 1B , the 70S dissociation was dependent on ObgE concentration . Further comparison of the extent of 70S disassembly in the presence of different nucleotides indicates that the 70S dissociation activity of ObgE is well correlated with its 50S binding ability ( Figure 1A ) . As expected from the results presented in Figure 1A , the maximal dissociation was observed when ObgE was saturated with ppGpp ( Figure 1C ) . Obg proteins from different species contain very diverse C-terminal domains ( CTDs ) , connected to the GDs by long flexible linkers [3] , [17] . We therefore tested whether the CTD plays any role for the dissociation of 70S ribosomes . As shown in Figure S1 , deletion of the CTD does not impair the binding of ObgE-NG to the 50S subunit ( Figure S1A ) , and ObgE-NG is sufficient to promote dissociation of 70S ribosomes in a way similar to full-length ObgE; the maximal dissociation is seen in the presence of ppGpp ( Figure S1B and S1C ) . Altogether , these data indicate that ObgE binds to the 50S subunit preferentially over the 70S ribosome , and the association is likely mediated by its conserved NTD . Most likely , by binding to the 50S subunit ObgE prevents the re-association of the 30S subunit thereby shifting the 70S dissociation equilibrium toward free subunits in the steady-state . We have studied the effect of ObgE in ribosomal subunit association using naked 30S subunits or an mRNA programmed 30S-preinitiation complex ( 30S-preIC ) containing fMet-tRNAfMet in the P-site . The kinetics of subunit association was followed by monitoring the increase in light scattering after rapid mixing of the 30S ( or 30S-preIC ) and 50S subunits in a stopped-flow instrument . In both cases , the rates of subunit association , in the absence of ObgE , matched closely with previously published results [18] , [19] . ObgE , as expected from its preferential binding to the 50S subunit , showed a strong inhibition on the subunit association both for naked 30S ( Figure 2A ) and 30S-preIC ( Figure 2B ) . The rates of the subunit association [ ( kobs 30S = 4±1 s−1 ) and ( kobs 30S-preIC = 20±0 . 2 s−1 ) ] decreased gradually with increasing concentration of ObgE ( Figure 2C ) . Subsequently , the mean time for subunit association , estimated as the reciprocal of the observed rate ( 1/kobs ) , increased linearly with ObgE concentration ( inset in Figure 2C ) , thereby suggesting that ObgE competitively inhibits the 30S subunit for binding to the 50S subunit . By comparing the concentration of ObgE required for half-maximal inhibition ( Figure 2C ) , it is evident that the ObgE-mediated inhibition is stronger in the 30S-preIC association than the naked subunit association . For the 30S-preIC association the Ki is around 0 . 3 µM , while the same for naked 30S association is 2 µΜ . Highly consistent with full-length ObgE , the C-terminal deleted ObgE-NG also showed inhibition of subunit association in an extent comparable to ObgE ( Figure 2A and 2B ) . For ObgE-NG the estimated Ki values for inhibition of 30S-preIC and naked 30S association are 1 and 3 . 7 µΜ , respectively ( Figure 2D ) . The inhibition was more profound in the presence of guanine nucleotides with ObgE , both GTP and ppGpp led to higher inhibition of subunit association ( Figure 2E ) , as expected from the higher affinity of ObgE to the 50S subunit in the presence of these guanine nucleotides ( Figure 1A ) . Thus , our results demonstrate that ObgE blocks subunit association and hence translation initiation by binding to the 50S subunit , and the NTD of ObgE is likely the main activity center . We have further tested the effect of ObgE in dipeptide synthesis starting from 30S-preIC or 70S-initiation complex ( 70S-IC ) . The formation of fMet-Leu ( ML ) dipeptide starting from 30S-preIC required about 200 msec ( kobs dipep 30S-preIC = 4 . 7±0 . 2 s−1 ) , which involved subunit association followed by peptide bond formation . In the presence of ObgE with 50S containing elongation mix ( see Materials and Methods for details ) , the rate of dipeptide formation slowed down about four times ( Figure 2F , kobs dipep 30S-preIC obgE = 1 . 2±0 . 05 s−1 ) . However , ObgE did not show any effect on the rate of dipeptide formation when 70S-IC was previously associated and ObgE was added with the ternary complex ( kobs dipep 70S-IC = 33±2 s−1 and kobs dipep 70S-IC ObgE = 30 . 3±1 s−1 ) ( Figure 2F , inset ) . Thus , our results suggest that the ObgE has no effect on peptide bond formation . The defect seen in dipeptide formation starting from 30S-preIC was due to its anti-association activity . When ObgE is bound to the 50S subunit it blocks association and consequently the downstream steps in protein synthesis get inhibited . Following these observations , we tested the effect of ObgE on translation in a multiple turn-over reaction . As expected , ObgE inhibits the in vitro translation of a reporter gene in a dose-dependent manner ( Figure S2 ) . Consistent with the in vitro data , overexpression of ObgE in E . coli cells leads to a slower growth ( Figure S3A and S3B ) and a substantial increase of free 50S fractions in the ribosome profile ( Figure S3C and S3D ) . Thus , both our in vitro and in vivo results suggest that ObgE acts as an anti-association factor . The role of IF3 as an anti-association factor is well-known [20] , which binds primarily to the 30S subunit and prevents premature association of the ribosomal subunits . We show that ObgE demonstrates a similar anti-association activity by binding to the 50S subunit . We next determined the cryo-EM structure of the 50S subunit bound with ObgE·GMPPNP . The cryo-EM density map was solved at a nominal resolution of 5 . 5 Å . As shown in Figure 3 , ObgE binds to the intersubunit face of the 50S subunit , at a position commonly used for the docking of translational GTPases . Specifically , the NTD of ObgE protrudes into the peptidyl-transfer center ( PTC ) , and its GD is situated between the bL12 ( adopted after a newly proposed ribosomal protein naming system described in [21] ) stalk base and the sarcin-ricin loop ( SRL ) of the 23S rRNA ( Figure 3A and 3B ) . However , we did not find extra densities that could be attributed to the CTD of ObgE , indicating that this domain is highly flexible . Structural superimpositions of ObgE with four translational GTPases ( IF2 , EF-Tu , EF-G , and RF3 ) on the 50S subunit all report a large steric clash ( Figure S4 ) , indicating that the binding of ObgE is incompatible with these translation factors on the 50S or 70S ribosomes . Both the 50S subunit and ObgE undergo conformational changes upon the complex formation . Compared with crystal structures of Obg proteins [10] , [17] , a large scale rotation between the NTD and GD ( Figure S5 ) is necessary to assume the 50S bound conformation . On the 50S subunit side , significant conformational changes are seen on uL1 stalk , bL12 stalk , helix 38 , helix 34 , helix 58 , as well as helix 69 of the 23S rRNA ( Figure 3C ) , all localized in the intersubunit face . These changes are well correlated with the local resolution map of the 50S·ObgE complex ( Figure 3D ) . Very interestingly , upon binding to ObgE , helix 69 is seen to have a massive movement by about 19 Å ( Figure 3C and 3F ) , which , if mapped on the 70S ribosome structure , would directly disrupt a strong intersubunit bridge ( B2a ) . The B2a is essential for intersubunit association to form the 70S ribosome [22] , and ribosome recycling factor ( RRF ) employs exactly the same mechanism to break the B2a during the ribosome recycling [23] . This large movement of helix 69 , as well as the observed conformational changes on bridges B1a ( helix 38 ) ( Figure 3C and 3E ) and B4 ( helix 34 ) ( Figure 3C and 3G ) , perfectly explains the anti-association activity of ObgE . Another intriguing conformational change takes place on the NTD of uL11 . Upon the binding of ObgE , the uL11-NTD becomes “invisible” in the cryo-EM density map ( Figure S6 ) . The flexibility of the uL11-NTD probably results from the interaction between the ObgE-GD and the bL12 stalk base ( H43-44 ) . The ObgE-NTD is composed of an eight-stranded β-barrel base and a unique glycine-rich protrusion containing six left-handed helices of the poly-Pro type II conformation ( Figure S5 ) [10] , [17] . At the tip of the protrusion , connecting the six helices are three loops ( Figure S5B ) . These loops are conserved in both sequence and length among Obg family proteins ( Figure S7 ) . The binding environment of the ObgE-NTD on the 50S subunit is exclusively in rRNA helices ( Figure 4 ) , including helix 89 , helix 90 , helix 91 , helix 93 , and the A-loop ( helix 92 ) . Consistent with our structural model , most of the conserved lysine and arginine residues in the ObgE-NTD are located at the rRNA interface ( Figures 4C–4E and S7 ) . Specifically , the tip of the NTD shows tight polar interaction with the PTC ( Figure 4C and 4D ) . At the opposite end , a short helical insertion from the β-barrel base is also seen to interact with the junction between helix 89 and helix 91 ( Figure 4E ) . To be more specific , several highly conserved basic residues from the three intervening loops , including R24 , R25 , K27 , and K31 from loop 1; R76 , K81 , and R82 from loop 2; and R136 and R139 from loop 3 , are within 4 Å distance from a number of the PTC residues , such as U2493 , G2494 , U2504 , A2602 , C2573 , U2555 , C2558 , and C2507 ( Figure 4 ) . To confirm these structural observations , we introduced mutations to a few selected arginine or lysine residues on the three loops and tested the binding of ObgE mutants to the 50S subunit ( Figure 5 ) . As a result , all of the mutations impaired the binding , and especially , loop 1 ( K27EK31E ) and loop 3 ( R136GR139G ) mutants exhibited almost abolished binding activity . The sequence ( Figure S7 ) and mutational data ( Figure 5 ) indicate that the polar interactions between the ObgE-NTD and the PTC are highly likely very specific and conserved across species , which suggests that the recognition of the PTC is a universal function for the NTDs of all Obg proteins . Interestingly , the loop regions of the ObgE-NTD occupy the space that accommodates the acceptor arm of the A-site tRNA ( Figure 6A ) . A comparison with the crystal structure of the 70S·RF2 complex [24] indicates that the tip of the ObgE-NTD overlaps exactly with the GGQ-motif containing domain of RF2 ( Figure 6B ) . A close inspection at the PTC suggests that the two factors employ a very similar way to interact with this functional center ( Figure 6C–6H ) . Specifically , when the P-site tRNA is superimposed with the 50S·ObgE structure , residues I29–K31 from loop 1 of ObgE are capable of forming interaction with CCA-end of the P-site tRNA , and K31 is inserted between A76 of the P-site tRNA and A2451 of the 23S rRNA ( Figure 6E ) , displaying an astonishing resemblance to the GGQ-motif of RF2 ( Figure 6D ) . Besides the possible interaction with the P-site tRNA , the ObgE-NTD also interacts with the A-loop of the 23S rRNA , in a strikingly similar way as RF2 does ( Figure 6G and 6H ) . The structural similarity between the acceptor arm of a tRNA and the NTD of ObgE indicates that like many translation factors , ObgE also adopts a tRNA mimicry strategy to interact with the ribosome . The GDs of classical translational GTPases , such as IF2 [25] , EF-G [26] , EF-Tu [27] , and RF3 [28] , [29] , all show only limited contact with the bL12 stalk base ( containing uL11 , H43 , and H44 ) on the ribosome ( Figure 7 ) . Unlike these , the ObgE-GD itself interacts directly with the bL12 stalk base , and bridges the gap between SRL and bL12 stalk base ( Figures 7A and S6 ) . Other than that , compared with translational GTPases , the ObgE-GD is distinctively orientated on the 50S subunit , which places the Switch regions and the nucleotide binding pocket of the GD rather distant from the conserved A2662 of the SRL ( Figure 7A and 7B ) . This unprecedented placement of the ObgE-GD on the 50S subunit immediately raises the question whether the 50S subunit serves as the GTPase-activating protein ( GAP ) for ObgE , as does for classical translational GTPases . Therefore , we performed a GTPase activity assay on ObgE in the presence or absence of purified 50S subunits . As expected , the basal GTPase level of ObgE is relatively low , similar to previous measurements [4] , [11] , [16] . The hydrolysis rate in the absence of the 50S subunit is ∼0 . 0268 min−1; with a sufficient time ObgE is capable of converting all GTP to GDP ( Figure 7G ) . In the presence of increasing amounts of 50S subunits , the phosphate production is accelerated accordingly ( Figure 7H ) . When supplied with equal amount of 50S subunits , the hydrolysis rate is stimulated by about 120-fold , being about 3 . 15 min−1 . This moderate stimulation by the 50S subunit is in sharp contrast to translational GTPases , e . g . , the GTP hydrolysis on EF-Tu and EF-G is enhanced by the 70S ribosome by over seven orders of magnitude [30] . More importantly , the binding and subsequent GTP hydrolysis of translational GTPases are regulated by the dynamic bL12 stalk on the ribosome in very distinct ways [19] , [31]–[33] . It remains to be tested whether the bL12 stalk has any role in activating the GTPase of ObgE . Nevertheless , it is clear that ObgE represents a novel class of ribosome-interacting GTPases , whose GTPase activation mechanism should be different from those of classical translational GTPases .
In the present work , we reveal that the interaction between the PTC and the evolutionarily conserved NTD of ObgE is highly specific ( Figures 4 and 5 ) . This suggests that a primary molecular role of ObgE is directly related to the ribosome , which is consistent with the proposed role of ObgE in the ribosome assembly [13] , [14] . The assembly function of ObgE started with a genetic study showing that overexpression of ObgE could suppress the slow growth and ribosome profile defect of the ΔrrmJ strain [16] . RrmJ is a 23S rRNA methyl-transferase ( U2552 of the A-loop ) required for late-stage 50S subunit assembly [34] . Later it was also shown that Obg homolog from S . typhimurium physically interacts with another 23S rRNA modification enzyme RluD ( pseudouridine synthatase for ψ1911 , ψ1915 , and ψ1917 of helix 69 ) in vitro [15] . Consistently , pull-down experiment indicates ObgE also co-localizes with factors required for the 50S assembly , such as CsdA and DnaK [14] . Further analysis of the pre-50S subunits accumulated in an ObgE mutant ( G80ED85N ) strain [13] revealed several 50S maturation defects , including reduced binding of several 50S proteins ( e . g . , bL33 , bL34 , and uL16 ) , impaired 23S rRNA processing , and prolonged association of RluC ( pseudouridine synthase for ψ955 , ψ2504 , and ψ2580 ) and RrmJ with the pre-50S subunits . From our structural data , the binding position of ObgE is exactly next to the modification sites of RrmJ , RluD , and RluC ( Figure S8 ) . The release of ObgE from the 50S subunit , therefore , might mark the finish point of the 50S assembly , considering that ObgE might act very late in the assembly pathway [13] . In this sense , ObgE could be a checkpoint protein during the late-stage assembly , which monitors the modification status of these critical residues and local conformation of the PTC . The correctly modified and assembled 50S subunit then signals the GTP-hydrolysis and subsequent release of ObgE ( Figure 8A ) . Escape from this quality control mechanism results in hypo-modified 50S subunits into the translation pool , leading to a profound impact on translation . For examples , lack of methylation at U2552 increases translation accuracy at the expense of efficiency [35] , and deletion of rluD gene in E . coli results in a defect in translation termination , with an increased rate of stop codon read-throughs [36] , [37] . Therefore , the reported diverse phenotypes , associated with various ObgE mutants , might be at least partially originated from subtle changes on cellular translatome profile , related to specific proteins in different cellular events . The stronger interaction of ObgE with the 50S subunit in the presence of ppGpp ( Figure 1 ) , therefore , highly likely reflects another level of regulation on the ribosome assembly by ( p ) ppGpp during stringent response , which is to delay the 50S assembly by a prolonged association of ObgE·ppGpp with the pre-50S subunits , in addition to the well-known direct role of the ( p ) ppGpp on rRNA transcription [7] . When GTP is plenty in the mid-log phase , ObgE functions primarily as a 50S assembly factor to facilitate the 50S subunit maturation . In contrast , when the cells are challenged by nutrient limitation or enter stationary phase , the intracellular level of ppGpp sharply rises , and ObgE is dominantly modulated by ppGpp . As an effector , ObgE·ppGpp over-stays on the 50S subunits , and consequently , downregulates the subunit production ( Figure 8B ) . In the process of eukaryotic ribosome assembly , many 60S and 40S maturation factors also possess anti-association activity , and some of them have functional implications in translation initiation ( reviewed in [38] , [39] ) . One such example is a 60S subunit assembly factor eIF6 ( Tif6 in yeast ) , which binds to the subunit interface of the pre-60S particles beside the SRL [40] , [41] and blocks the subunit association and downstream initiation events . Mammalian eIF6 has also been shown to have a profound role in translation control ( reviewed in [42] ) . Interestingly , in yeast the release of Tif6 from the pre-60S particles by an assembly GTPase Efl1 is triggered by the maturation state of the P-site on the 60S subunit [43] . Taken together , it is apparent that there are common quality control mechanisms in the assembly of bacterial and eukaryotic ribosomes , which make use of the maturation states of functional centers ( such as PTC ) on the subunits as structural checkpoints . Our biochemical results demonstrate that ObgE possesses anti-association activity in addition to its role in the 50S subunit maturation . By binding to the 50S subunit ObgE prevents association of the naked 30S subunit as well as the properly programmed 30S-preIC to the 50S subunit . Our structural analysis shows that ObgE obstructs the binding of the 30S subunit by inducing significant conformational changes at several intersubunit bridging contacts on the 50S subunit , including B1a , B2a , and B4 ( Figure 3 ) . The anti-association activity of ObgE can have deep implications in protein synthesis and bacterial physiology . Under normal growth conditions , the binding of ObgE to pre-50S subunits prevents defective subunits from being engaged in translation , thereby minimizing the chance of inefficient and faulty protein synthesis . However , under stress conditions , when ppGpp level increases in the cell , ppGpp bound ObgE not only delays the maturation of the 50S subunit , but also sequesters a large number of mature 50S subunits from taking part into translation , thereby lowering the number of active 70S ribosomes and thus , regulating the rate of total protein synthesis in the cell . Given the universal distribution of Obg proteins and ( p ) ppGpp system in bacteria and eukaryotic organelles [44] , the action of ObgE might represent a conserved regulatory mechanism on translation in response to fluctuations in cellular energy level caused by nutrient availability . It must be noted that in bacteria both ( p ) ppGpp synthetase RelA and hydrolase SpoT are associated with the ribosome , and especially , the ( p ) ppGpp production by RelA is well documented to be strictly dependent on deacylated A-site tRNA on the 70S ribosome [45] , [46] . Our structural data show that the NTD of ObgE exactly mimics the CCA-end of the A-site tRNA . Furthermore , ObgE and SpoT were shown to co-fractionate with the pre-50S fractions [5] , [11] . All these pieces of information seem to suggest that ObgE might also have additional functional interplay with RelA and SpoT . Whether or not ObgE could directly act as a regulator of the ( p ) ppGpp pathway remains to be investigated . A handful of genetic studies have also implicated Obg proteins in other cellular processes , such as DNA replication , chromosome segregation , and other stress response pathways ( reviewed in [2] , [3] ) . Interestingly , many of these pleiotropic phenotypes associated with Obg dysfunction appear to be species-specific , and could be attributed to the high sequence diversity within the CTDs of Obg proteins ( Figure S7 ) . For example , dysfunction of ObgE in E . coli causes cellular defects in DNA replication and chromosome segregation [47]–[50] . Intriguingly , these mutations with defects in DNA synthesis are primarily located to the CTD and GD of ObgE . For another example , Obg proteins in B . subtilis and M . tuberculosis were demonstrated to be involved in σB-controlled general stress response [51] , [52] , and again , the CTD of the B . subtilis Obg is required for the binding to the anti-σB factor RsbW [53] . However , it must be stressed that many reported functions of Obg proteins are not independent of the ribosome or ( p ) ppGpp-mediated pathways . The DNA replication is known to be regulated by ( p ) ppGpp [54] , and regulators of the σB-dependent general stress response in B . subtilis also appear to be ribosome-associated [55] , [56] . Our structural data suggest that the binding of ObgE induces a conformational change on the uL11-NTD , resulting in the displacement of the uL11-NTD from its normal position ( Figure S6 ) . This is highly consistent with roles of uL11 as key regulators in both the stringent response in E . coli [46] and the σB-dependent general stress response in B . subtilis [57] . In addition , we show that the conserved function of the ObgE-NTD is to interact with the 50S subunit , in a similar way as the A-site tRNA does , and the CTD is not required for its 50S binding and anti-association activity . Therefore , the species-specific functions of Obg proteins suggest that Obg proteins might act as a specialized translation factor , partnering , through their CTDs , with distinct players in different growth control pathways to regulate ribosome assembly and protein synthesis at given energy status [3] .
The gene for ObgE was amplified from E . coli DH5α genomic DNA using PCR with the following two primers: 5-GCCATATGATGAAGTTTGTTGATGAA-3 and 5-GCGGATCCTTAACGCTTGTAAATGAA-3 . The 1 . 17 Kb PCR products were digested by NdeI and BamHI ( New England Biolabs ) and ligated into the pET28a vector ( Novagen ) . A CTD deleted construct of ObgE , including coding sequence for residues 1–340 ( ObgE-NG ) , was similarly constructed . For site-directed mutations , pET28a-obgE was used as PCR templates and the following primers were designed: ObgE-K27EK31E , 5-CGCCGCGAAGAGTATATTCCGGAAGGCGGC-3 , and 5-GCCGCCTTCCGGAATATACTCTTCGCGGCG-3; ObgE-R76GR82G , 5-GCAAGCGGCGACTGTACCGGTAAGGGCGGTAAA-3 , and 5-TTTACCGCCCTTACCGGTACAGTCGCCGCTTGC-3; ObgE-R136GR139G , 5-TCCGTTAACGGTACACCGGGGCAGAAAACC-3 , and 5-GGTTTTCTGCCCCGGTGTACCGTTAACGGA-3 ( mutated bases were underlined ) . The PCR products were digested with DpnI ( New England Biolabs ) to remove the template . The mutant of ObgE-K27EK31E/R76GR82G/R136GR139G was generated similarly using double-mutant plasmids as templates . E . coli BL21 ( DE3 ) cells transformed with wild type pET28a-obgE were grown in 1 . 0 liter LB medium at 37°C to OD600 of approximately 0 . 6 to 0 . 8 . Protein expression was induced with 1 mM isopropyl-β-D-thiogalactopyranoside ( IPTG ) at 30°C for 5 h . Cells were collected at 5 , 000 rpm in a JLA 10 . 500 rotor ( Beckman Coulter ) for 10 min and resuspended in 40 ml lysis buffer ( 20 mM Tris-HCl [pH = 7 . 5] , 500 mM NaCl and 50 mM imidazole ) . Cells were lysed by sonication , and clarified lysates were obtained by centrifugation at 15 , 000 rpm in a JA 25 . 50 ( Beckman Coulter ) for 30 min . Lysates were loaded onto a Ni-NTA column ( GE Healthcare ) , washed with 20 ml lysis buffer , and eluted with 10 ml elution buffer ( 20 mM Tris-HCl [pH = 7 . 5] , 500 mM NaCl and 500 mM imidazole ) . Protein fractions were desalted through a desalting column ( HiPrep 26/10 Desalting , GE Healthcare ) with desalting buffer ( 20 mM Tris-HCl [pH = 7 . 5] , 150 mM NaCl ) , then subjected to a RESOURCE Q column ( 1 ml , GE Healthcare ) , and eluted with a 20 ml linear gradient of NaCl from 150 to 1 , 000 mM . Mutant variants of ObgE ( point mutations and CTD truncation ) were similarly expressed and purified . Purified proteins were finally concentrated to 10 mg/ml on a 6 ml spin filter ( Satorius Stedim Biotech ) . E . coli 70S ribosomes were purified as described previously [58] . Purified 70S ribosomes were further centrifuged through a 10%–40% sucrose gradient with 2 mM MgCl2 to obtain separated 30S and 50S subunits . 50S fractions were pooled and with buffer changed to Binding buffer ( 20 mM Tris-HCl [pH = 7 . 5] , 100 mM NH4Cl , 10 mM MgCl2 ) . E . coli cells of the BL21 strain , with or without the pET28a-obgE were grown at 37°C to OD600 of ∼0 . 5 . The cell cultures were diluted to a series of concentrations ( 10−3 , 10−4 , 10−5 , 10−6 , and 10−7 ) . 2 µl of each dilution was dropped on LB plate ( 1 mM IPTG ) and incubated at 37°C for 10 h . E . coli cells of BL21 strain with or without pET28a- obgE were grown at 37°C to OD600 of ∼0 . 5 , and 1 mM IPTG was added to both cultures . After 5 h incubation at 30°C , cells were lysed by sonication , and clarified at 15 , 000 rpm in a JA 25 . 50 ( Beckman Coulter ) for 30 min . Equal amounts of cell extracts were loaded gently onto the top of a12 ml 10% to 40% sucrose gradient in Binding buffer , and the gradients were centrifuged in a SW41 rotor ( Beckman Coulter ) for 3 . 5 h at 39 , 000 rpm and 4°C . Gradients were analyzed using a Teledyne Isco fractionation system ( Teledyne Isco ) . Each reaction contained a mixture of equal amounts of purified 50S subunits and His-ObgE ( ∼30 pmole , final concentration 1 µM ) in the absence or presence of 0 . 5 mM GTP , GMPPNP , GDP ( Sigma-Aldrich ) , or ppGpp ( TriLink BioTechnologies ) . After incubation in Binding buffer for 10 min at 37°C , samples were gently loaded onto the top of 150 µl 33% sucrose cushion in Binding buffer and centrifuged at 330 , 000 g at 4°C for 4 h in a TLA120 . 1 rotor ( Beckman Coulter ) . The supernatants were rapidly removed and the pellets were resuspended with 20 µl of Binding buffer . 10 µl of resolved pellets were loaded onto a 12% SDS-PAGE and the presence of His-ObgE was examined by Western blot analysis using a mouse anti-His antibody as the primary antibody , and a goat anti-mouse IgG ( coupled to horseradish peroxidase ) as the secondary antibody . Each reaction contained a fixed amount of 70S ribosomes ( 90 pmole , final concentration 1 µM ) , with varying amounts of ObgE or ObgE-NG , in the absence or presence of 2 mM GTP , GMPPNP , GDP , or ppGpp . The mixtures were incubated at 37°C for 10 min , loaded onto the top of a 10%–40% sucrose gradient in Binding buffer , and centrifuged in an SW41 rotor ( Beckman Coulter ) for 3 . 5 h at 39 , 000 rpm and 4°C . Gradients were analyzed using a Teledyne Isco fractionation system ( Teledyne Isco ) . In Figure 1D , fractions were treated with 20% trichloroacetic acid at 4°C overnight . The pellets were isolated by centrifugation , resolved by SDS-PAGE , and then examined by Western blot analysis . >All translation components were from E . coli and purified as previously described [18] , [19] . Two reaction mixes were prepared in HEPES polymix buffer ( pH 7 . 5 ) [18] . Mix A contained either only 30S subunit ( 0 . 5 µM ) ( referred to as “naked subunit” ) or a 30S-preinitiation complex ( 30S-preIC ) containing 30S ( 0 . 5 µM ) , 2 µM of each of XR7 mRNA encoding MLL , fMet-tRNAfMet initiation factors 1 and 2 ( IF1 and IF2 ) , and GTP ( 100 µM ) . Mix B contained 50S subunit ( 0 . 5 µM ) alone , with ObgE or ObgE-NG in varying concentrations ( 0 . 5–20 µM ) . After 5 min incubation at 37°C , equal volumes of mix A and B were mixed rapidly in a stopped flow instrument equipped with fluorescence detector set at 37°C . The extent of 70S formation was monitored by following the increase in Rayleigh light scattering at 425 nm and the observed rates ( kobs ) were derived by fitting the data using subunit association model described in [59] . The rates of subunit association were plotted as a function of final concentration of ObgE ( or ObgE-NG ) . The Ki values were estimated from the midpoint of the curves fitted with hyperbolic equation . To check the effect of guanine nucleotides , naked subunit association was followed in the absence or presence GTP and ppGpp ( 100 µM ) in the 50S mix containing ObgE ( 2 . 5 µM ) . The formation of fMet-Leu ( ML ) dipeptide was followed starting either from 30S-preIC or 70S-initiation complex ( 70S-IC ) . The initiation complexes were prepared by incubating 30S subunit or 70S ribosome ( 1 µM ) with XR7 mRNA encoding MLL ( 1 µM ) , f[3H]Met-tRNAfMet ( 2 µM ) , IF1 ( 1 µM ) , and IF2 ( 2 µM ) at 37°C for 10 min . In parallel , an elongation mix was prepared containing leu-tRNA synthetase ( 0 . 5 µM ) , tRNALeu ( 10 µM ) , leucine ( 0 . 2 mM ) , EF-Tu ( 10 µM ) , and EF-Ts ( 5 µM ) . For dipeptide formation starting from 30S-preIC , 50S ( 1 µM ) was added in the elongation mix . Both mixes were prepared in HEPES polymix buffer and contained GTP ( 1 mM ) , ATP ( 1 mM ) , phosphoenol pyruvate ( 10 µM ) , pyruvate kinase ( 50 µg/ml ) , and myokinase ( 2 µg/ml ) . For specific reactions ObgE ( 10 µM ) was incubated with the elongation mix at 37°C for 10 min . Equal volumes of the initiation and the elongation mixes were rapidly mixed in a quench-flow instrument ( RQF-3; KinTek Corp . ) . After definite time intervals the reactions were quenched , precipitated , and the peptides were analyzed by RP-HPLC as described in [18] . The PURExpress in vitro protein synthesis kit ( New England Biolabs , E6800S ) was used for analyzing the effect of ObgE on translation . Transcription and translation components from the kit were mixed with purified ObgE , in a ribosome to ObgE ratio of 1∶1 , 1∶2 , 1∶5 , 1∶10 , or 1∶20 . Reactions were started by adding a plasmid DNA template of dihydrofolate reductase ( DHFR ) , and carried out at 37°C for 1 or 2 h and terminated on ice . The reaction mixtures were separated by 12% SDS-PAGE to examine the production of DHFR . 7-methyl-6-thioguanosine ( MESG ) assay [60] was used to determine the GTPase activity of ObgE by measuring the absorbance increase at 360 nm . In this system , the inorganic phosphate ( Pi ) release during GTP hydrolysis by ObgE was quantified by a coupled enzyme reaction , in which a purine nucleoside phosphorylase ( PNPase ) and its chromogenic substrate MESG were used . The 1 . 6 ml reaction system contained 100 mM MOPS ( pH = 7 . 0 ) , 100 mM NaCl , 10 mM MgCl2 , 100 µM MESG , 0 . 1 mg/ml PNPase , and 7 . 5 µM GTP . Reactions were initiated by adding ObgE to a final concentration of 3 . 5 µM and carried out at 25°C . The time courses of absorbance change were monitored using a Pharmacia Ultraspec III spectrometer and the Pi release was estimated with the molar extinction coefficient 11 , 200 M−1cm−1 for a phosphate-dependent reaction at 360 nm and pH 7 . 0 . To determine whether the 50S subunit could simulate the GTP hydrolysis by ObgE , increasing amounts of the 50S subunits ( 0 . 02 µM , 0 . 1 µM , 0 . 69 µM , 1 µM ) were mixed with 1 µM ObgE . The initial rates of reactions were calculated from the linear parts of the progress curves obtained . Nonenzymatic GTP hydrolysis was corrected by measuring the control reaction in the absence of ObgE . To reconstitute the 50S·ObgE·GMPPNP complex , purified 50S subunits ( 50 nM ) were incubated with ObgE in a ratio of 1∶80 in the presence of 2 mM GMPPNP for 10 min at 37°C . Aliquots ( 5 µl ) of samples were applied to carbon-coated Quantifoil 2/2 grids ( Quantifoil Micro Tools GmbH ) , and cryo-grids were prepared as previously described [58] . The specimen was examined on an FEI Titan Krios at 300 kV at liquid-nitrogen temperature . Images were recorded on an FEI eagle CCD camera ( 4K × 4K ) under low dose conditions ( ∼20 e−/Å2 ) , using a nominal 59 , 000× magnification ( effective pixel size of 1 . 502 Å ) . The data collection was performed with the software AutoEMation [61] . Micrographs were processed following standard reference projection matching procedures using SPIDER [62] with some modifications . Particles were first picked using a method based on a locally normalized cross-correlation function [63] with a 256×256 window size , subjected to correspondence analysis and then manually verified [64] . Due to the sub-stoichiometric binding of ObgE , all particles ( 223 , 274 in number ) were first classified in two groups , according to the presence or absence of ObgE on the 50S subunit using a modified supervised classification method ( Figure S9 ) [65] . The resulting 188 , 814 ObgE-containing particles were further applied to another round of 3D classification using RELION [66] . The particles were finally split into four groups in 30 iterations using a final angle sampling of 1 . 8 degree . One of the four groups , which displays the highest ObgE occupancy , with a total particle number of 102 , 814 , was used for final refinement ( Figure S10 ) . The refinement was performed using RELION [66] with the final sampling angle of 0 . 1 degree . The final resolution was reported by gold-standard FSC calculations [67] , as 5 . 5 Å according to FSC 0 . 143 criterion ( Figure S11 ) . Amplitude correction using the B-factor sharpening approach [68] was applied to the final volume . Local resolution map was calculated using the blocres program of the Bsoft package [69] . The atomic model of the E . coli ObgE was built with MODELLER [70] , using the B . subtilis and Thermus thermophiles Obg crystal structures ( PDB IDs 1LNZ and 1UDX ) [10] , [17] as templates . A crystal structure of the 50S subunit ( PDB ID 3OFC ) [71] , NTD and GD of the ObgE model were first manually docked as rigid bodies into the cryo-EM density map . The flexible fitting was performed using the molecular dynamics flexible fitting approach [72] . PyMol [73] and Chimera [74] were used for graphic visualization and figure preparation . Cryo-EM map of the 50S·ObgE complex has been deposited in the EMDataBank ( EMD-2605 ) . The atomic model has been deposited in the Protein Data Bank ( 4CSU ) . | GTPases commonly act as molecular switches in biological systems . By oscillating between two conformational states , depending on the type of guanine nucleotide bound ( GTP or GDP ) , GTPases are essential regulators of many aspects of cell biology . Additional levels of regulation can be acquired through the synthesis of other guanine nucleotide derivatives that target GTPases; for instance , when nutrients are limited , bacterial cells produce guanine tetraphosphate/pentaphosphate— ( p ) ppGpp—as part of the “stringent response” to adjust the balance between growth and survival . ObgE is a GTPase with many reported cellular functions that include ribosome biogenesis , but none of its functions is understood at the molecular level . Here we characterize , both biochemically and structurally , the binding of ObgE to its cellular partner , the 50S ribosomal subunit . Our results show that ObgE is an anti-association factor , which binds to the 50S subunit to block the formation of the 70S ribosome , thereby inhibiting the initiation of translation . Furthermore , the binding and anti-association activities of ObgE are regulated by guanine nucleotides , as well as by ( p ) ppGpp . We thus propose that ObgE is a checkpoint protein in the assembly of the 50S subunit , which senses the cellular energy stress via levels of ( p ) ppGpp and links ribosome assembly to other global growth control pathways . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"protein",
"interactions",
"molecular",
"complexes",
"enzymes",
"regulatory",
"proteins",
"enzymology",
"microbiology",
"ribozymes",
"developmental",
"biology",
"protein",
"synthesis",
"protein",
"structure",
"microbial",
"growth",
"and",
"development",
"microbial",
"physio... | 2014 | Structural and Functional Insights into the Mode of Action of a Universally Conserved Obg GTPase |
The blast fungus Magnaporthe oryzae threatens global food security through the widespread destruction of cultivated rice . Foliar infection requires a specialized cell called an appressorium that generates turgor to force a thin penetration hypha through the rice cuticle and into the underlying epidermal cells , where the fungus grows for the first days of infection as a symptomless biotroph . Understanding what controls biotrophic growth could open new avenues for developing sustainable blast intervention programs . Here , using molecular genetics and live-cell imaging , we dismantled M . oryzae glucose-metabolizing pathways to reveal that the transketolase enzyme , encoded by TKL1 , plays an essential role in facilitating host colonization during rice blast disease . In the absence of transketolase , Δtkl1 mutant strains formed functional appressoria that penetrated rice cuticles successfully and developed invasive hyphae ( IH ) in rice cells from primary hyphae . However , Δtkl1 could not undertake sustained biotrophic growth or cell-to-cell movement . Transcript data and observations using fluorescently labeled histone H1:RFP fusion proteins indicated Δtkl1 mutant strains were alive in host cells but were delayed in mitosis . Mitotic delay could be reversed and IH growth restored by the addition of exogenous ATP , a metabolite depleted in Δtkl1 mutant strains . We show that ATP might act via the TOR signaling pathway , and TOR is likely a downstream target of activation for TKL1 . TKL1 is also involved in controlling the migration of appressorial nuclei into primary hyphae in host cells . When taken together , our results indicate transketolase has a novel role in mediating - via ATP and TOR signaling - an in planta-specific metabolic checkpoint that controls nuclear migration from appressoria into primary hyphae , prevents mitotic delay in early IH and promotes biotrophic growth . This work thus provides new information about the metabolic strategies employed by M . oryzae to enable rice cell colonization .
The rice blast fungus Magnaporthe oryzae causes the most serious disease of cultivated rice [1] , [2] and is a significant challenge to global food security [3] , [4] . Infection involves the elaboration of a specialized structure , the appressorium , from a germinating three-celled conidium ( spore ) on the surface of the leaf [1] , [5] , [6] . Each cell of the conidium contains a nucleus , the most apical of which migrates into the germ tube during appressorium development where it undergoes closed mitosis [7] . One of the resulting daughter nuclei returns to the conidium to be degraded during autophagic cell-death [8]; the other enters the incipient appressorium [7] to become the source of M . oryzae genetic material during host infection . Disrupting autophagy or blocking the cell cycle at three checkpoints ( S-phase , mitosis and exit from mitosis ) abolishes infection [8] , [9] . At maturation , enormous turgor is generated in the appressorium through the accumulation of glycerol and the ingress of water that becomes trapped due to a layer of melanin deposited on the wall of the appressorium [10] , [11] . This hydrostatic pressure acts on a penetration peg [5] , [6] , forcing it through the leaf cuticle . In the cell lumen , the penetration peg becomes a thin primary hypha that invaginates the host plasma membrane before differentiating into bulbous invasive hyphae ( IH ) that are sealed in the plant-derived extra-invasive hyphal membrane ( EIHM ) compartment [12] . IH moves into adjacent cells by crossing the cell wall at plasmodesmata , and the biotrophic invasion process is repeated in successive rice cells [12] . After four to five days of biotrophic growth in susceptible cultivars , host cells die and M . oryzae enters its necrotrophic growth phase , causing spreading necrotic lesions on the leaf surface from which conidia are produced on aerial hyphae [1] . During appressorium development , lipid and glycogen stores in the spore are mobilized and transported to the appressorium in a manner dependent on the MAP kinase- and cAMP-dependent protein kinase A-signaling pathways [13] , [14] . Following lipolysis in the incipient appressorium [15] , a large body of evidence indicates that organelles and biochemical pathways that metabolize fatty acids , such as mitochondrial and peroxisomal β-oxidation and the glyoxylate cycle , are essential for the development of infection-competent appressoria [15]–[20] . Thus , fatty acid β-oxidation and the distribution of the resulting acetyl-CoA subunits into cellular processes such as cell wall biosynthesis [17] and pyruvate formation [19] are dominant biochemical pathways during appressorial development . In contrast to appressorial function ( which requires lipid catabolism through β-oxidation and the glyoxylate cycle ) , post-penetrative host infection is dependent on glucose 6-phosphate ( G6P ) sensing by trehalose-6-phosphate synthase 1 ( Tps1 ) and the concomitant production of NADPH through the pentose phosphate pathway [2] , [21]–[24] . The binding of G6P by Tps1 leads , via an NADPH-dependent signaling pathway , to the induced expression of genes encoding NADPH-requiring enzymes [22] , [24] and the repression of genes required for utilizing alternative sources of carbon [23] . At least two NADPH-requiring processes under Tps1 control , the glutathione and thioredoxin antioxidation systems , are glucose-responsive and required for biotrophic growth in rice cells [24] . Taken together , we have proposed that these observations are consistent with the occurrence of a metabolic shift from lipid metabolism by M . oryzae on the host surface to glucose metabolism in the host cell [25] . This metabolic reprogramming hypothesis is supported by the observations that the glyoxylate enzyme isocitrate lyase is required for appressorium function but is not required for growth in planta [16] . However , which glucose metabolizing pathways are active during M . oryzae growth in planta , and whether glucose metabolism contributes to appressorial development , is not known . Here , we sought to bolster our understanding of the metabolic strategies governing M . oryzae in planta growth by focusing on how glucose-metabolizing pathways contribute to rice blast infection . We used gene functional analysis and live-cell imaging to show how glucose-metabolizing pathways are not required for appressorium formation and function . Instead , glucose metabolism via transketolase ( encoded by TKL1 ) is essential for the post-invasive colonization of rice cells by M . oryzae . The loss of transketolase function resulted in Δtkl1 mutant strains that could form appressoria , penetrate host cells and elaborate IH , but were attenuated for growth and cell cycle progression in planta . Δtkl1 strains were depleted for ATP , and IH growth was restored in planta following the treatment of Δtkl1 strains with exogenous ATP , suggesting ATP acts downstream of transketolase to signal growth [26] , [27] . Moreover , evidence is provided to suggest ATP acts on the TOR signaling pathway and together our data indicate a functional connection between TKL1 , ATP and the TOR signaling pathway . These results are consistent with a novel role for transketolase in mediating an in planta-specific metabolic checkpoint that regulates the cell cycle , via TOR activation , in order to permit rice cell colonization by M . oryzae . This work thus gives fresh insights into the metabolic demands of biotrophy and points to novel signaling roles for transketolase that will be important for other areas of biology .
To develop a more robust understanding of the biochemical processes underlying plant infection by M . oryzae , we first sought to determine what glucose-metabolizing pathways might be important for rice infection . Following Berg and associates , [28] , we considered four metabolic destinations for G6P in M . oryzae: into glycolysis via the action of phosphoglucose isomerase 1 ( Pgi1 ) if more ATP than NADPH or ribose 5-phosphate is needed ( Mode 1 in Figure 1 ) ; into early glycolysis and the non-oxidative pentose phosphate pathway if ribose 5-phosphate production for nucleotide biosynthetic purposes is predominant ( Mode 2 in Figure 1 ) ; into the pentose phosphate pathway ( PPP ) and recycling via late gluconeogenesis until G6P is fully oxidized to CO2 if NADPH production is paramount ( Mode 3 in Figure 1 ) ; into the PPP and glycolysis to form pyruvate if NADPH and ATP production is required ( Mode 4 in Figure 1 ) . Searching the M . oryzae genome [29] , we found single orthologues of FBP1 ( MGG_08895 ) encoding fructose 1 , 6-bisphosphatase in the gluconeogenic pathway , PGI1 ( MGG_12822 ) involved in both glycolysis and gluconeogenesis , and TKL1 ( MGG_02471 ) encoding transketolase , a non-oxidative PPP enzyme linking the PPP and glycolysis . Using our high-throughput , PCR-based split marker method for gene deletion [22] , we replaced part of the coding regions of FBP1 and PGI1 with the ILV1 gene conferring sulphonyl urea resistance , and the coding region of TKL1 with the Bar gene conferring bialaphos resistance . Figure 2A shows that the resulting three mutant strains - Δfbp1 , Δpgi1 and Δtkl1- grew without impediment on glucose-rich complete media ( CM ) compared to wild type Guy11 strains . Despite these similarities in growth , Figure 2B shows that both PGI1 and TKL1 , but not FBP1 , contributed significantly ( Student's t-test p≤0 . 05 ) to sporulation rates on CM media . This indicates nucleotide biosynthesis via Mode 2 ( Figure 1 ) might be important for sporulation , perhaps due to the nucleotide demands of DNA replication during spore production . Interestingly , all the mutant strains in this study formed melanized appressoria on hydrophobic surfaces ( Figure 2C and 2D ) at rates that were indistinguishable from Guy11 ( Figure 2C , Student's t-test p>0 . 05 ) . This indicates that the glucose utilization pathways shown in Figure 1 are not required for appressorium formation and is consistent with previous work emphasizing the requirement for lipid metabolism and the glyoxylate cycle , rather than sugar metabolism , during appressorium formation [15]–[20] . Taken together , G6P metabolism via Mode 2 contributes to sporulation , but neither Mode described in Figure 1 is necessary for appressorial formation . Although all the mutant strains were able to form appressoria on artificial hydrophobic surfaces , Figure 3A shows that when spores were applied to whole plants , only the Δtkl1 mutant strain was unable to cause disease . TKL1 is thus revealed here as an essential and previously unknown determinant of pathogenicity by the rice blast fungus . The inability of Δtkl1 strains to infect rice leaves was not due to defective appressorium formation or function on host surfaces because , like Δpgi and Δfbp1 mutant strains , Δtkl1 formed appressoria on rice leaf surfaces at the same rates as Guy11 ( Figure 3B; Student's t-test p = 0 . 64 ) . Also , the rates of cuticle penetration on detached rice leaf sheaths were not significantly different ( Student's t-test p = 0 . 37 ) for Guy11 and any of the mutant strains generated for this study , including Δtkl1 ( Figure 3C ) . If appressorium formation and function did not depend on a functional Tkl1 enzyme , how , then , does TKL1 contribute to pathogenicity ? To address this question , we continued to characterize the role of theTKL1 gene in pathogenicity using live-cell imaging of detached rice leaf sheaths colonized with either Δtkl1 or Guy11 strains . At 48 hr post inoculation ( hpi ) , Δtkl1 mutant strains were found to elaborate bulbous IH from primary hyphae , but IH growth was severely restricted to the first infected cell compared to Guy11 ( Figure 3D ) . At 48 hpi , more than 80% of Guy11 , Δpgi and Δfbp1 primary infection sites had resulted in IH spreading to adjacent cells , but IH movement to adjacent cells by Δtkl1 mutant strains was not observed ( Figure 3E ) . Δtkl1 strains had also not spread to adjacent cells by 65 hpi ( Figure 3F ) . Growth rates in rice cells at 48 hpi was scored for 50 infected cells/strain ( repeated in triplicate ) using our 4 point scale ( where 1 = IH length shorter than 10 µm with no branching; 2 = IH length is 10–20 µm with 0–2 branches; 3 = IH length is longer than 20 µm and/or with more than 2 branches within one cell; 4 = IH has spread to adjacent cells ) . The average growth rate score for Δtkl1 strains in rice cells at 48 hpi was significantly reduced compared to Guy11 ( Figure 3G; Student's t-test p = 0 . 0078 ) . Therefore , TKL1 is not required for appressorium formation or function , or for IH elaboration , but is essential for strong biotrophic growth in rice cells and the movement of IH to neighboring cells . Figure 3A and 3E shows that early glycolytic and late gluconeogenic steps are dispensable for biotrophic growth in rice cells and disease development , whereas the non-oxidative PPP enzyme Tkl1 is essential . This indicates that G6P metabolism via Mode 4 ( Figure 1 ) is likely the dominant pathway for glucose utilization by M . oryzae during host infection . Further experimental support for Mode 4—which would involve Tkl1 connecting the PPP and later glycolytic steps to yield NADPH and ATP from glucose [28]—is shown in Figure 3G where , following axenic growth in minimal media ( MM ) , Δtkl1 mycelia was found by LC-MS/MS analysis to contain significantly less ATP than the mycelia of Guy11 strains ( Figure 3H ) . Therefore , ATP production is disrupted in Δtkl1 strains . We next sought to determine the molecular basis for the observed attenuated growth of Δtkl1 mutant strains in planta ( Figure 3 ) . First , we wished to ascertain if Δtkl1 mutants were senescent or killed in host cells after a short period of initial IH growth . We used quantitative real time PCR ( qPCR ) to measure M . oryzae actin ( MoACT1 ) gene expression in Guy11 and Δtkl1 mutant strains , normalized against rice actin ( OsACT2 ) gene expression , during the infection of rice epidermal cells at 48 hpi . The ratio of MoACT1 to OsACT2 gene expression was significantly higher in rice cells infected with Guy11 than Δtkl1 ( Figure 4A ) . This was to be expected considering the greater growth rate of Guy11 in rice cells ( Figure 3G ) . Nonetheless , MoACT1 gene expression was detectable in rice cells infected with Δtkl1 mutant strains . CT values for MoACT1 expression , before rice actin normalization , were 28 . 5±0 . 2 for rice cells infected with Δtkl1 mutants compared to 23 . 7±0 . 1 for those infected with Guy11 . These results indicate Δtkl1 mutant strains were alive in rice cells at 48 hpi and undertaking some gene transcription despite the reduced growth rate of these strains ( Figure 3G ) . Because Δtkl1 mutant strains were alive in rice cells , we next asked what molecular mechanism might account for their attenuated growth in planta . Quiescence is associated with arrested growth and has been described as either an extended G1 stage or a distinct G0 state [26] , [27] . Severely reduced growth states such as those exhibited by Δtkl1 strains also exhibit some features of quiescence [26] , [27] , [30] and it is not known if quiescence is a distinct state from slow growth [31] . Cellular quiescence occurs in yeast and mammals via the inactivation of the TOR signaling pathway in response to nutrient depletion . Conversely , the activated TOR signaling pathway promotes growth under favorable nutrient conditions . TOR inactivation results in , amongst other outcomes , cell cycle arrest [32] and translation suppression [33] , [34] . Translation suppression occurs at the level of ribosome biogenesis , including TOR-dependent downregulation of ribosomal gene expression [35] , [34] . To determine if the TOR signaling pathway might play a role in the poor growth of Δtkl1 strains in rice cells , we first examined the expression of two ribosomal genes , RS2 and RS3 ( encoding ribosomal protein S2-like and 40S ribosomal protein S3 , respectively ) , and the translation initiation factor eIF4G , in Guy11 and Δtkl1 mutant strains in planta at 48 hpi . eIF4G expression is reduced when mammalian mTOR is inhibited in an immortalized breast epithelial cell , and eIF4G depletion results in impaired cell proliferation and increased autophagy [36] . Compared to Guy11 , and following normalization against MoACT1 , both ribosomal protein genes ( Figure 4B ) and eIF4G ( Figure 4C ) were downregulated in Δtkl1 mutant strains in planta . To ensure that the observed downregulation of gene expression in Δtkl1 mutants in planta was not attributable to a general decrease in MoACT1 expression in Δtkl1 strains compared to Guy11 , we analyzed the expression of a putative glucose transporter encoded by RGT2 , previously shown to be highly and constitutively expressed during rice infection [37] . Figure 4D shows that the ratio of MoACT1 gene expression to RGT2 gene expression , in planta , is the same for both Guy11 and Δtkl1 , indicating MoACT1 gene expression was not reduced in Δtkl1 strains compared to Guy11 . We next sought to determine if RS2 , RS3 and eIF4G gene expression was under TOR signaling control in M . oryzae . We studied the expression levels of these genes under optimal growth conditions ( ie . minimal media containing the preferred carbon and nitrogen sources glucose and ammonium , respectively [2] , [23] ) , and following growth on the same media containing the specific TOR kinase inhibitor rapamycin [33] , [35] . Figure 4E shows that Guy11 grew poorly on preferred media when 50 nM rapamycin was added , confirming that optimal growth requires an activated TOR signaling pathway . Figure 4F and 4G shows , consistent with observations in yeast and mammalian cells , that poor growth in media containing rapamycin was associated with RS2 , RS3 and eIF4G downregulation . Taken together , Figure 4 suggests Δtkl1 mutant strains might be downregulated - in a TOR-dependent manner - for translation initiation and protein biosynthesis during growth in rice cells . In response to nutrient deprivation , TOR inactivation can arrest yeast cells in G1 to induce quiescence [33] . In addition , TOR inactivation prolongs the G2/M transition in yeast [33] . Thus , cell growth and cell cycle progression are linked by common signaling pathways , including TOR [33] . In order to continue investigating the role of TKL1 during in planta growth , we next turned our attention to the cell nucleus . We performed live-cell imaging of M . oryzae IH using a Guy11 strain expressing a histone H1 protein fused to the tdTomato variant [38] of red fluorescent protein ( H1:RFP ) - described in [9] - and the same strain lacking a functional copy of TKL1 due to homologous recombination with the ILV1 gene conferring sulphonyl urea resistance ( Δtkl1 H1:RFP ) . The individual nuclei of these strains were visualized by epifluorescence microscopy , and Figure 5A shows that at 32 hpi , when both strains had elaborated IH in rice cells and growth differences between the two strains were not as pronounced as at later timepoints , the IH of Guy11 H1:RFP contained significantly more ( Student's t-test p≤0 . 05 ) nuclei , per infected rice cell , than the IH of Δtkl1 H1:RFP strains ( quantified below ) . We also quantified the rate of mitosis in both strains at 32 hpi by measuring the number of nuclei per unit length of mycelia . Figure 5B shows that 10 µm lengths of Δtkl1 H1:RFP IH carried significantly less nuclei ( Student's t-test p≤0 . 05 ) than 10 µm lengths of Guy11 H1:RFP IH . Thus , attenuated in planta growth by Δtkl1 strains is accompanied by delayed ( but not arrested ) mitotic progression . Interestingly , no significant differences in nuclei number ( Student's t-test p>0 . 05 ) were observed in Δtkl1 H1:RFP vegetative hyphae compared to Guy11 H1:RFP following the growth of both strains in glucose rich CM ( Figure S1A and B ) or in defined glucose minimal media ( GMM ) with varying concentrations of glucose ( Figure S2 ) . This is consistent with Figure 2A that showsTKL1 is not required for optimal radial growth on CM . Together with the transcript data presented in Figure 4 , and the physiological data in Figure 3 , the results presented in Figure 5 support the notion that a functional TKL1 gene is required for cell cycle progression and growth during the biotrophic colonization of host rice cells . Why does loss of TKL1 result in mitotic delay and growth attenuation ? We speculated that , considering the Tkl1 enzyme is involved in glucose metabolism , the lack of a metabolite ( s ) or metabolic pathway ( s ) downstream of Tkl1 might impact , directly or indirectly , cell cycle progression . From Figure 3H we knew that Δtkl1 strains were significantly reduced in ATP production compared to Guy11 , such as would be predicted from perturbations to Mode 4 ( Figure 1 ) . In the absence of evidence for other metabolic changes , we hypothesized , then , that treatment of Δtkl1 mutant strains with exogenous ATP might remediate IH growth and affect mitotic rates upon host infection . To test this hypothesis , we first needed to determine if M . oryzae was capable of acquiring or metabolizing ATP from the external milieu . In a previous study , we had shown how an adenine-requiring mutant , Δade1 , could not grow on minimal media unless supplemented with adenosine or adenine [37] . Figure 6A shows how this same Δade1 mutant strain was remediated for growth on minimal media in the presence of ATP at a final concentration of 5 mM . This suggests that M . oryzae is capable of acquiring exogenous ATP directly from the media and , in adenine-requiring Δade1 mutant strains , this remediates axenic growth by converting ATP to adenine through the purine salvage pathway [37] . In addition , because Δade1 mutant strains can penetrate host cells but fail to establish infection , the growth tests in Figure 6A indicate plant sources of ATP , like adenine and adenosine [37] , are not available to M . oryzae during growth in rice cells . To determine whether exogenous ATP might interfere with the cAMP-signaling pathway controlling appressorium development [1] , we added increasing amounts of ATP to Guy11 spores and applied them to leaf surfaces . No significant ( Student's t-test p≤0 . 05 ) inhibition of appressorium formation was observed using final concentrations up to and including 5 mM ATP ( Figure 6B ) . Δtkl1 spores treated with the same concentrations of ATP were then applied to detached leaf sheaths . Live-cell imaging showed cell-to-cell movement was improved in Δtkl1 mutant strains following treatment with ATP in a dose-dependent manner ( Figure 6C and 6D ) . We next added 5 mM ATP to spores of our RFP labeled histone H1 strains and applied them to detached rice leaf sheaths . Figure 7A and 7B shows how the addition of exogenous ATP increased the total number of nuclei in the IH of Δtkl1 H1:RFP , in rice cells at 32 hpi , to levels comparable to those observed in the IH of Guy11 H1:RFP . The rate of mitosis was also remediated in Δtkl1 H1:RFP strains by ATP treatment ( Figure 7C ) . Similarly , at 48 hpi , the mitotic rate of untreated Δtkl1 H1:RFP was significantly less ( Student's t-test p≤0 . 05 ) than Guy11 H1:RFP but was restored by ATP treatment ( Figure S3A ) . However , at 65 hpi , there was no significant difference ( Student's t-test p≤0 . 05 ) in the rates of mitosis between Δtkl1 H1:RFP and Guy11 H1:RFP strains in untreated samples ( Figure S3B ) , suggesting mitotic delay is surmounted in Δtkl1 strains at later timepoints , perhaps in response to other metabolites . The remediating effect of ATP on Δtkl1 H1:RFP IH nuclear proliferation at 32 hpi was also observed when Δtkl1 H1:RFP spores were treated with the related compound adenosine ( possibly due to its conversion to ATP through the purine salvage pathway [37] ) , but not when treated with 0 . 4 mM of the unrelated ROS quencher diphehyleneiodonium chloride ( data not shown ) . Thus , we conclude exogenous sources of ATP can stimulate cell cycle progression in Δtkl1 mutant strains . The preceding results demonstrated that following host penetration , nuclear division rates in the IH of Δtkl1 H1:RFP mutant strains were reduced compared to those of Guy11 H1:RFP , but this mitotic delay could be avoided by treatment with ATP prior to infection . However , untreated Δtkl1 mutant strains could still form infection-competent appressoria on the host surface ( Figure 3B and C ) . This was interesting considering appressorial developmental involves , amongst other processes , cAMP signaling [1] and an active cell cycle [7] , [9] . How can these observations be reconciled ? To address this question , we applied spores of Guy11 H1:RFP and Δtkl1 H1:RFP strains to detached leaf surfaces and observed nuclei behaviour during appressorium development and early infection . Figure S4 shows that , up to 21 hpi , there was little difference in the behaviour of the nuclei of either strain , with each strain demonstrating three conidial nuclei at 13 hpi , and a punctate , appressorial nucleus at 18 hpi . Figure 8A and Figure S4 shows that by 21 hpi , no nuclei remained in the conidium of either strain , indicating that appressorial development and autophagic cell death of the conidium and its contents [8] had progressed normally in both strains . The timing of these events on rice cuticles is consistent with what has previously been demonstrated during appressorial formation on artificial hydrophobic surfaces [9] . However , at 21 hpi , whereas H1:RFP was still localized to a single , punctate appressorial nucleus in Guy11 H1:RFP strains , H1:RFP was diffused throughout the appressoria of Δtkl1 H1:RFP strains ( Figure 8A and Figure S4 ) . Differences were also seen at 24 hpi when , by this time , the single nucleus of each Guy11 H1:RFP appressorium had migrated from the leaf surface to primary hyphae in the host cell ( Figure 8A ) , but no nuclear migration was evident for Δtkl1 H1:RFP ( Figure 8A ) . However , Δtkl1 H1:RFP strains treated with ATP displayed a single , punctate appressorial nucleus at 21 hpi which had migrated into primary hyphae by 24 hpi ( Figure 8A ) . These results suggest TKL1 is not required for the cell cycle events that give rise to the single nucleus of the appressorium , or for autophagic cell death and degradation of the remaining nuclei in the conidium . Instead , the results presented here indicate TKL1 acts after autophagy , via ATP , to maintain a punctate appressorial nucleus and control its migration into primary hyphae in the host cell . Additional evidence that ATP acts after appressorial formation comes from the use of the non-hydrolysable analogue of ATP , adenosine 5′-adenylyl imidodiphosphate ( AMP-PNP ) . Treating Guy11 spores with AMP-PNP did not prevent appressorium formation on detached leaf sheaths but did abolish appressorium penetration of host surfaces ( Figure 8B ) . This could indicate AMP-PNP is not taken up from the environment until after the appressorium has formed , but this is not likely considering cAMP is readily taken up into the cell prior to appressorial development [25] . Rather , the results with AMP-PNP are consistent with TKL1 and ATP being necessary for cuticle penetration and post-penetration development but not appressorial development and autophagy . The results presented above suggest TKL1 temporally controls the migration of appressorial nuclei into primary hyphae , and is required for the subsequent establishment of IH biotrophic growth , using a mechanism that involves ATP . Because TOR inactivation slows growth and extends mitosis [33] , [34] , we considered that a functional relationship between TKL1 , ATP and the TOR signaling pathway might exist to promote biotrophic growth in planta . To tests this hypothesis and thus gain more mechanistic insights into the regulation of M . oryzae gene expression during biotrophic growth , we studied the transcript levels for a number of known TOR read-out genes [34] , [35] , [39] in Guy11 and Δtkl1 mutant strains in planta at 48 hpi . These included M . oryzae genes shown in a previous study [39] , or known from yeasts studies [35] , to be positively regulated by the activated TOR pathway , such as those involved in nitrogen uptake and utilization ( Nii1 encoding nitrate reductase , an aspartate semi-aldehyde dehygrogenase-encoding gene and GAP1 encoding the general amino acid permease [39] ) , a laccase putatively involved in cell wall modifications [39] , and the single M . oryzae TOR-encoding gene TOR1 [35] . We also studied the expression of the ATG8 gene involved in autophagy [8] , a process upregulated when TOR signaling is inactivated [34] . Figure 9A shows that all TOR readout genes were downregulated in expression in Δtkl1 strains , in planta , compared to Guy11 except ATG8 , which was more highly expressed in Δtkl1 strains . These expression patterns are consistent with reduced TOR activity in Δtkl1 strains compared to Guy11 during in planta biotrophic growth at 48 hpi . Gene expression patterns were restored in Δtkl1 mutant strains when the infection was repeated using Guy11 and Δtkl1 spores treated with 5 mM ATP ( Figure 9B ) . These results provide evidence that TKL1 activates the TOR signaling pathway via ATP ( or an ATP derivative ) in order to regulate gene expression in planta , promote biotrophic growth and prevent mitotic delay . Further evidence for a connection between ATP and the TOR signaling pathway in M . oryzae is shown in Figure 10 . Compared to axenic growth in liquid minimal media with glucose and ammonium ( preferred carbon and nitrogen sources for M . oryzae [2] , [23] ) , growth on the same media with 50 nM rapamycin [33] , [35] resulted in elevated expression of the autophagy gene ATG8 . This observation is consistent with studies in yeast , which showed that rapamycin inhibits the TOR signaling pathway and activates autophagy [33]–[35] . However , rapamycin induction of ATG8 expression in M . oryzae was not observed when exogenous ATP was also present ( Figure 10 ) . Thus in M . oryzae , at least under some growth conditions , exogenous ATP can positively influence the TOR signaling pathway and override negative-acting signals such as those deriving from rapamycin treatment . In summary , the results presented above suggest that TKL1 and ATP influence the expression of at least some known TOR read-out genes during biotrophic growth . How ATP affects the TOR signalling pathway , and the implications for biotrophy , require future elucidation .
M . oryzae is a serious threat to world rice harvests and spends most of its infection cycle as a symptomless biotroph growing cell-to-cell in rice leaves before the onset of necrosis . Compared to appressorial development on the host surface , little is known about the metabolic and physiological demands of the fungus during in planta growth [25] . Improving our basic understanding of the biology of biotrophy might contribute to devising effective , long-term resistance strategies using novel chemical or biological interventions . Here , we aimed to add to our knowledge of M . oryzae by describing how some glucose-metabolizing pathways contribute to rice blast disease . In doing so , we revealed that TKL1 , encoding transketolase , is essential for pathogenicity and controls biotrophic growth in rice cells via a mechanism involving ATP , TOR and the regulation of cell cycle progression . This work started with the hypothesis that M . oryzae undergoes metabolic reprogramming during infection , switching from lipid metabolism on the surface of the leaf during appressorium formation to glucose metabolism in host cells . On the leaf surface , β-oxidation and the glyoxylate cycle are required for developing infection-competent appressoria [16] , [17] , but at least the glyoxylate enzyme isocitrate lyase is not required for growth in planta [16] . In contrast , our genetic dissections of glucose metabolism in M . oryzae show that early glycolysis , late gluconeogenesis and the PPP are not required for appressorium formation or function on the rice leaf surface ( Figure 3B and C ) . Rather , glucose utilization through transketolase ( but not early glycolysis or late gluconeogenesis ) is essential for growth in host cells ( Figure 3A and D ) . Glucose utilization through transketolase likely occurs in the direction of glycolysis because ATP levels are reduced in the mycelia of Δtkl1 strains compared to Guy11 ( Figure 3H ) . Together , the genetic and biochemical evidence point to transketolase being important in connecting the PPP with glycolysis during in planta growth ( Mode 4 , Figure 1 ) . The critical post-penetration , in planta-specific role of TKL1 indicates that once in the host cell , M . oryzae metabolism might be dedicated to NADPH and ATP production from glucose . NADPH is likely important , amongst other processes , for recycling antioxidation systems [40] , [41] and maintaining redox balance during biotrophic growth [24]; ATP is likely required for meeting the energetic demands of the fungus in planta . A major and novel finding of this work is that ATP produced via Tkl1 can also act as a signal , likely upstream of the TOR pathway , to promote cell cycle progression and trigger the biotrophic growth of M . oryzae in planta . Delayed mitotic progression was reversed by the addition of exogenous ATP , but not unrelated compounds , to Δtkl1 spores , thus providing a functional connection between TKL1 , ATP and TOR . Indeed , exogenous ATP was shown to override the specific TOR kinase inhibitor rapamycin . Although we do not know how ATP acts on TOR , these results are consistent with observations in mammalian cells , where inactivated mTOR promotes cellular quiescence while mTOR activation is dependent on the monitoring of ATP levels , either directly [42] or via the AMP-sensing TOR regulator AMPK [43] . Therefore , one explanation for the results presented here could be that , following the transition of the fungus into the host cell , TKL1 is required to metabolize glucose and provide an ATP signal that mediates a metabolic checkpoint - via TOR - in order to control the cell cycle and promote hyphal growth in planta . Metabolic checkpoints sense metabolites to regulate cellular functions and have recently been implicated in cell cycle quiescence in hematopoietic stem cells [44] . However , although TOR can control quiescence in yeast [33] , [34] , Δtkl1 strains are likely in a slow growing rather than quiescent state ( Figure 3F and 3G ) . Moreover , in addition to extending G1 , inactivating TOR can affect the G2/M transition , resulting in G2-delay in yeast [33] , [45] . Therefore , we cannot state at this time where in the cell cycle of M . oryzae the TOR pathway regulates mitosis in response to ATP , except that its abrogation results in mitotic delay . Additional evidence that TKL1 mediates an in planta-specific metabolic checkpoint in M . oryzae comes from the observations that outside the host plant , TKL1 was not required for cell cycle regulation or autophagic cell death of the spore during appressorium formation and rice leaf penetration . Instead , TKL1 ( or exogenous ATP treatment of Δtkl1 strains ) was required to maintain a punctate appressorial nucleus and ensure the correct timing of nuclear migration from the appressorium into the primary hyphae in the plant . H1:RFP was not localized to a punctate nuclear structure in Δtkl1 H1:RFP strains following host penetration at 21 hpi and was instead diffused throughout the appressorium . In fibroblasts , different histone H1 subtypes have different cellular localizations during cell division . For example , Histone H1 . 2 is associated with chromatin during prophase but cytoplasmically localized during metaphase and early anaphase [46] . Histone H1 . 5 is partitioned between chromatin and cytoplasm during metaphase and early anaphase [46] . In addition , histone H1 diffusion dynamics in HeLa cells are affected by ATP depletion [47] . Thus , H1:RFP might be localized throughout the appressorium in Δtkl1 H1:RFP strains at 21 hpi due to altered cell cycle progression and/or altered histone dynamics in response to ATP ( compared to Guy11 H1:RFP at 21 hpi ) . This hypothesis remains to be tested . In Δtkl1 strains in the absence exogenous ATP , at least one nucleus has migrated into the IH of Δtkl1 mutant strains by at least 32 hpi ( Figure 5 ) , suggesting the TKL1-dependent metabolic checkpoint is reversible and perhaps responds to other metabolites . Reversibility is a hallmark of slow growth and quiescence and distinguishes it from other non-growing cell states such as senescence , apoptosis or terminal differentiation [31] . Taken together , our data suggests the following testable model of infection . Appressorial formation occurs under the nutrient-starvation conditions found on the host surface and requires autophagic cell death of the spore and recycling of the spore contents into the incipient appressorium [8] . β-oxidation and the glyoxylate cycle are the dominant metabolic pathways during this stage of development [25] . A single mitotic cell cycle event occurs in the germ tube and one daughter nucleus migrates into the incipient appressorium [9] . The remaining conidial nuclei are degraded during autophagy [8] , resulting in a sole appressorial nucleus . Following host penetration , and perhaps in response to G6P sensing by Tps1 [25] , M . oryzae metabolism switches to glucose metabolism through the PPP and transketolase , resulting in NADPH production for redox and ATP production for satisfying the energetic demands of the growing fungus . ATP is also a signal , likely acting via TOR pathway activation , to control the migration of the appressorial nucleus into primary hyphae in the host cell . Once IH have developed in the host cell , transketolase is further required to propagate the ATP signal and , via TOR pathway activation , prevent delayed mitotic progression in order to permit vigorous cell-to-cell growth in planta . Deleting TKL1 or treating spores with the non-hydrolysable ATP analogue AMP-PNP does not impact appressorium development but subsequent infection steps are delayed or abolished . The work presented here has highlighted mechanisms controlling the transition from appressorial development to biotrophic growth in rice cells , but important questions remain . We wish to understand 1 ) whether ATP affects TOR signaling due to a direct interaction of ATP and the Tor1 kinase , or indirectly due to an interaction between ATP and additional TOR interacting protein ( s ) or pathways; 2 ) how the germ tube mitotic cell cycle event occurs during autophagy when TOR signaling is presumably inactive , whereas nuclear division in IH requires an activated TOR pathway; 3 ) what is the relationship between autophagy in the spore , G6P sensing by Tps1 in planta , and the TOR signaling pathway; 4 ) how does G6P sensing by Tps1 switch carbon metabolism from β-oxidation in mitochondria and peroxisomes to glucose metabolism through the PPP in the cytoplasm . Addressing these points will require further , detailed explorations of the biology of M . oryzae biotrophy . In addition to shedding new light on the metabolic strategies governing rice infection , the characterizations of TKL1 function in M . oryzae might also have important clinical relevance . Metabolizing glucose through the PPP and transketolase to generate NADPH is a metabolic strategy observed in cancer cells [48] , activated macrophages [49] and the growth of T cells in response to pathogen challenge [50] , [51] . Similar to our observations with M . oryzae , the transition of naive T cells to actively growing T cells is accompanied by a metabolic shift from β-oxidation to glucose utilization via the PPP [50] , [51] . Moreover , transketolase , which controls the PPP along with G6PDH [52] , is upregulated in certain tumors [53] and has been implicated in metastasis [52] . Transketolase inhibitors have been shown to reduce the rate of proliferation of pancreatic adenocarcinoma cells in culture [54] while , conversely , stimulating transketolase activity in cancer cells using thiamine promoted tumor growth in mice [55] . Thus , switching to glucose utilization via transketolase might be a conserved feature of proliferating cells . Consequently , small-molecule inhibitors of transketolase , such as those developed as anti-cancer therapies [54] , [56] , could hold promise for combating recalcitrant intracellular fungal pathogens . The work presented here also contributes to our understanding of metabolic checkpoints by revealing a previously unknown functional relationship between TKL1 and TOR that provides new information on the regulation of the TOR pathway . This connection might have importance beyond the field of plant pathogenicity due to the role activated TOR plays in some cancers and other serious disorders such as cardiovascular disease [57] . Moreover , metabolic checkpoints are required for T cell differentiation and immune responses [51] , and this requires signaling through a metabolic checkpoint mediated by the mTOR pathway [51] . In conclusion , this study gives fresh insights into the metabolic strategies controlling , and committing , intracellular pathogens to biotrophic growth . This knowledge is likely applicable to a wide range of important fungal pathogens . Moreover , our characterization of transketolase as a metabolic checkpoint mediator might inform other areas of biology where this enzyme and its downstream targets are important , such as during cancer cell proliferation .
All strains used in this study are stored in the Wilson lab . Guy11 was used as the wild-type isolate [58] and all mutant strains described in this study were generated from the Guy11 parental strain ( Table S1 ) . M . oryzae was cultured and stored using standard procedures [37] . Strains were grown on complete medium ( CM ) or Cove's minimal media ( MM ) , as described previously [59] . Glucose was used in MM at a final concentration of 1% w/v . Nitrogen sources were used in MM at a final concentration of 10 mM . Physiological analyses were performed on media as described previously [24] . Plates were imaged with a Sony Cyber-shot digital camera , 14 . 1 mega pixels , after 10 days of growth . For sporulation rates , strains were grown on at least three independent CM plates for 12 days before the spores were harvested and spore concentrations were measured using a hemocytometer ( Corning ) . Appressorium developmental assays were performed on hydrophobic microscope coverslips ( Fisherbrand ) . Spores were harvested and diluted into 1×105 spores ml−1 in sterile distilled water . 200 µl of spore suspension was placed on three plastic coverslips per strain and placed in a plastic box with a wet paper towel on the bottom simulating a humid chamber . After 24 hr of incubation , the number of appressoria formed from 50 spores was determined for each replicate , and an average percent value generated . The non-hydrolysable ATP analogue adenosine 5′-adenylyl imidodiphosphate ( AMP-PNP ) was purchased from Sigma and added to Guy11 spores to a final concentration of 5 mM . The NADPH oxidase inhibitor diphehyleneiodonium chloride was purchased from Sigma and added to Guy11 spores to a final concentration of 0 . 4 mM , following [60] . Transformation-competent protoplasts were produced as previously described [22] . Guy11 was grown in liquid CM with agitation at 150 rpm for 48 hr . Mycelia was harvested and treated with lytic enzymes ( Glucanex , Sigma ) for 3 hr at 29°C in order to produce protoplasts . All targeted gene deletions mentioned in this study were generated using the split marker approach in which a selectable marker replaces the native gene of interest in the Guy11 genome ( following [22] ) . PGI1 , and FBP1 genes were replaced in Guy11 with the ILV1 gene conferring resistance to sulphonyl urea [22] . TKL1 was replaced in Guy11 with the Bar gene confirming resistance to bialaphos , and replaced in Guy11 H1:RFP with the ILV1 gene conferring resistance to sulphonyl urea [22] . In all cases , primers were designed to amplify a 1 kb sequence upstream and a 1 kb sequence downstream of the gene of interest ( Table S2 ) . These primers were used in a first round PCR reaction , which amplified both the right and left flanks of the gene of interest . The thermocycler conditions for the first round were 1 min at 95°C initial denaturation , followed by 34 cycles of 95°C for 30 sec denaturation , 63°C for 30 sec annealing and 68°C for 1 min extension . A second round of PCR was conducted , in which each flanking region of the gene of interest was fused to one of two overlapping pieces of the ILV1 or Bar gene . For the second round PCR , similar thermocycler conditions as first round were used except for a 3 min extension time . The resulting PCR products were transformed directly into protoplasts of M . oryzae . Homologous gene replacements by the ILV1 or Bar resistance markers were initially selected on the basis of sulphonyl urea or bialaphhos resistance , respectively . Strains carrying homologous gene replacement of the gene of interest were identified by PCR as described by Wilson et al . [22] using the oligonucleotide primers shown in Table S2 . The conditions were 2 min at 95°C initial denaturation , followed by 35 cycles of 95°C for 30 sec denaturation , 63°C for 1 min annealing and 68°C for 5 min extension . A minimum of two transformants was analyzed per gene of interest . To ensure that the observed phenotype for Δtkl1 mutant strains were solely the result of TKL1 deletion , a TKL1 complementation vector was constructed using the primers in Table S2 , following the protocol outlined in [24] , and introduced into Δtkl1 mutant strains to restore virulence . Rice blast pathogenicity assays were performed as described previously [37] . Briefly , three to four week old rice seedlings from a susceptible cultivar , CO-39 , were spray inoculated with spore suspensions at a rate of 1×105 spores ml−1 in a 0 . 2% gelatin ( Difco ) solution . Plants were placed in a growth chamber with 12 hr light/dark periods . After five days , the infected leaves were collected and scored for disease symptoms . Images of the infected leaves were taken by using an Epson Workforce scanner at a resolution of 600 dpi . Live-cell imaging of fungal colonization of rice epidermal cells was achieved using detached rice leaf sheaths from the susceptible cultivar CO-39 . Leaf sheaths were inoculated with fungal spores ( 1×105 spores ml−1 in 0 . 20% gelatin ) in the hollow interior of the sheaths as described previously [37] . Infected sheaths were kept horizontal in a glass container with humid conditions for up to 48 hr . Starting at 24 hpi , the rice sheaths were excised and observed under a light microscope ( Zeiss AxioSkop ) . For each strain , appressorium development from 50 spores was measured on each of three independent leaf surfaces at 24 hpi to obtain an average rate of appressorium formation . Appressorium penetration rates at 30 hpi , and IH growth rates and movement to adjacent cells at 48 hpi , were determined from fifty appressoria per treatment , repeated in triplicate , following [59] . IH growth rates were determined using a 4-point scale where 1 = IH length shorter than 10 µm with no branching; 2 = IH length is 10–20 µm with 0–2 branches; 3 = IH length is longer than 20 µm and/or with more than 2 branches within one cell; 4 = IH has spread to adjacent cells . Images were taken using a Nikon A1 laser scanning confocal mounted on a Nikon 90i compound microscope ( software version: NIS Elements 4 . 13 Build914 ) at the University of Nebraska-Lincoln Microscopy Center . A 1 . 5 zoom Z series step ( 1 µm ) was used in the 60× lens . Transmitted light and fluoresce for td tomato were imaged with a 561 . 5 nm laser . td tomato fluoresce was detected at 570–620 nm . The number of nuclei per 10 µm of IH within infected rice cells was quantified using ImageJ software ( rsbweb . nih . gov/ij ) . For all the images , the scale was set ( Analyze-Set scale ) and a 10 µm line was drawn inside the hyphae ( Analyze-Measure ) . The number of nuclei along the 10 µm line was counted for each strain at 32 , 48 and 65 hpi . For each timepoint , six independent replications were analyzed . Total RNA from infected leaf tissue or fungal mycelium was extracted using the RNeasy Plant Mini Kit from Qiagen . For leaf RNA extractions , deteached rice leaf sheaths were inoculated with 1×105 spores ml−1 of the appropriate strain , isolated at 48 hpi , frozen in liquid nitrogen and ground in a mortar with a pestle . For mycelial RNA extraction , strains were grown in CM for 48 hr before switching to Cove's minimal media containing 5 nM rapamycin , 5 mM rapamycin +5 mM ATP , or no treatment , for 16 hr following [23] . Mycelia was harvested , frozen in liquid nitrogen , and lyophilized for 24 hr . A total of 100 mg of each mycelial sample was used to perform RNA extractions . RNA from mycelia or rice leaf tissue was treated with DNase I ( Invitrogen ) and converted to cDNA using the qScript reagents from Quantas . The cDNA reactions were performed in 20 µl reaction volumes containing 1× of qScrip cDNA Super Mix ( 5× ) and 1–10 µg of total treated RNA . cDNA synthesis conditions were: 5 min at 25°C , 30 min at 42°C and 5 min at 85°C . After completion of cDNA synthesis , the first reaction was diluted 5 fold for PCR amplification . The resulting cDNA was analyzed by quantitative real-time PCR ( qPCR ) in an Eppendorf Mastercycler ep Realplex real-time PCR system . Reactions were performed in a 25 µl reaction containing 12 . 5 µl 2× Quantifast SYBR Green PCR Master Mix ( Qiagen ) , 100 nM of oligonucleotide primers and 2 . 5 µl ( ≤100 ng ) of cDNA . The primer sequences are provided in Table S2 . Thermocycler conditions were: 5 min at 95°C initial denaturation , followed by 40 cycles of 95°C for 30 sec denaturation , 63°C for 30 sec annealing and 72°C for 1 min extension . Gene expression of each gene was normalized against the expression levels of either the M . oryzae actin gene ( MoACT1 ) for in planta analysis or the M . oryzae β-tubulin gene ( TUB2 ) for mycelial transcript analysis , as described previously [23] . Results are the average of three technical replications and at least two biological replications . The analysis of nucleotides in mycelial samples was performed using LC-MS/MS by separation of the nucleotides using hydrophilic interaction chromatography ( HILIC ) . Samples of ground , lyophilized mycelia were weighed at about 10 mg per sample in seal proof Eppendorf tubes and extracted at 4°C with 0 . 6 mL of 80% MeOH/20% H2O ( both LC-grade , Fisher Scientific ) . The samples were subjected to extraction at room temperature using a Bullet Blender Homogenizer ( Averill Park , NY ) after the addition of ∼100 µL of zirconia beads . The samples were then centrifuged for 10 minutes , the supernatants were removed , and a second extraction with 0 . 4 mL of 50% MeOH/50% H2O was performed by the same procedure . After centrifugation of the samples , aliquots were transferred into plastic LC-vials capped with sealed septums and loaded onto the LC-MS/MS system . The latter consisted on an AbSciex 4000 Qtrap Hybrid LC-MS/MS system ( Framingham , MA ) operating in triple quad mode using a multiple reaction monitoring ( MRM ) method . The instrument was interfaced to an Agilent LC1200 which included an autosampler with the samples thermostated at 4°C . The nucleotides were separated using a Phenomenex ( Torrance , CA ) Luna-NH2 column with dimensions of 2×250 mm; 100 Å pore size and 5 µm particle size . The MRM transitions were monitored in positive mode following [61] . The transitions were optimized with pure nucleotide solutions by infusion of 10 µM solutions directly into the mass spectrometer , followed by final ion source optimization by loop injection of those solutions . Stock solutions of ATP were prepared and quantified by their UV/visible spectra using a Cary-100 spectrophotometer with available extinction coefficients . External calibration curves were generated for 2 transitions of the nucleotide by performing injections of serial dilutions of the stock using the HILIC column interfaced with the LC-MS/MS system . Plots of integrated areas vs . Concentration in the range from 50–1000 pmoles yielded linear fits with correlation coefficients >0 . 99 . The amounts of nucleotide was estimated from the integration of the more predominant MRM transition by comparison with the calibration curve and normalized by the total weight of the samples . Reinjection of the samples after 24 hr yielded the same areas , indicating that no significant hydrolysis of the nucleotides was occurring during LC runs . All solvents were purchased from J . T . Baker and were of LC-MS Grade Purity . | The blast fungus Magnaporthe oryzae destroys rice and wheat harvests and could compromise global food security . Following penetration into the rice cell , M . oryzae elaborates bulbous invasive hyphae that grow in living rice cells for most of the infection cycle without causing disease symptoms . Little is known about the physiological processes governing this important biotrophic stage of fungal growth . Here , we used gene functional analysis to show how the primary metabolic enzyme transketolase is essential for hyphal growth in rice cells . Loss of transketolase did not affect the ability of the fungus to gain entry into rice cells , but invasive hyphal growth was curtailed in transketolase null mutants . Biotrophic growth was restored in transketolase mutants by the addition of exogenous ATP . We conclude that M . oryzae metabolism is dedicated to metabolizing glucose through transketolase in planta in order to provide ATP as a trigger for biotrophic growth and infection . This work is significant because it reveals important—but previously unknown—metabolic strategies employed by M . oryzae to facilitate rice infection . These strategies might be open to abrogation by chemical or biological means and are likely relevant to other rapidly proliferating intracellular pathogens . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"biochemistry",
"plant",
"science",
"model",
"organisms",
"biology",
"and",
"life",
"sciences",
"microbiology",
"research",
"and",
"analysis",
"methods"
] | 2014 | Evidence for a Transketolase-Mediated Metabolic Checkpoint Governing Biotrophic Growth in Rice Cells by the Blast Fungus Magnaporthe oryzae |
Leishmania major parasites reside and multiply in late endosomal compartments of host phagocytic cells . Immune control of Leishmania growth absolutely requires expression of inducible Nitric Oxide Synthase ( iNOS/NOS2 ) and subsequent production of NO . Here , we show that CD11b+ CD11c+ Ly-6C+ MHC-II+ cells are the main iNOS-producing cells in the footpad lesion and in the draining lymph node of Leishmania major-infected C57BL/6 mice . These cells are phenotypically similar to iNOS-producing inflammatory DC ( iNOS-DC ) observed in the mouse models of Listeria monocytogenes and Brucella melitensis infection . The use of DsRed-expressing parasites demonstrated that these iNOS-producing cells are the major infected population in the lesions and the draining lymph nodes . Analysis of various genetically deficient mouse strains revealed the requirement of CCR2 expression for the recruitment of iNOS-DC in the draining lymph nodes , whereas their activation is strongly dependent on CD40 , IL-12 , IFN-γ and MyD88 molecules with a partial contribution of TNF-α and TLR9 . In contrast , STAT-6 deficiency enhanced iNOS-DC recruitment and activation in susceptible BALB/c mice , demonstrating a key role for IL-4 and IL-13 as negative regulators . Taken together , our results suggest that iNOS-DC represent a major class of Th1-regulated effector cell population and constitute the most frequent infected cell type during chronic Leishmania major infection phase of C57BL/6 resistant mice .
Leishmania spp . are protozoan parasites belonging to the Trypanosomatidae family . They are transmitted by phlebotomine sand flies to a variety of mammals , including humans and mice ( reviewed in references [1] , [2] , [3] ) . These organisms , under amastigote form , reside and multiply in late endosomal compartments of host phagocytic cells . Clinical manifestations of Leishmania infection vary with regards to the particular parasite species , the host immune response , and genetic factors , and much information has been gleaned from murine models of Leishmania major infection . The control of L . major and the development of long-lasting resistance require the interleukin ( IL ) -12 dependent differentiations of type 1 CD4+ T helper cells ( Th1 ) . The secretion of interferon ( IFN ) -γ by Th1 cells induces the expression of inducible nitric oxide synthase ( iNOS , also termed NOS2 ) by phagocytic cells , leading to the production of nitric oxide ( NO ) [4] . iNOS expression remains high in chronically infected , but clinically healthy mice , and is absolutely crucial for the sustained control of L . major [5] , [6] , [7] . Genetically resistant mouse strains ( e . g . C57BL/6 ) develop a strong Th1 response and restrict the spread of local parasite infection . In contrast , non-healing mouse strains ( e . g . BALB/c ) mount a Th2 response associated with high level IL-4 and IL-13 production by CD4+ T cells . C57BL/6 mice lacking MyD88 [8] , CD40 [9] , IL-12 [10] , IFN-γ [11] or CCR2 [12] display a Th2-skewed response , associated with a severe reduction in iNOS expression and high tissue parasite burdens . In turn , BALB/c mice lacking IL-4 [13] , IL-13 [14] or STAT-6 [15] develop a Th1-skewed response and are resistant to Leishmania infection . Dendritic cells ( DC ) play an essential role in initiating and shaping Th1 protective responses in Leishmania infection , mostly through production of IL-12p70 [16] , [17] , [18] . In the last decade , it has become clear that DC represent a highly heterogeneous cell population , with various subsets being defined by their differential expression of various cell surface markers and specialized functions [19] , [20] , [21] . Recently , a population of DC expressing CD11b+ CD11c+ LY-6C+ MHC-II+ and high levels of iNOS protein ( termed inflammatory DC or TNF-iNOS-producing DC ( Tip-DC ) ) [22] , [23] has been implicated in the resistance to infection by intracellular bacteria ( e . g . Listeria monocytogenes [22] and Brucella melitensis ) [24] . These observations suggest that these cells might be a potential source of iNOS during infection by Leishmania . In the present study , using immunofluorescent microscopy and ex vivo flow-cytometric analysis , we demonstrated that inflammatory DCs are the main producers of iNOS in vivo during the course of L . major infection . The recruitment of inflammatory DC was dependent upon CCR2 expression , and the induction of iNOS expression in these cells required the development of a local Th1 microenvironment , as demonstrated by the reduced frequency of iNOS+ inflammatory DC in MyD88−/− , CD40−/− , IL-12−/− and IFN-γ−/− mice . In contrast , a Th2 environment inhibited the local differentiation of iNOS+ inflammatory DC , as an enhanced frequency of iNOS+ inflammatory DC was observed in STAT-6−/− BALB/c mice .
Monitoring of L . major-induced lesion size in wild-type C57BL/6 ( B6 . WT ) , B6 . iNOS , B6 . TNF-α mice and wild-type BALB/C ( BC . WT ) mice confirmed the previous results of our group and others , showing an important contribution of both iNOS enzymatic activity [5] , [6] , [7] and TNF-α [25] in the resistance of B6 . WT mice to L . major infection ( Figure S1 . A ) . However , the severity of the lesions in the footpads four weeks post-infection ( p . i . ) highlighted a more crucial role for iNOS compared to TNF-α in this pathogenic model ( Figure S1 . A and S1 . B ) . To this point , the exact nature of the cell type ( s ) responsible for the production of iNOS and TNF-α in vivo has remained unclear , and we therefore attempted to directly identify them in the draining lymph node ( LN ) four weeks after primary L . major infection ( Figure 1A , gate R1 and R2 , respectively ) . Approximately 8500 cells per 2×106 ( 0 . 4% ) of the cells analyzed appeared positive for iNOS after intracellular staining , whereas 20-fold fewer cells expressed detectable level of TNF-α . A time course analyses of iNOS and TNF-α expression by LN cells from B6 . WT and BC . WT infected mice demonstrated that four weeks p . i . was the peak expression for these two proteins ( data not shown ) . Next , the phenotype of the iNOS+ and TNF-α+ LN cells in resistant B6 . WT and susceptible BC . WT mice was examined . The majority of iNOS+ and TNF-α+ cells expressed high levels of CD11b and CD11c , characteristic of inflammatory DC ( Figure 1B ) . Among the iNOS+ ( R1 gate ) and TNF-α+ ( R2 gate ) LN cells , resistant B6 . WT mice displayed ∼10-fold and ∼3-fold more CD11bhi CD11chi cells compared to susceptible BC . WT mice , respectively ( Figure 1B ) . The analysis of TNF-α and iNOS expression in the CD11bhi CD11chi cells-gated population of infected B6 . WT resistant mice demonstrated that a high frequency ( 37 . 9% ) of these cells expressed iNOS , whereas only a negligible fraction expressed either TNF-α ( 3 . 3% ) or both iNOS and TNF-α ( 1 . 6% ) ( Figure 1C , gate R3 ) . Together , these results demonstrated that iNOS-producing CD11chi CD11bhi DC are recruited to the draining LN to a much higher degree in infected B6 . WT mice compared to infected BC . WT mice , and that iNOS expression by these cells was much more prominent than TNF-α expression . To further characterize iNOS+ cells in the draining LN of L . major infected B6 . WT mice , an extensive phenotypic characterization of these cells was performed by flow cytometry ( Figure 2 ) . iNOS-producing cells also expressed F4/80 , CD115 , 7/4 , Ly-6C and Mac3 markers , but did not stain for CD4 , CD8α and Ly-6G markers . Additionally , CD40 and MHC-II molecules were also highly expressed , suggesting a potential antigen presenting function of these cells . Thus , the iNOS-producing cells induced during L . major infection phenocopied the surface phenotype of the “TNF-α/iNOS-producing DC , TipDC” or “inflammatory DC” , described in other infectious [22] , [24] and non infectious models [26] . To gain further insight into the infection level of iNOS-expressing DC in vivo , we infected mice with a strain of L . major which stably expresses the Discosoma Red ( DsRed ) fluorescent tracer protein ( DsRed-Leish ) [27] . As reported by the group of C . Ardavin [18] , flow cytometric characterization of infected cells in draining lymph nodes showed CD11b+ DsRed-Leish+ cells and CD11c+ DsRed-Leish+ cells ( Figure S2 . A ) . However , we also observed the unexpected presence of IgD+ DsRed-Leish+ cells and CD3ε+ DsRed-Leish+ cells ( Figure S2 . A ) . As this approach did not exclude the possibility that during purification ( i ) non infected phagocytic cells might become infected in vitro or ( ii ) extracellular parasite adhere to cells , we performed an experiment to detect “false-positive” DsRed-Leish+ cells ( Figure S2 . B ) . When LN cells from naïve uninfected BALB/C mice expressing the T cell congenic marker CD90 . 1 are mixed during ex vivo purification with LN cells from DsRed-Leish-infected BALB/C expressing the congenic marker CD90 . 2 , we found DsRed-Leish+ LN CD90 . 1+ cells . This result clearly demonstrated that the manipulation of infected cells/tissues ex vivo can result in the apparition of “false-positive” DsRed-Leish+ cells that are normally not infected in situ . Moreover , when cytospins were performed on sorted DsRed-Leish+ cells from infected BC . WT LN , we observed that a fraction of these cells do not exhibit an intracellular localisation of the parasite , but just display it bound to their cell surface ( Figure S2 . C ) . Consequently , these very important technical considerations led us to restrict our analysis of the cellular tropism of L . major infection in draining LN to in situ analysis . Large DsRed-Leish infected tissue sections were scanned at high resolution with a motorized fluorescent microscope . Resulting images were analyzed with the “Colocalization” module ( an AxioVision program , Zeiss ) to precisely identify the phenotype of both infected and iNOS-producing cells . These newly developed analysis techniques provide quantitative and statistically significant data , and avoid the potential loss of cell populations that might occur during the harvesting and processing of tissue before ex vivo analyses , such as cytofluorometric analysis . By performing a colocalization analysis of large scanned LN surfaces , it was confirmed that iNOS expression largely colocalized with CD11b+ and CD11c+ cells ( 75% and 65–70% of the iNOS+ surface , respectively ) , with approximately 40% and 35% also exhibiting DsRed signal and high expression levels of MHC-II , respectively ( Figure 3A ) . Using the same analysis technique , ∼75% of the DsRed-Leish+ surface colocalized with iNOS+ , CD11b+ and CD11c+ cells , but <30% did with Ly-6G and MHC-IIhi ( Figure 3B ) . The Figures 3C–E depicted representative examples of colocalization between CD11c+ , iNOS+ and DsRed-Leish+ in infected draining LN sections . They also illustrated the aggregated distribution of these cells within the draining LN . CD45R/B220 expression ( largely a B-cell marker ) displayed very little colocalization with iNOS+ and DsRed-Leish+ surface ( <5% , Figure 3A and 3B ) , and likely represented the general degree of non-specific colocalization ( due to cell superposition in the section ) when analyzing tissue sections averaging 10 µm thickness . We did not observe any iNOS staining in B6 . iNOS−/− mice ( data not shown ) . Analyses of tissue sections showed that approximately 70% of iNOS+ cell surface expressed CD11b and CD11c markers , yet only ∼35% colocalized with MHC-IIhi+ surface . This correlated with the flow cytometric analysis where MHC-II expression is decreased by ∼50% in the CD11b+ LN cells from infected B6 . WT mice compared to uninfected mice ( Figure S3A ) . MHC-II downregulation was observed in both resistant and susceptible mice ( data not shown ) . Figure S3 . B illustrated representative serial sections where colocalization is seen for CD11b+ , iNOS+ , MHC-IIhi+ and DsRed+ LN surfaces . Altogether , these results demonstrated that CD11b+ CD11c+ cells are by far the most abundant cells expressing iNOS , and are the most highly infected population of cells infected with L . major , within the LN at four weeks p . i . of B6 . WT mice . Few studies have investigated the phenotype of iNOS+ cells [7] or infected cells [28] in the L . major-induced cutaneous lesion . Using the same immunofluorescence microscopy technique developed for the analysis of the LN , we investigated which cells were infected in the cutaneous lesion of the B6 . WT footpad four weeks p . i . As in the draining LN , the majority of the iNOS staining colocalized with CD11b ( ∼75% ) , CD11c ( ∼60% ) , MHC-IIhi ( ∼60% ) and DsRed signal ( ∼40% ) ( Figure 4A ) . The surface occupied by CD11c+ iNOS+ staining corresponds to 3 . 55%+/−0 . 9% ( mean of eight lesion sections ) of the total section surface , which is determined by actin staining . We also observed that most of the DsRed-Leish+ surfaces colocalized with CD11b ( ∼65% ) , CD11c ( ∼55% ) and MHC-IIhi ( ∼55% ) , whereas only 20–25% colocalized with Ly-6G and iNOS ( Figure 4B ) . The Figures 4C–E showed representative examples of colocalization between CD11c+ , iNOS+ , MHC-II+ and DsRed+ signals in serial footpad sections . Figure S4A and S4B depicted colocalization between DsRed-Leish+ surface and CD11b+ , CD11c+ , MHC-II+ and Ly6G+ , or MHC-II , CD11c and Ly-6G in the footpad , respectively . Together , these analyses revealed that ( i ) a very similar pattern of expression exists for CD11c and MHC-II ( ii ) the majority of the Ly-6G+ area does not colocalize with CD11c or MHC-II and ( iii ) CD11b+ is expressed by CD11c+/MHC-II+/Ly-6G− ( DC ) and CD11c−/MHC-II−/Ly-6G+ ( granulocytes ) populations . In summary , these data revealed for the first time that inflammatory DC are the most abundant cell type expressing iNOS , whereas inflammatory DC and granulocytes are the most commonly infected cell type within the infected footpad of B6 . WT mice 4 weeks p . i . The lesion size during L . major infection was monitored in B6 . WT , B6 . TLR2/4 , B6 . TLR9 , B6 . MyD88 , B6 . TRIF mice and BC . WT mice . Our group and others have previously observed that MyD88 adaptor molecule [8] , and to a lesser extent TLR9 [29] , [30] , are critical innate sensing molecules that promote resistance to L . major in B6 . WT mice ( Figure S5A and S5B ) . Immunofluorescent microscopy was utilized to examine the levels of DsRed expressing L . major and to determine whether the size of the lesion directly correlated with increased L . major replication within the footpad . Indeed , the DsRed-Leish+ surface per footpad sections in B6 . TLR9 ( 19 . 52% ) and B6 . MyD88 ( 24 . 91% ) mice was increased when compared to B6 . WT ( 8 . 81% ) mice ( Figure S5C ) . Statistical analysis of these sections further confirmed this correlation ( Figure S5C ) and also suggested a negative role of TRIF in controlling L . major growth . In turn , the same analysis performed on the infected draining LN also highlighted a key role for MyD88 and TLR9 as well as a minor contribution of TLR2/4 in the control of L . major burden ( Figure S5 . E ) . Similar results were obtained when we determined the number of parasites per infected LN using a limit dilution assay ( Figure S6 ) . MyD88−/− and to a lesser extent TLR9−/− mice showed the higher level of living parasites . TLR2/4−/− mice displayed a slightly higher , but significant , level of living parasite per LN when compared to WT mice . TLR2−/− and TLR4−/− mice did not present any enhanced parasite count ( data not shown ) . We also excluded a possible contribution of the MyD88-dependent inflammatory pathway as infected IL-18 , IL-1beta-converting enzyme ( ICE ) -deficient mice did not exhibit increased parasite count compared to infected B6 . WT . Next , we investigated whether TLR9-MyD88 signaling was linked to iNOS production by inflammatory DCs . B6 . TLR9 and B6 . MyD88 mice , but not B6 . TLR2/4 and B6 . TRIF mice , showed statistically significant reductions in the number of iNOS+ CD11b+ CD11c+ cells when compared to B6 . WT mice , suggesting that L . major-derived PAMPs detected by this innate-sensing pathway lead directly to iNOS production by inflammatory DC ( Figure 5A ) . For decades , immune control of L . major infection has been associated with the development of a Th1-mediated response in B6 . WT mice , and the production of IFN-γ by CD4+ T cells [1] , [2] . In contrast , the Th2 cytokine profile ( i . e . IL-4 and -13 ) observed in L major infected , susceptible BC . WT is promoted through a STAT6-dependent signaling pathway [15] . Therefore , we examined the importance of various factors , i . e . cytokines and chemokines , implicated in the establishment of protective Th1 responses and playing a role in cellular recruitment , to determine which pathways might regulate the recruitment of iNOS+ CD11b+ CD11c+ cells to the infected draining LN . We found that IL-12p35 , CD40 , IFN-γ and CCR2 expression were absolutely required for recruitment/development of iNOS-expressing inflammatory DC in the infected LN , while TNF-α played a less important role in this process ( Figure 5B ) . Consistent with these results , STAT6-deficient BALB/C ( BC . STAT6 ) mice , which are defective in IL-4 and IL-13 signaling , showed higher levels of iNOS+ CD11b+ CD11c+ cells , when compared to infected BC . WT mice ( Figure 5C ) . These observations suggest that the presence of iNOS+ inflammatory DC in the infected LN is largely dependent upon the development of an IFN-γ-mediated Th1 protective response against L . major infection . In agreement with this hypothesis , we observed a statistical reduction in the frequency of IFN-γ+ TCRβ+ CD4+ LN T cells in B6 . TLR2/4 , B6 . TLR9 , B6 . MyD88 , B6 . CCR2 , B6 . IL-12p35 and B6 . CD40 mice when compared to B6 . WT mice ( Figure S7A–C ) . Moreover , BC . STAT6 mice displayed an increased number of IFN-γ-producing CD4+ T cells compared to BC . WT mice ( Figure S7 . D ) . In summary , these observations strongly suggest that the resistance to L . major infection is closely associated with the presence of iNOS-producing inflammatory DC , which seems dependent on the development of a Th1 microenvironment by IFN-γ-producing CD4+ T cells . To further dissect the mechanism for iNOS-mediated control of L . major infection , we examined whether the factors that were required to promote increased frequencies of iNOS-producing inflammatory DC in the draining LN functioned at the level of recruitment only , or whether they might directly induce iNOS production once the cells were present in the infected LN . We found that only CCR2 deficiency decreased the frequency of the CD11b+ CD11c+ cells in infected LN ( Figure 6A ) , while in turn STAT6 deficiency in susceptible BC . WT favored their recruitment ( Figure 6B ) . Figure 6C depicts representative flow cytometric analyses , summarizing the role of specific factors implicated in the recruitment of CD11c+ CD11b+ DC ( R1 gate ) .
Reactive oxygen intermediates ( ROI ) and reactive nitrogen intermediates ( RNI ) can damage DNA and several chemical moieties necessary for the replication/division of both host cells and pathogens , and their production constitutes an essential arm of the immune response to microbial infections . High level production of ROI and RNI is typical of infected phagocyte cells , including granulocyte and monocyte-derived cells [4] . ROI and RNI production seems to be largely redundant , as illustrated by the fact that gp91phox−/− ( ROI-deficient ) and iNOS/NOS2−/− ( RNI-deficient ) mice are viable in normal housing conditions [4] . However , in the case of Leishmania infection , gp91phox−/− C57BL/6 mice largely control Leishmania growth , whereas iNOS−/− C57BL/6 mice display a quite dramatic phenotype during the first weeks of infection [5] , [6] , [7] . Lesions in iNOS−/− mice appear highly necrotic 3–4 weeks after L . major infection ( [5] , [6] , [7] and Figure S1 ) and mutilation is observed before 5 weeks . In comparison , TNF-α−/− [25] , [28] , MyD88−/− [8] , IL-12−/− [10] or IFN-γ−/− [11] C57BL/6 mice , normally thought of as highly susceptible mice , display necrotic lesions only after 4 or 6 weeks of infection and mutilation is observed only after 8–10 weeks . These observations demonstrate that among Th1 effectors , RNI constitutes a non redundant and crucial immune mechanism for control of L . major growth . This is also substantiated by the fact that L . major infection can be reactivated in chronically infected healthy C57BL/6 mice following iNOS inhibitor treatment [7] . To this point , cells expressing iNOS in vivo during L . major infection have been only generally characterized [7] as macrophages and dendritic cells ( DCs ) based on their expression of F4/80 and NLDC-145 markers , respectively . DCs were originally described as the population of splenocytes which were responsible for promoting the mixed lymphocyte reaction . Such splenic DC , known as “conventional” DC , cDC , are present in all lymphoid organs and are essential for the induction of immunity [19] , [20] . However , the term “DC” now refers to a group of several cell populations in addition to cDCs that differ in their cellular origin , their localization and their role in immune response [23] . The antigen-presenting cell ( APC ) function of nearly all DC populations seems to remain their main general characteristic . Among DC subsets , inflammatory DCs ( also termed TNF-α iNOS-producing DC , TipDC ) produce TNF-α , nitric oxide ( NO ) , IL-12 and can stimulate T cells [18] , [22] . They are mainly defined by the expression of CD11b , CD11c , CD115 , MHC-II and Ly-6C markers and most likely derived from CD11b+ CD11c− CD115+ Ly-6C+ “inflammatory” monocytes that are recruited to inflamed tissues , spleen and lymph nodes [23] . They represent the major source of iNOS in the spleen from Listeria monocytogenes [22] and Brucella melitensis [24] infected mice . A recent work by the group of C . Ardavin [18] has reported that cells expressing the cell surface phenotype of inflammatory DC were recruited in skin lesions and draining lymph nodes of L . major infected mice . In our study , we formally demonstrated that these inflammatory DC are the main source of iNOS protein ( 70–75% of total iNOS-producing cells on tissue section and more than 90% by flow cytometry analyses ) and , therefore , potentially represent a key effector cells for the defence to L . major infection . Flow cytometry analyses of draining lymph node cells from C57BL/6 infected mice showed that 80–90% of iNOS+ cells express the typical phenotype of inflammatory DC: CD4− CD8α− CD11b+ CD11c+ CD40+ CD62L− CD115+ F4/80+ 7/4+ Ly-6C+ Ly-6G− Mac3+ and MHC-II+ . In agreement with this previous result , we confirmed that 60 to 70% of iNOS+ cells detected in tissue sections express CD11c+ and CD11b+ using immunohistofluorescence techniques and colocalization analyses . In comparison , granulocytes identified by Ly-6G expression represent only 10–15% of iNOS+ cells . Interestingly , the frequency of inflammatory DC expressing detectable level of TNF-α protein appears very small when compared to the frequency of iNOS+ cells , 2–6% and 30–40% , respectively ( Figure 1 . C ) . Among iNOS-producing inflammatory DC , the frequency of TNF-α+ cells is 0 . 5–2% , suggesting that in L . major model , like in B . melitensis model [24] , inflammatory DC can be mainly characterized by their iNOS production . Despite advances made in mouse models of L . major infection , many parameters regarding the nature and cell surface phenotype of infected cells remain poorly characterized . Initially , L . major-infected cells were largely thought to be macrophages . However , we reported that cells expressing high level of CD11c , a DC specific characteristic , are the most frequently infected cells in the draining lymph nodes of infected mice [28] . In the work from the group of C . Ardavin [18] , DsRed-expressing L . major has been used for flow cytometric characterization of infected cells in the lesions and draining lymph nodes . Monocytes ( CD11b+ CD11c− F4/80int Ly-6Chigh ) , macrophages ( CD11b+ CD11c− F4/80high Ly-6Cint ) and inflammatory DC ( CD11b+ CD11c+ F4/80int Ly-6Chigh ) were found infected . However , this approach did not exclude the possibility that during purification ( i ) non infected phagocytic cells might become infected in vitro , ( ii ) extracellular parasite adheres to cells or ( iii ) that cells infected in vivo might be lost ex vivo . Using flow cytometric analyses , we have detected the presence of “false-positive” DsRed-expressing cells in LN cells isolated from infected mice . This demonstrated that the manipulation of infected cells/tissues ex vivo can result in the infection of cells or adherence of parasite to cells that are normally not infected in situ . In agreement , when cytospins were performed on sorted DsRed-Leish+ cells from infected mice , we observed that a fraction of these cells do not show intracellular presence of parasite , but just display the parasite bound to their cell surface . Consequently , these very important technical considerations led us to restrict our analysis about the cellular tropism of L . major infection in tissues and draining LN to in situ analysis . We observed that 70–80% of DsRed signal colocalized with CD11b and CD11c inflammatory DC markers in tissue section from draining lymph node of C57BL/6 mice while only 20–30% costained with Ly-6G granulocyte marker and less than 5% with B220 marker ( used as negative control ) . More interestingly , 30–40% of iNOS+ signal overlap with DsRed and 70–80% of DsRed signal with iNOS . In lesion tissue sections , the phenotype of DsRed cells appeared similar to that observed in the draining lymph node with the exception of the iNOS marker . Only 20–30% of DsRed signal colocalized with iNOS . This difference could be explained by the very high level of DsRed signal found in lesion ( 5–10% of DsRed+ surface among lesion surface ) when compared to draining lymph node ( 0 . 01–0 . 1% of DsRed+ surface among lymph node surface ) . In total , these data demonstrate that inflammatory DCs are oftentimes infected in vivo by L . major , the iNOS+ subset being the most frequent of these and constituting the major infected cell population in draining lymph node . Recent studies in L . monocytogenes [31] and T . gondii [32] model have shown that Ly-6Chigh inflammatory monocyte recruitment to sites of infection involved CCR2-mediated emigration of monocytes from the bone marrow into the bloodstream . In agreement , we also observed a drastic inhibition of inflammatory DC recruitment into the draining lymph node of CCR2−/− C57BL/6 mice in the L . major model . Factors regulating the activation of effector functions of inflammatory DCs in vivo remain largely undetermined . In vitro studies have shown that regulation of iNOS gene expression is very complex . The murine iNOS gene promoter contains nearly 30 consensus binding sites for known transcriptional factors [33] , [34] . In the L . monocytogenes model , iNOS production by inflammatory DCs appeared MyD88 dependent [35] . In our L . major model , we observed a close association between susceptibility to infection and reduced iNOS production by inflammatory DCs . BALB/c susceptible mice displayed decreased recruitment and activation of inflammatory DCs when compared to resistant C57BL/6 mice . We took advantage of this model to try to identify important factors for regulating iNOS expression by inflammatory DCs . A defect in iNOS production , but not in recruitment , for inflammatory DC was observed in C57BL/6 mice deficient for MyD88 , TLR9 , CD40 , IL-12 , IFN-γ and TNF-α . In contrast , in STAT-6−/− BALB/c mice , that are defective for IL-4 and IL-13 signal transduction , the frequency of iNOS-producing inflammatory DCs is clearly enhanced when compared to wild-type BALB/c mice . In summary , these results demonstrated that Th1 and Th2 responses have opposite effect on effector function of inflammatory DC . Deficiencies in IFN-γ or factors affecting its production ( e . g . CD40 , IL-12 , and MyD88 ) in C57BL/6 mice negatively affect the frequency of iNOS-producing DC . As IFN-γ is mainly produced in our model by CD4+ T cells , this suggests that these cells have an important role in the regulation of inflammatory DC . Interestingly , the study from C . Ardavin group suggests that inflammatory DC could be responsible to the Th1 differentiation of CD4+ T cells during L . major infection because they produce IL-12 and display L . major-derived antigens associated to MHC-II molecules [18] . Thus , their data as well as ours suggest a positive cross-regulation between inflammatory DCs and CD4+ T cells during L . major infection . iNOS production by inflammatory DCs also required TNF-α as demonstrated by the fact that TNF-α−/− C57BL/6 mice display reduced frequency of iNOS-producing inflammatory DCs , despite of an extremely high frequency of IFN-γ-producing CD4+ T cells . On the contrary , neutralisation of Th2 responses enhances iNOS-expressing inflammatory DC frequency in BALB/c mice . These observations are supported by several in vitro studies on established cell lines showing that iNOS gene expression is positively regulated by IFN-γ [36] and TNF-α [37] and negatively regulated by IL-4 [36] and IL-13 [38] . In summary , our study showed a strong association between the recruitment and activation of inflammatory DC and the resistance to L . major . In addition , we showed that iNOS production by inflammatory DCs is positively regulated by Th1 response and negatively by Th2 response . Taken together , our results provide new insight into how innate and adaptive immune responses fight L . major infection . A better understanding of the mechanisms regulating inflammatory DC recruitment and activation could lead to new therapeutic strategies against Leishmania infection .
Genetically deficient mice in C57BL/6 background: TLR2/4−/− mice from Dr . T . van der Poll ( Academic Medical Center , The Netherlands ) , TLR9−/− [39] and MyD88−/− [40] were obtained from Dr . S . Akira ( Osaka University , Japan ) . TRIF−/− mice [41] were a kind gift from Dr . B . Beutler ( The Scripps Research Institute , CA ) , TNF-α−/− mice [42] from Dr . S . Magez ( Vrije Universiteit Brussel , Belgium ) , IL-12p35−/− mice [10] from Dr . B . Ryffel ( University of Orleans , France ) , iNOS mice [6] from Dr . G . Lauvau ( Université de Nice-Sophia Antipolis , France ) , IFN-γ−/− mice [43] from Dr . M . Moser ( Université Libre de Bruxelles , Belgium ) , CCR2−/− mice [44] from Dr . G . Brusselle ( Universitair Ziekenhuis Gent , Belgium ) . STAT-6−/− BALB/c mice [45] were obtained from The Jackson Laboratory ( Bar Harbor , ME ) . Wild type C57BL/6 mice and BALB/c mice , purchased from Harlan ( Bicester , UK ) , were used as control . All mice used in this study were bred in the animal facility of the Free University of Brussels ( ULB , Belgium ) . The maintenance and care of mice complied with the guidelines of the ULB Ethic Committee for the use of laboratory animals . Leishmania major promastigotes ( World Health Organization strain WHOM/IR/-/173 ) were grown in M199 medium containing 20% FCS . Discosoma Red ( DsRed ) Protein expressing promastigotes [27] were selected as previously described [46] . Leishmania major parasites were harvested in stationary phase after 6 to 8 days of culture growth , centrifuged ( 2 , 500 rpm , 10 min , 20°C ) and washed in PBS ( buffer ) . Promastigotes were purified by 10% Polysucrose ( Sigma ) gradient and washed three times in PBS before being used for infection . Mice were infected s . c . in the hind footpad with 106 promastigotes in a final volume of 25 µl . The thickness of infected footpads was weekly monitored with a metric caliper ( in mm; Kroeplin , Schlüchtern , Germany ) . Mice were killed at indicated times by cervical dislocation . Footpad lesions ( cut tangentially to the bone ground ) and popliteal draining lymph nodes were collected for cytofluorometric and microscopic analyses . Tissue parasite burden was determined by limiting dilution analysis Popliteal draining lymph nodes were harvested and digested with a cocktail of DNAse I fraction IX ( Sigma-Aldrich Chimie SARL , Lyon , France ) ( 100 µg/ml ) and 1 . 6 mg/ml of collagenase ( 400 Mandl U/ml ) at 37°C for 30 min . After washing , lymph node cells were filtered and first incubated in saturating doses of purified 2 . 4G2 ( anti-mouse Fc receptor , ATCC ) in 200 µl PBS 0 . 5% BSA 0 . 02% NaN3 ( FACS buffer ) for 10 minutes on ice to prevent antibody binding to Fc receptor . 3–5×106 cells were stained on ice with various fluorescent mAbs combinations in FACS buffer and further collected on a FACScalibur cytofluorometer ( Becton Dickinson , BD ) . We purchased the following mAbs from BD Biosciences: Biotin-coupled 53-2 . 1 ( anti-CD90 . 2 ) , AFS98 ( anti-CD115 ) , AL-21 ( anti-Ly-6C ) , M5/114 . 15 . 2 ( anti- IA/IE ) , 3/23 ( anti-CD40 ) , Fluorescein ( FITC ) -coupled OX-7 ( anti-CD90 . 1 ) , 1A8 ( anti-Ly-6G ) , RM4-5 ( anti-CD4 ) , 53-6 . 7 ( anti-CD8α ) , M1/70 ( anti-CD11b ) , Phycoerythrin ( PE ) -coupled HL3 ( anti-CD11c ) . Allophycocyanin ( APC ) -coupled BM8 ( anti F4/80 ) . Biotin-coupled 7/4 ( anti-neutrophil ) was obtained from Caltag Laboratories . Biotin-coupled mAbs were stained with FITC or PE-coupled streptavidin from BD Biosciences . The cells were analyzed on a FACScalibur cytofluorometer . Cells were gated according to size and scatter to eliminate dead cells and debris from analysis . Lymph node cells were treated as previously described [24] . Lymph node cells were incubated for 4 h in RPMI 1640 5%FCS with 1 µl/ml Golgi Plug ( BD Pharmingen ) at 37°C , 5%CO2 . The cells were washed with FACS buffer and stained for cell surface markers before fixation in PBS/1% PFA for 15–20 min on ice . These cells were then permeabilized for 30 min using a saponin-based buffer ( 1× Perm/Wash , BD Pharmingen in FACS buffer ) and stained with one or a combination of the following intracellular mAbs: Phycoerythrin-coupled M3/84 ( anti-Mac3; BD Biosciences ) , Phycoerythrin-coupled MP6-XT22 ( anti-TNF-α; eBioscience ) , allophycocyanin-coupled MP6-XT22 ( anti-TNF-α; BD Biosciences ) , allophycocyanin-coupled XMG1 . 2 ( anti-IFN-γ; BD Biosciences ) , purified M-19 ( rabbit polyclonal IgG anti-NOS2; Santa Cruz Biotechnology ) stained with Alexa Fluor 647 goat anti-rabbit ( Molecular Probes ) . After final fixation in PBS/1% PFA , cells were analyzed on a FACScalibur cytofluorometer . No signal was detectable with control isotypes . Draining lymph node cells from four weeks infected mice were washed 3 times in PBS , and spun down onto glass slides . Slides were air-dried overnight , fixed in acetone , stained with hematoxylin/eosin ( Vector Laboratories Inc . , Burlingame , CA ) and dehydrated in ethanol series . Slides were mounted and digitized image were captured using Zeiss inverted microscope ( Axiovert 200 ) equipped with high resolution monochrome camera ( AxioCam HR , Zeiss ) . Footpad lesions and lymph nodes were fixed for 3 h at 4°C in 1% paraformaldehyde ( pH 7 . 4 ) , washed in PBS , incubated overnight at 4°C in a 20% PBS-sucrose solution under agitation , and washed again in PBS . Tissues were embedded in the Tissue-Tek OCT compound ( Sakura ) , frozen at −80°C , and cryostat sections ( 10 µm ) were prepared . Tissues sections were rehydrated in PBS , then incubated successively in a PBS solution containing 1% blocking reagent ( Boeringer ) ( PBS-BR 1% ) and in PBS-BR 1% containing Alexa Fluor 488 phalloidin ( Molecular Probes ) and any of the following mAbs: purified 1A8 ( anti-Ly-6G ) , or rabbit polyclonal antibodies anti-NOS2 ( Calbiochem ) ( note that M-19 anti-NOS2; used for cytofluorometric analysis is not use for immunofluorescence microscopy ) , biotin-coupled M1/70 , HL3 and RA3-6B2 ( anti-CD45R/B220 , BD Biosciences ) as well as APC-coupled BM8 and M5/114 . 15 . 2 Uncoupled 1A8 mAb and anti-NOS2 polyclonal antibodies were detected using biotin-coupled R67/1 . 30 ( mouse anti-rat IgG2a , BD Biosciences ) and Alexa Fluor 647-coupled goat anti-rabbit IgG ( Molecular Probes ) in PBS-BR 1% , respectively . Biotin-coupled mAbs were amplified using Alexa Fluor 350 or Alexa Fluor 647 Streptavidin ( Molecular Probes ) in PBS-BR 1% . When two biotin-coupled mAbs were used , free biotin sites were saturated with an avidin-biotin blocking kit ( Vector ) . Slides were mounted in Fluoro-Gel medium ( Electron Microscopy Sciences , Hatfield , PA ) . Labeled tissues sections were visualized under a Zeiss fluorescent inverted microscope ( Axiovert 200 ) equipped with high resolution monochrome camera ( AxioCam HR , Zeiss ) . All images were acquired with 63× objective at maximal camera resolution . Acquisition of entire tissue section surface by automatic scanning and measurement of colocalization between two staining was realized using MosaiX module and Colocalization module , respectively , from AxioVision program ( Zeiss ) . When images were treated with The Colocalization module , double positive surface was stained in white ( as indicated in Figures ) . We have used a ( Wilcoxon- ) Mann-Whitney test provided by GraphPad Prism program to statistically analyze our results . Each group of deficient mice was compared to wild type mice . We also compared each group to each other and displayed the result when it is required . Values of p<0 . 05 were considered to represent a significant difference . * , ** , *** denote p<0 . 05 , p<0 . 01 , p<0 . 001 , respectively . | Leishmania spp . are protozoan parasites infecting a variety of mammals , including humans and mice . Much information has been gleaned from murine models of Leishmania major infection . The control of L . major infection by resistant C57BL/6 mice requires the secretion of type 1 ( Th1 ) cytokines ( i . e . IFN-γ ) by T cells as well as the expression of inducible nitric oxide synthase ( iNOS ) by phagocytic cells . Conversely , susceptible BALB/c mice are unable to control infection and develop a type 2 ( Th2 ) immune response characterized by the secretion of IL-4 and IL-13 cytokines . In this study , we showed that the main iNOS-producing cells in the lesion and the draining lymph node are phenotypically similar to iNOS-producing “inflammatory” dendritic cells ( DC ) , which are already described in the mouse models of Listeria monocytogenes and Brucella melitensis infection . Our data also highlighted a strong association between the recruitment and activation of these inflammatory DC and the resistance to L . major infection . In addition , we showed that iNOS production by these inflammatory DC is positively regulated by Th1 response and negatively by Th2 response . Taken together , our results provide new insights into how innate and adaptive immune responses fight L . major infection . A better understanding of the mechanisms regulating inflammatory DC recruitment and activation could lead to new therapeutic strategies against Leishmania infection . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"infectious",
"diseases/protozoal",
"infections",
"immunology/immunity",
"to",
"infections",
"immunology/immune",
"response",
"immunology/innate",
"immunity"
] | 2009 | iNOS-Producing Inflammatory Dendritic Cells Constitute the Major Infected Cell Type during the Chronic Leishmania major Infection Phase of C57BL/6 Resistant Mice |
Although Bcl-XL and Bax are structurally similar , activated Bax forms large oligomers that permeabilize the outer mitochondrial membrane , thereby committing cells to apoptosis , whereas Bcl-XL inhibits this process . Two different models of Bcl-XL function have been proposed . In one , Bcl-XL binds to an activator , thereby preventing Bax activation . In the other , Bcl-XL binds directly to activated Bax . It has been difficult to sort out which interaction is important in cells , as all three proteins are present simultaneously . We examined the mechanism of Bax activation by tBid and its inhibition by Bcl-XL using full-length recombinant proteins and measuring permeabilization of liposomes and mitochondria in vitro . Our results demonstrate that Bcl-XL and Bax are functionally similar . Neither protein bound to membranes alone . However , the addition of tBid recruited molar excesses of either protein to membranes , indicating that tBid activates both pro- and antiapoptotic members of the Bcl-2 family . Bcl-XL competes with Bax for the activation of soluble , monomeric Bax through interaction with membranes , tBid , or t-Bid-activated Bax , thereby inhibiting Bax binding to membranes , oligomerization , and membrane permeabilization . Experiments in which individual interactions were abolished by mutagenesis indicate that both Bcl-XL–tBid and Bcl-XL–Bax binding contribute to the antiapoptotic function of Bcl-XL . By out-competing Bax for the interactions leading to membrane permeabilization , Bcl-XL ties up both tBid and Bax in nonproductive interactions and inhibits Bax binding to membranes . We propose that because Bcl-XL does not oligomerize it functions like a dominant-negative Bax in the membrane permeabilization process .
Apoptosis is a form of programmed cell death important for development and tissue homeostasis , and its deregulation has been implicated as the cause of many disease processes . Apoptosis may be initiated by a developmental program and by many diverse forms of cell stress . A critical feature of metazoan apoptosis is the permeabilization of intracellular organellar membranes that leads to the egress of intraorganellar components that activate the proteases responsible for cell death . Bcl-2 is the founding member of a family that includes members that either prevent ( e . g . , Bcl-2 or Bcl-XL ) or promote ( e . g . , Bax or Bak ) the membrane permeabilization that leads to apoptosis . Another large subgroup of the Bcl-2 family ( BH3-only proteins; e . g . , tBid ) initiates apoptosis through binding to Bax and/or Bcl-XL . Even though Bcl-XL and Bax are structurally similar , experiments with protein fragments and peptides or full-length protein in the absence of membranes has led to the elaboration of models in which the functional relevance of binding partners for Bcl-XL differ . The direct activation [1 , 2] or hierarchical model [3] states that Bcl-XL and other antiapoptotic Bcl-2 family proteins inhibit apoptosis primarily through interactions with a subclass of BH3-only proteins termed “activators” , preventing them from activating Bax and Bak . The indirect activation model [4 , 5] proposes that the function of all BH3-only proteins is to displace the inhibitory Bcl-XL from inherently active Bax or Bak . Thus , these two models differ primarily in whether Bax/Bak is constituively active or requires activation and which interactions are crucial to the antiapoptotic function of Bcl-XL . However , there is also considerable overlap between these two competing models: both models recognize that Bcl-XL directly binds to and inhibits a proapoptotic Bcl-2 family protein that is directly involved in membrane permeabilization , while other proapoptotic Bcl-2 family proteins indirectly induce apoptosis by binding to Bcl-XL and preventing this function . Moreover , some authors postulate that both models may be relevant in different cell types or in the same cell type under different circumstances ( e . g . , normal versus cancerous ) [2 , 6] . Our previous results have demonstrated the importance of dynamic interaction of Bcl-2 family members with membranes [7–9] . On the basis of these data and evidence from the literature , we have proposed a model termed “embedded together” , in which interaction of these proteins with each other changes after binding to the membrane as this causes conformational changes that alter and/or expose new binding surfaces [10 , 11] . Accordingly , a key feature for resolving differences in the components and mechanisms of physiologically relevant interactions between Bcl-2 family members is to examine these interactions in membranes . To examine the molecular mechanism of membrane permeabilization for full-length Bcl-XL and Bax , we used recombinant proteins and measured permeabilization of both liposomes and mitochondrial outer membranes in vitro . Recombinant tBid was used as an activator protein . Our results demonstrate that Bcl-XL and Bax , despite having opposite effects on apoptosis , share many functionally similar features: in the absence of an activator , neither protein bound tightly to membranes , whereas the addition of membrane-bound tBid to Bax or Bcl-XL recruited a similar molar excess of the soluble protein to membranes . However , after binding to membranes , only Bax formed large oligomers and permeabilized the membrane . Bcl-XL competes with Bax for binding to both tBid and membrane-bound Bax as well as for binding to membranes . In all cases , interaction of Bax with Bcl-XL is nonproductive for membrane permeabilization , presumably because Bcl-XL does not oligomerize . Thus , we propose that Bcl-XL functions similarly to a dominant-negative Bax in tBid-initiated membrane permeabilization .
The earliest step in the process leading to membrane permeabilization that could be inhibited by Bcl-XL is binding of tBid or Bax to the membrane . In incubations of 20 nM tBid and liposomes , tBid bound effectively to liposomes as assayed by gel filtration chromatography ( Figure 1B ) . Without tBid , Bax ( 100 nM ) did not bind to liposomes , as expected from results seen in vitro and in cells [20] . However , when the two proteins were added together , 20 nM tBid caused most of the Bax to bind liposomes . The stoichiometry of this interaction confirms that one tBid molecule can directly or indirectly recruit multiple Bax molecules to membranes [21] . The resulting membrane permeabilization released ANTS/DPX and larger encapsulated fluorophores ( Figure S1A ) as well as fluorescent proteins of similar mass to the proteins released from the intermembrane space of mitochondria during apoptosis ( unpublished data ) . Addition of recombinant Bcl-XL inhibited tBid/Bax-mediated liposome permeabilization in a concentration-dependent manner ( Figure 1A ) with an IC50 of ∼25 nM at the 2 h end point . Furthermore , the shape of the inhibition curve ( Figure S1B ) suggests that Bcl-XL inhibits ANTS/DPX release competitively . At all effective concentrations of Bcl-XL , membrane binding by Bax was efficiently inhibited ( Figures 1B and 3 ) , as has been observed previously in cells [22] and for mitochondria [18] . While Bax membrane binding was inhibited , Bcl-XL had no effect on tBid binding to membranes ( Figure 1B ) . These results argue that Bcl-XL inhibits membrane permeabilization by preventing the binding of Bax , but not tBid , to membranes . To confirm that Bcl-XL inhibited tBid-induced binding of Bax to biological membranes , mitochondria purified from mouse liver ( MLM ) of bak –/– mice were assayed . MLM from bak –/– mice also lack Bax as it is a cytoplasmic protein in liver . Like liposomes , these mitochondria were resistant to tBid , as outer mitochondrial membrane ( OMM ) permeabilization , measured by the release of cytochrome c , required the addition of both tBid and Bax ( Figure 1C ) . As seen with liposomes , Bcl-XL inhibited tBid/Bax-mediated cytochrome c release . To assess the binding of recombinant Bax to mitochondria , the organelles were incubated with or without tBid or tBid and Bcl-XL , pelleted by centrifugation , and analyzed by immunoblotting ( Figure 1D ) . The addition of tBid increased the level of membrane-bound Bax and caused the integration of Bax into the OMM , as assessed by resistance to carbonate extraction , while the addition of Bcl-XL inhibited these effects , confirming our results obtained with liposomes . As expected , tBid was sufficient to induce permeabilization of OMM isolated from wild-type mice ( as these mitochondria contained Bak ) , and Bcl-XL inhibited this permeabilization ( unpublished data ) . The correlation between tBid concentration ( 20 nM ) and the IC50 of ∼25 nM Bcl-XL for inhibition of liposome permeabilization is consistent with other published models suggesting Bcl-XL sequestration of Bax/Bak activators ( in this case tBid ) as one mechanism by which Bcl-XL could inhibit membrane binding by Bax . Alone , Bcl-XL showed minimal liposome binding , consistent with cytoplasmic or loosely membrane-bound localizations reported for Bcl-XL in live cells [23 , 24] ( Figure 1B ) . However , 20 nM tBid caused migration of ∼80 nM Bcl-XL to liposomes both when the two proteins were added together ( Figure 1B , quantified in Figure 4B ) and when tBid was bound to liposomes before Bcl-XL was added ( unpublished data ) . The stoichiometry of this interaction indicates that , similar to the effect on Bax , one tBid molecule recruits multiple Bcl-XL molecules . When MLM from bak –/– mice were used as the membrane source , the addition of tBid markedly increased the levels of membrane-integrated Bcl-XL . As a result , the Bcl-XL , which was almost entirely peripherally attached in the absence of tBid , integrated in the OMM in the presence of tBid ( Figure 1E ) . When both proteins were in the membranes , Bcl-XL bound to tBid , as assessed by co-immunoprecipitation . This interaction was not dependent on the detergent used to solubilize the liposomes ( unpublished data ) , and at concentrations of 100 nM Bcl-XL and 20 nM tBid the interaction was not affected by the addition of Bax ( Figure 2A , left panel , lanes 1 and 2 ) . In the absence of membranes , interaction between Bcl-XL and tBid could not be detected by co-immunoprecipitation ( unpublished data ) . Therefore , sequestering membrane-bound tBid represents one mechanism whereby Bcl-XL could inhibit apoptosis ( Figure 6C , step 4 ) . The recruitment of multiple Bcl-XL molecules by a single molecule of membrane-bound tBid increases the likelihood that tBid will be sequestered by Bcl-XL , thereby increasing how effectively Bcl-XL competes with Bax for tBid binding . Furthermore , when Bcl-XL prevented tBid/Bax-mediated membrane permeabilization ( Figure 1A ) , Bcl-XL was overwhelmingly membrane-bound ( Figure 1B ) . Thus , contrary to models that propose that membrane integration would inactivate Bcl-XL [25] , our results suggest that tBid is required to cause a conformational change in Bcl-XL that allows it to insert into membranes where it may inhibit tBid ( Figure 6B ) . Although sequestration of membrane-bound tBid by Bcl-XL appears to account for its antiapoptotic function in both liposomes and MLM , we sought to determine whether Bcl-XL also could interact stably with Bax . When tested in the absence of tBid , a stable interaction between Bax and Bcl-XL was not detected ( Figure 2A , left panel , lane 3 ) , suggesting that Bcl-XL does not sequester Bax in solution . However as expected , co-immunoprecipitation of Bcl-XL and Bax was observed in control experiments where membranes were solubilized with the nonionic detergent NP-40 , known to induce a conformational change in Bax required for heterodimerization with Bcl-XL that is also seen in cells when apoptosis is induced [20 , 26] . When tested in the presence of tBid ( 20 nM ) and membranes , we did not detect an interaction between Bcl-XL ( 100 nM ) and Bax ( 100 nM ) ( Figure 2A , left panel , lane 2 ) . Rather , Bcl-XL out-competed Bax for binding to both tBid and membranes . In these incubations , Bcl-XL may not bind Bax because they are not in the same compartment; Bcl-XL is almost completely membrane-bound while Bax is soluble ( Figure 1B ) . Therefore , the same experiment was performed using 20 nM Bcl-XL , a concentration at which Bcl-XL competes less efficiently with Bax for binding to tBid , allowing some Bax to become membrane-bound . Under these conditions , Bcl-XL marginally inhibited membrane permeabilization ( Figure 1A ) and coprecipitated with Bax in CHAPS buffer ( Figure 2A , right panel , lane 2 ) . The addition of Bax also reduced the amount of tBid bound to Bcl-XL ( compare lanes 1 and 2 ) , suggesting that under conditions where Bcl-XL cannot fully out-compete Bax for access to tBid the presence of membrane-bound Bax provides a second target for binding by Bcl-XL ( Figure 6C , step 5 ) . Taken together , our results indicate that Bcl-XL and Bax compete for binding to tBid and that Bcl-XL prevents membrane permeabilization by forming stable heterodimers with the membrane-bound forms of both tBid and Bax . However , because both interactions are present when Bcl-XL is active , it is possible that one particular interaction contributes more to the antiapoptotic function of Bcl-XL than the other , as has been proposed alternately by two competing models of apoptosis [3 , 4] . To assess the relative contributions of Bcl-XL binding to tBid and Bax to antiapoptotic function , these interactions were selectively removed through mutagenesis . A tBid mutant with substitution of two residues within the BH3 domain ( M97A/D98A of murine Bid , denoted tBid-mt1 ) does not bind stably to Bcl-XL but binds to Bax [27] . Consistent with this previous report , in our assay system tBid-mt1efficiently recruited Bax to membranes where Bax was activated ( unpublished data ) , but tBid-mt1 did not coprecipitate with Bcl-XL ( Figure 2B ) . Thus , tBid-mt1 allows the selective removal of the Bcl-XL–tBid interaction . Conversely , a mutation within the BH3 binding pocket of Bcl-XL ( Y101K ) prevents stable binding to Bax [28] . As expected , Bcl-XL Y101K bound tBid but did not bind to activated Bax ( Figure 2C ) , even in the presence of NP-40 detergent . Therefore , using Bcl-XL Y101K allows the selective removal of Bcl-XL–Bax binding . Combining these two mutants creates a situation where Bcl-XL does not bind stably to either tBid or Bax ( unpublished data ) . As a negative control , we used Bcl-XL with a deletion of the BH4 domain ( amino acids 4–24 ) that has been reported to remove binding to both Bax and the BH3-only protein Bad [29] . In our assay , ΔBH4 Bcl-XL did not coprecipitate tBid and bound very inefficiently to activated Bax ( Figure 2D ) . Addition of the Y101K point mutation to ΔBH4 Bcl-XL to further remove Bax binding did not reduce this residual interaction ( unpublished data ) , indicating that it is nonspecific . Therefore , these mutants behave similarly in live cells and in our cell-free system . By assay of ANTS/DPX release from liposomes for different combinations of these proteins , the residual function of Bcl-XL can be measured in the absence of stable interaction with tBid , Bax , or both . The relative effects of each mutant were tested by examining the concentration-dependent inhibition of tBid/Bax-mediated liposome permeabilization by Bcl-XL ( Figure 3A ) . When tBid was replaced by tBid-mt1 , Bcl-XL continued to inhibit membrane permeabilization and bound Bax activated by tBid-mt1 ( Figure 2B ) . This indicates that a significant amount of Bcl-XL activity does not require Bcl-XL–tBid binding , as has been suggested previously by some [30] but not other models [3 , 31 , 32] . Removal of the Bcl-XL–Bax interaction by using Bcl-XL Y101K decreased the activity of Bcl-XL somewhat ( Figure 3A ) , consistent with previously published results [33] . Nevertheless , the protein offered significant protection from tBid/Bax-induced dye release from liposomes . Thus , the loss of tBid or Bax binding to Bcl-XL can be compensated by the other binding interaction . The combination of tBid-mt1 and Bcl-XL Y101K or ΔBH4 Bcl-XL ( situations in which Bcl-XL binds stably to neither tBid nor Bax ) greatly reduced the activity of Bcl-XL but did not eliminate it completely . To address the possibility that this remaining activity was the result of residual Bcl-XL–tBid or Bcl-XL–Bax interactions at elevated concentrations of Bcl-XL , we examined the effects of the tBid-mt1 and Bcl-XL Y101K mutations on Bcl-XL–tBid and Bcl-XL–Bax interactions by co-immonoprecipitation at Bcl-XL concentrations higher than those shown in Figure 2 ( Figure S2 ) . To detect the interactions of Bcl-XL with tBid-mt1 as well as Bcl-XL Y101K with Bax required elevated concentrations of Bcl-XL and/or prolonged immunoblot exposures compared to the wild-type proteins , suggesting that these residual interactions either are nonspecific or contribute very little to the remaining antiapoptotic activity seen with the tBid-mt1/Bcl-XL Y101K combination . To test this possibility further , we combined the tBid-mt1/Bcl-XL Y101K and ΔBH4 Bcl-XL mutations . This combination of mutations ( tBid-mt1 with ΔBH4 Bcl-XL Y101K ) did not further diminish Bcl-XL function ( unpublished data ) , suggesting that this remaining activity does not involve stable protein–protein interactions . This residual function of Bcl-XL is addressed below . The effects of these proteins also were assessed for the regulation of cytochrome c release from isolated mitochondria ( Figure 3B–D ) . By the use of bak –/– MLM ( Figure 3B , quantified in Figure 3C ) , Bcl-XL inhibited tBid/Bax-induced membrane permeabilization in a dose-dependent manner , as was seen in liposomes . Substitution of tBid-mt1 or Bcl-XL Y101K for the respective wild-type proteins only slightly reduced the prevention of cytochrome c release by Bcl-XL . Using these two mutants together markedly reduced but did not eliminate the inhibition of cytochrome c release by Bcl-XL , similar to the residual activity noted with liposome permeabilization . When wild-type MLM that contained BAK were used , it was necessary to reduce the concentration of tBid 10-fold to 25 pM to prevent it from activating sufficient Bak to cause OMM permeabilization . At this concentration of tBid , cytochrome c release required the addition of Bax ( unpublished data ) , and the activity of Bcl-XL was similar to that seen with bak –/– MLM ( Figure 3D ) . In these incubations , substitution of the wild-type proteins with tBid-mt1 , Bcl-XL Y101K , or both showed similar effects to those seen in bak –/– MLM and liposomes ( Figure 3C ) . Thus , in contrast to widely promulgated models for the antiapoptotic mechanism of Bcl-XL , our results with purified proteins in the presence of relevant membrane targets indicate that Bcl-XL binding to Bax and tBid are both functionally relevant , but neither is paramount . Although our results indicate that Bcl-XL inhibits tBid-mediated activation of Bax by sequestering tBid in a stable complex , it is unclear how Bcl-XL inhibits Bax binding to membranes . For example , Bcl-XL might inhibit Bax binding to membranes by preventing it from interacting with tBid . Alternatively or in addition , Bcl-XL might directly inhibit Bax binding to membranes . To determine the mechanism ( s ) involved , we measured Bax liposome binding by gel filtration chromatography for reactions containing the different mutant proteins ( Figure 4A ) . The absence of a stable interaction between tBid-mt1 and Bcl-XL significantly decreased Bcl-XL-mediated inhibition of Bax binding to membranes at all tBid concentrations assayed . Consequently , significantly more Bax bound to membranes in reactions containing tBid-mt1 compared to otherwise identical reactions containing tBid ( p < 0 . 005 for 100 nM Bcl-XL , p < 0 . 05 for 40 nM Bcl-XL , p < 0 . 1 for 20 nM Bcl-XL ) . However , contrary to the prediction of some models [34] , Bcl-XL sequestering of tBid is only a contributing factor , as in the absence of tBid binding Bcl-XL still dramatically reduced Bax binding to membranes ( Figure 4A , compare lanes 5–8 ) . It has been suggested that the multi-BH-region proapoptotic proteins Bax and Bak autoactivate after tBid ( or another BH3-only protein ) initiates the process and that autoactivation is inhibited by Bcl-2 [21 , 35] . In our system , one consequence of Bax autoactivation would be recruitment of soluble Bax to membranes . Therefore , in the absence of Bcl-XL/tBid heterodimerization , Bcl-XL may inhibit recruitment of Bax by binding to membrane-bound Bax and inhibiting Bax autoactivation . To address this possibility , we substituted Bcl-XL with the Bax-binding-deficient Bcl-XL Y101K and assayed Bax binding to liposomes in the presence of tBid-mt1 ( Figure 4A , lanes 9–12 ) . In reactions containing Bcl-XL Y101K , significantly more Bax bound membranes ( p < 0 . 005 for 100 nM Bcl-XL , p < 0 . 05 for 40 nM Bcl-XL , p < 0 . 1 for 20 nM Bcl-XL ) , indicating that the interaction of Bcl-XL and Bax inhibits further Bax binding to membranes . To determine whether the Bcl-XL/Bax heterodimer also prevented the subsequent oligomerization of Bax , we examined oligomerization by cross-linking . In these experiments , the cross-linker was added to reactions containing an equal amount of membrane-bound Bax in the absence or presence of Bcl-XL ( Figure S3 ) . In these reactions , membrane-bound Bcl-XL inhibited Bax oligomerization , as detected by cross-linking concomitant with inhibition of dye release from liposomes . Taken together , these results suggest that , when bound to Bcl-XL , Bax function is neutralized , both in recruitment of other Bax molecules through autoactivation and in oligomerization to permeabilize membranes ( Figure 6C , step 6 ) . Similar to results obtained by examining liposome permeabilization , there is a residual activity of the tBid-mt1/Bcl-XL Y101K combination that prevented Bax binding to membranes even in the absence of a stable interaction of Bcl-XL Y101K with either tBid or Bax . This activity also was observed for a ΔBH4 mutant of Bcl-XL ( Figure 4A , lanes 13 , 14 ) . At the onset of apoptosis in cells , Bcl-XL binds to membranes [23] . The addition of tBid is sufficient to trigger Bcl-XL to bind to membranes in vitro ( Figure 1B ) . Because the membrane appears to be the active locus , we used the mutant versions of Bcl-XL and tBid to examine the importance of stable binding to tBid or Bax for Bcl-XL to bind tightly to membranes ( Figure 4B ) . In a negative control experiment without other added proteins , less than 10 nM of the added Bcl-XL ( 100 nM ) bound to membranes ( Figure 4B , lane 1 ) . Addition of Bax to the incubation did not increase membrane binding by Bcl-XL ( lane 2 ) . As noted above ( Figure 1B ) , a substoichiometric amount of tBid ( lane 3 ) caused ∼80 nM Bcl-XL to bind to membranes , suggesting that each tBid molecule recruited three to four Bcl-XL molecules . Membrane binding by Bcl-XL was reduced greatly when tBid was substituted by tBid-mt1 , indicating that stable Bcl-XL–tBid binding can recruit Bcl-XL ( lane 5 ) . However , the addition of Bax and tBid-mt1 ( lane 6 ) caused the near complete recruitment of Bcl-XL ( ∼85 nM membrane-bound ) . Under these conditions , the concentration of membrane-bound , activated Bax is only ∼15 nM ( Figure 4A , lane 8 ) , yet the concentration of membrane-bound Bcl-XL increases by ∼60 nM above that seen with tBid-mt1 , indicating that similar to tBid each activated Bax molecule recruits about four Bcl-XL molecules . Thus , like tBid , tBid-activated Bax recruits both soluble Bax and Bcl-XL . Surprisingly , when both soluble Bax and Bcl-XL are present , it appears that activated Bax recruits Bcl-XL more efficiently than it recruits Bax . Therefore , another way that Bcl-XL inhibits recruitment of Bax is by competing with it for activated Bax on the membrane . The Bax-binding-deficient mutant Bcl-XL Y101K directly and efficiently binds liposomes ( Figure 4B , lanes 7–9 ) . This is presumably due to the location of this mutated residue in the BH3 binding pocket suggested to house the C-terminal membrane anchor of soluble Bcl-XL based on structural similarity with Bax [36] or to bind the C-terminal membrane anchor of an another Bcl-XL molecule as a cytoplasmic homodimer [33] . In either case , spontaneous binding of Bcl-XL Y101K to membranes may be due to displacement of the C-terminal membrane anchor as a result of the Y101K mutation . The functional importance of membrane binding by Bcl-XL is supported further by our observations that when Bcl-XL inhibits membrane permeability in assays containing tBid and Bax almost all of the Bcl-XL is membrane-bound and is in stoichiometric excess over membrane-bound tBid and Bax ( Figure 4A and 4B , black bars ) , even when using mutants that prevent heterodimerization with one or both of the binding partners . Consistent with a role for membrane binding , removing the C-terminal tail that mediates membrane binding impaired but did not abolish Bcl-XL function ( Figure S4 ) , similar to the results obtained in cells using the same mutant [37] . However , removal of the C-terminal tail in the context of the tBid-mt1 and Bcl-XL Y101K mutations completely abolishes the remaining activity of Bcl-XL . It is possible therefore that the antiapoptotic function of Bcl-XL that is independent of stable interactions with tBid and Bax requires that Bcl-XL is membrane-bound . In that case , the excess Bcl-XL bound to membranes in the presence of tBid or activated Bax could contribute to this function of Bcl-XL ( see below ) . Taken together , our results suggest that Bcl-XL prevents recruitment of Bax to membranes by multiple mechanisms ( Figure 6C ) . These include: competition with Bax for binding to and thereby sequestering tBid , inhibition of Bax autoactivation by competing with soluble Bax for binding membrane-bound Bax , and an unidentified mechanism requiring membrane-bound Bcl-XL . We hypothesized that the residual antiapoptotic function of membrane-bound Bcl-XL that does not require stable heterodimerization with tBid or Bax involves inhibiting the transient Bax conformational change that occurs when Bax interacts with the membrane surface and exposes an epitope bound by the 6A7 monoclonal antibody [7] . Binding of 6A7 has been used widely as a marker for one of the stages required for activation of Bax ( reviewed in [10] ) . In vivo , this conformational change has been observed for membrane-bound Bax actively involved in membrane permeabilization [38 , 39] . However , in experiments using liposomes , it is evident that the exposure of this epitope occurs prior to and is independent of Bax membrane insertion . The change is presumed to result from an interaction of Bax with the surface of the membrane because it rapidly reverses when Bax is separated from the liposomes [7] . This liposome-induced conformational change does not require tBid; however , the epitope remains exposed after tBid-induced Bax membrane insertion ( Figure 5A , lanes 1 and 2 ) . Unlike the conformational change that accompanies tBid-induced insertion of Bax into membranes , the liposome-induced conformational change also disappears if liposomes are solubilized in CHAPS prior to immunoprecipitation ( Figure 5A , compare lanes 1 and 3 ) . As shown above , addition of Bcl-XL prevents tBid-induced insertion of Bax into membranes and instead results in Bcl-XL inserting into the membrane ( Figures 1 and 4 ) . However , under these conditions the soluble Bax does not undergo the liposome-induced conformation change ( Figure 5B ) , suggesting that the presence of membrane-bound Bcl-XL either inhibits this conformational change or shifts the equilibrium of Bax molecules interconverting between the two forms sufficiently towards the 6A7-negative conformer that the antibody no longer has sufficient access to the epitope . As previously described , soluble Bcl-XL had little effect on the liposome-induced conformational change of Bax ( Figure 5C , top panel ) [7] . In contrast , the spontaneous membrane binding Bcl-XL Y101K mutant inhibits this conformational change in Bax ( Figure 5C , bottom panel ) . To investigate further the effects of the membrane surface on Bax and the inhibition of these effects by Bcl-XL , cross-linking experiments using disuccinimidyl suberate ( DSS ) were performed . Cross-linking of Bax into higher-order structures after Bax binds to membranes has been observed previously [18] . The incubation of Bax with liposomes alone does not cause sufficiently tight membrane binding by Bax to survive gel filtration chromatography ( Figure 1B ) . Nevertheless , incubation with liposomes did result in the cross-linking of Bax into higher-order complexes ( Figure 5D , left panel ) . As expected from previous results [7 , 9] , the interactions between Bax monomers induced by incubation with membranes were not resistant to detergent solubilization prior to cross-linking . The Bax–Bax cross-links were reduced in the absence of liposomes ( Figure 5E ) , suggesting that , similar to binding by the 6A7 antibody , they result from a liposome-induced conformational change in Bax . Addition of tBid to Bax and liposomes resulted in a similar cross-linking pattern , but these Bax oligomers were resistant to solubilization of the membrane with detergent ( Figure 5D , middle panel ) . Membrane-bound Bcl-XL not only prevented the formation of detergent-resistant Bax cross-links but also prevented the cross-linking of Bax that resulted when Bax contacted the membrane surface ( Figure 5D , right panel ) . These results further demonstrate that Bcl-XL in the membrane prevents or reverses the conformational change in soluble Bax that occurs upon exposure to a membrane surface . To determine whether the effect of membrane-bound Bcl-XL on this transient , liposome-induced conformational change in Bax could be regulated , BH3 peptides were used to induce selectively Bcl-XL binding to membranes ( Text S1 ) . On the basis of the effects previously published for mutations in BH3 domains of proapoptotic proteins and peptides , we selected two peptides that both caused membrane binding by Bcl-XL ( Figure S5A ) but differed in their functional effects ( Figure S5B ) . One peptide , designated m1Bid BH3 , containing a single mutation for a conserved and critical leucine residue , did not interfere with the antiapoptotic function of Bcl-XL . The other peptide , Bak BH3 , effectively eliminated the antiapoptotic activity of Bcl-XL as assayed by liposome permeabilization assays . The effects of these peptides indicate that membrane-bound Bcl-XL can exist in both functional and nonfunctional states . To determine whether the functional state of membrane-bound Bcl-XL affected the inhibition of the liposome-induced conformational change in Bax , the m1Bid BH3 and Bak BH3 peptides were used to trigger membrane binding by Bcl-XL in the presence of soluble Bax . In the absence of Bcl-XL , neither peptide induced Bax membrane binding or Bax-dependent membrane permeabilization ( unpublished data ) . When the m1Bid BH3 peptide triggered Bcl-XL binding to membranes , the Bcl-XL still inhibited the liposome-induced Bax conformational change ( Figure 5F , top panel , lanes 2 and 3 ) . In contrast , when the Bak BH3 peptide was added , membrane-bound Bcl-XL did not prevent the liposome-induced Bax conformational change ( Figure 5F , bottom panel , lanes 2 and 3 ) . Therefore , inhibition of the liposome-induced conformational change in soluble Bax by membrane-bound Bcl-XL correlates with the functional status of Bcl-XL on the membrane . Taken together , these results indicate that this activity is a regulatable function of membrane-bound Bcl-XL and is not merely the result of changes in the biophysical properties of liposomes after Bcl-XL binding ( Figure 6D ) .
To date , it has been a paradox that Bcl-XL and Bax have very similar structures , yet proposed models for their function all suggest that they behave very differently . Taken together , our results suggest that they actually function similarly to the extent that they compete with each other at most steps of the process but with the major difference being that Bcl-XL is defective for membrane permeabilization . Thus , for most functions Bcl-XL behaves in a manner conceptually similar to a dominant-negative Bax . Bcl-XL inhibited membrane binding by Bax by competing with soluble Bax for recruitment to membranes by either tBid or membrane-bound Bax ( Figure 6 ) . Thus , similar to Bax , Bcl-XL binds to both activated , membrane-bound Bax and membrane-bound tBid . Furthermore , Bcl-XL binding to Bax inhibits the subsequent oligomerization of membrane-bound Bax . Our results with mutations that disable either of these interactions individually indicate that in membranes both are functionally important . Pioneering experiments to identify relevant binding partners for Bcl-2 and Bcl-XL using immunoprecipitation in transfected cells suggested a lack of correlation between Bax binding and inhibition of apoptosis , as only certain Bcl-XL point mutants that could no longer bind to Bax lost function [40] . Further analysis of two of these Bcl-XL mutants that did not bind to Bax ( the F131V/D133A mutant that remained functional and the inactive G138E/R139L/I140N mutant ) suggested that prevention of apoptosis required binding to the BH3-only proteins tBid , Bim , and Bad , a function that was specifically lost in the latter mutant [19] . However , as these experiments were conducted on whole cells where Bax was also present , it is possible that the lack of function of this mutant is caused by the loss of binding to both BH3-only proteins ( e . g . , tBid ) and Bax , a result entirely consistent with our observations . Conversely , early work with the M97A/D98A Bid mutant that does not bind to Bcl-XL [27 , 30] that we have used in our study ( tBid mt1 ) indicated that Bcl-XL inhibited the apoptosis caused by this Bid mutant , in cells and in purified mitochondria , implying that the interaction with activated Bax rather than Bid was critical to the antiapoptotic function of Bcl-XL in this context . Our model reconciles these disparate results by postulating that only after loss of both interactions does the antiapoptotic function of Bcl-XL become severely diminished . While Bcl-XL may inhibit OMM permeabilization initiated by the BH3-only proteins Bid , Bim , and Puma by sequestering these proteins , a variety of evidence now exists for alternate Bax activation pathways ( reviewed in [10] ) that do not rely on these BH3-only proteins . For many of these known and potentially unknown pathways , Bcl-XL may not directly sequester the initial activators of Bax , and hence inhibition of membrane permeabilization may rely on the sequestration of activated Bax to inhibit Bax autoactivation , oligomerization , and membrane permeabilization . Our results clearly show that the major functional interactions of Bcl-XL occur after the protein has migrated to the membrane , a process initiated in our in vitro system by binding to either membrane-localized tBid or membrane-localized ( activated ) Bax . Membrane insertion of both tBid [41] and Bax [42] exposes the proapoptotic BH3 region of each protein , an event that is critical for interaction with Bcl-XL and likely initiates a conformational change in the antiapoptotic protein that is required for membrane insertion . Bcl-XL is found in both the cytoplasm as well as attached to ( but not inserted into ) the mitochondrial membrane in healthy cells . Migration of the cytoplasmic fraction to membranes with insertion occurs during apoptosis [23 , 43 , 44] for Bcl-XL as well as other antiapoptotic proteins such as Mcl-1 [44] and Bcl-w [25 , 45] . For Bcl-w , interaction with a tethered Bim BH3 peptide displaces the C-terminal insertion sequence from the BH3 binding pocket within the protein triggering insertion of the protein into the membrane [25 , 46] . Although a structure has not been reported for Bcl-XL containing the C-terminal insertion sequence , the structure of the truncated protein is sufficiently similar to that of Bcl-w [46] and proapoptotic Bax [36] to suggest displacement of the insertion sequence of Bcl-XL as a mechanism that drives Bcl-XL into the membrane . In vitro binding assays using protein lacking the C-terminal insertion sequence have shown that Bcl-XL binds a variety of BH3 peptides [2 , 34] . For several of the peptides , structural studies have confirmed that the peptides bind in the hydrophobic pocket that is also believed to bind the Bcl-XL insertion sequence [47–49] . Furthermore , it has been reported previously that overexpression of Bad caused the membrane insertion of Bcl-XL in HeLa cells [33] . We therefore determined whether peptides from the BH3 regions of Bid , Bim , Bad , Bax , and Bak caused insertion of Bcl-XL into liposomes , assayed by flotation on a sucrose gradient ( Figure S6 ) . All five peptides caused Bcl-XL to insert into membranes , while a mutant Bid peptide that fails to bind Bcl-XL [2] did not . In cells , it is likely that other BH3-only proteins will substitute in many cases for tBid , and unlike Bid some of these proteins may not be required themselves to bind to membranes to expose the BH3 sequence . Nevertheless , whether initiated in the cytoplasm or at the surface of the membrane , interaction of Bcl-XL with a proapoptotic BH3 sequence is likely to be at least one of the factors that contributes to the insertion of Bcl-XL into membranes . Previous reports using transfected cells have indicated that the removal of the C-terminal insertion sequence of Bcl-XL severely impairs membrane insertion but has varying effects on Bcl-XL function , from moderate [29] to severe [50] reduction in function . Consistent with this variability , we have shown previously [37] that the function of ΔTM Bcl-XL compared to that of wild type is heavily dependent on the stimulus used to initiate apoptosis ( and therefore possibly the BH3 protein ( s ) involved ) . In our in vitro system , ΔTM Bcl-XL showed a severe loss of function ( Figure S3 ) as the relevant interactions here occur on membranes . However , in certain circumstances in cells ΔTM Bcl-XL may retain at least part of its antiapoptotic function in the cytoplasm depending on the location of relevant binding partners at the onset of apoptosis . While we have focused our experiments on examining tBid and its interactions with Bax and Bcl-XL , our studies do not imply that tBid is an essential component of the membrane permeabilization process . Instead , we predict that various aspects of the functions of Bcl-XL and Bax revealed here can be triggered differentially by the different BH3 proteins that have been identified . Moreover , different BH3-only proteins would be expected to exhibit differences in the relative affinities for and activities on Bax and Bcl-XL , as we have described in our “embedded together” model [10] . However , the previously reported binding affinities between BH3 peptides and other Bcl-2 family members are unlikely to accurately reflect binding affinities at the physiologically relevant locus of membranes . Clearly , it will be important to determine more appropriate quantitative estimates for the interactions of Bcl-2 family proteins in membranes . Inhibition of the transient conformation change in Bax that occurs at the membrane surface ( Figure 6D ) is the only step in which inhibition of Bax activation is not due to Bcl-XL functioning similar to a defective version of Bax . It is likely that the conformational change in Bax at the membrane surface enhances its activation by membrane-bound tBid . Consistent with this interpretation , tethering of the soluble Bid BH3 peptide to liposomes increased its potency by several orders of magnitude [51] . It is not known how membrane-bound Bcl-XL prevents a transient change in the conformation of Bax . However , it is not unreasonable to speculate that it might do so by interacting transiently with Bax to shift the equilibrium in Bax conformers sufficiently toward the 6A7-negative conformation that the antibody does not have sufficient access to the epitope to bind it . Our in vitro system recapitulates the core features of organelle permeabilization by activated Bax and its inhibition by Bcl-XL and has allowed us to identify and examine many of the individual steps . The liposomes that we used to model physiologic membranes have a high intrinsic curvature and lipid composition that facilitate membrane binding of the recombinant proteins and induction of the 6A7 conformational change in Bax . In cells , these feature are likely represented by complex and dynamic physiologic processes such as mitochondrial membrane fission and fusion shown to be important in apoptosis [52–54] and the interaction of mitochondria with other membrane systems [39 , 52 , 55] . The major observations reported here were confirmed for cytochrome c release from mitochondria ( Figure 3B–D ) , suggesting that they will be relevant in live cells . Moreover , the simplicity and power of our in vitro system using recombinant proteins and membranes has allowed us to identify and measure functionally important and potentially “druggable” interactions important for the regulation of apoptosis ( Figure 6 ) .
The bak –/– mice were purchased from Jackson Laboratories . ANTS and DPX were purchased from Molecular Probes . BH3 peptides blocked at both ends ( Ac-peptide-amide ) were obtained from Dalton Chemicals . The monoclonal Bax antibodies 2D2 and 6A7 were generous gifts from Richard Youle [26] . The monoclonal tBid antibody 5C8 was obtained from Exalpha Biologicals . The rabbit polyclonal antibody to Bcl-XL and the sheep polyclonal antibody to cytochrome c were produced in our laboratory . Immunoblotting of Bax was carried out using 2D2 at a dilution of 1:10 , 000 . Immunoblotting of Bcl-XL , tBid , and cytochrome c was performed at dilutions of 1:10 , 000 , 1:2 , 000 , and 1:5 , 000 , respectively , with the appropriate antibody . Secondary antibodies conjugated to horseradish peroxidase were purchased from Jackson Immuno Research Laboratories and were used at dilutions of 1:10 , 000 . Immunoblots were analyzed using ImageQuant ( version 5 . 2 , Molecular Dynamics ) . Statistical analysis was performed using a one-way ANOVA model . All lipids were obtained from Avanti Polar Lipids . DSS was purchased from Pierce . Recombinant full-length human Bcl-XL ( or Bcl-XL Y101K ) with no additional amino acids was expressed in Escherichia coli as a C-terminal intein/chitin-binding domain fusion and purified by affinity chromatography on a chitin column followed by further purification on a phenyl-Sepharose column , similar to a method described previously [7] but with a final dialysis step to remove detergents . For ΔTM Bcl-XL ( and the ΔTM Y101K mutant ) , the phenyl-Sepharose chromatography step was omitted . Recombinant full-length human Bax and murine tBid ( or tBid-mt1 ) with no additional amino acids were purified as described previously [7 , 9] . Liposomes were composed of the following molar percentages of lipids: phosphatidylcholine , 48%; phosphatidylethanolamine , 28%; phosphatidylinositol , 10%; dioleoyl phosphatidylserine , 10%; and tetraoleoyl cardiolipin , 4% . Liposome preparation , including ANTS/DPX charged liposomes , was essentially as in [7] , with the exception that liposomes were prepared in assay buffer ( 10 mM HEPES ( pH 7 ) , 200 mM KCl , 5 mM MgCl2 , and 0 . 2 mM EDTA ) . Samples ( 50 μM total lipids ) were prepared in assay buffer with all sample components ( buffers , liposomes , etc . ) added prior to the addition of recombinant proteins . Fluorescence ( λex = 355 nm and λem = 520 nM ) was measured for 30 min at 37 °C in the presence of Bcl-XL but in the absence of Bax and tBid to obtain background values ( F0 ) . Bax and tBid ( in that order ) were added at t = 0 , and fluorescence was measured for 2 h at 37 °C . Triton X-100 was added to a final concentration of 0 . 2% ( w/v ) , and fluorescence was measured for 10 min at 37 °C ( F100 ) . The percentage release of ANTS/DPX was calculated as percentage release = ( ( F – F0 ) / ( F100 – F0 ) ) × 100 . Membrane fractions containing mitochondria from C57bl6 mouse liver were purified as described previously [56] . Samples were diluted to 1 mg/ml total protein ( Bradford assay ) , incubated with purified proteins , added in the order Bcl-XL , Bax , and tBid , for 1 h at 30 °C , then centrifuged at 13 , 000 g for 10 min . The supernatant and pellet were separated and analyzed . Samples containing liposomes ( 300 μM total lipids ) were prepared in assay buffer and incubated at 37 °C for 2 h . All sample components ( buffers , liposomes , etc . ) were added prior to the addition of recombinant proteins , which were added in the order Bcl-XL , Bax , and tBid . Membrane-bound protein was separated from soluble ( “free” ) protein using gel filtration chromatography on Sepharose CL-2B resin . Membrane binding was measured by comparing the intensities of membrane-bound proteins ( fractions 3 and 4 ) with total proteins ( fractions 3 and 4 plus fractions 8–11 ) . Separation of membrane-bound protein from soluble ( “free” ) protein by liposome floatation on a sucrose density gradient was performed as previously described [57] . To assess the membrane binding and membrane insertion of Bax and Bcl-XL into mitochondria , purified proteins , added in the order Bcl-XL , Bax , and tBid , were incubated for 1 h at 30 °C with mitochondria ( 5 mg/ml ) from bak –/– mice . Mitochondria were centrifuged at 7 , 000 g for 10 min . Pellets were resuspended ( 50 μl ) and treated with 800 μl of carbonate buffer ( 200 mM sodium carbonate ( pH 11 . 5 ) , 10 mM DTT , and 2% glycerol ) for 30 min on ice . Samples ( 750 μl ) were then overlayed onto a 0 . 5 M sucrose cushion in carbonate buffer ( 250 μl ) and centrifuged at ∼200 , 000g for 30 min . The supernatant and sucrose cushion were separated , neutralized with glacial acetic acid ( 5 μl per 250 μl sample ) , and stored at −80 °C , or the proteins were precipitated with trichloroacetic acid ( 18 . 75% ) . All immunoblots are representative of at least three independent experiments . Samples containing liposomes ( 300 μM total lipids ) were prepared and incubated as described in the “Membrane binding assays” section . Immunoprecipitation of Bcl-XL was performed using the polyclonal Bcl-XL antibody in assay buffer containing either 2% CHAPS or 0 . 2% NP-40 . Immunoprecipitates were collected as previously described [7] and washed three times in assay buffer containing the appropriate detergent . Immunoprecipitation using the conformation-specific 6A7 Bax antibody was performed on whole membranes and washed three times with assay buffer containing 2% CHAPS . All immunoblots are representative of at least three independent experiments . Samples containing liposomes ( 300 μM total lipids ) were prepared and incubated as described in the “Membrane binding assays” section . DSS cross-linking was performed at a concentration of 2 mM ( or DMSO control ) for 30 min at room temperature . The cross-linker was quenched by the addition of Tris-Cl ( pH 8 ) to a final concentration of 20 mM . To examine the effects of CHAPS solubilization prior to cross-linking , samples were mixed with an equal volume of assay buffer or assay buffer containing 4% CHAPS , and the extent of cross-linking was analyzed . All immunoblots are representative of at least three independent experiments . | During development and under stress , cells can become committed to die via programmed cell death ( apoptosis ) . In most cases , the permeabilization of the outer mitochondrial membrane is a key component of this commitment . The membrane permeablization step is both positively and negatively regulated by members of the Bcl-2 family of proteins . One member of this protein family with only a BH3 region , such as tBid , activates another family member , Bax , causing it to form large complexes that generate membrane-spanning pores , hence making the membrane permeable . Antiapoptotic members of the Bcl-2 family , such as Bcl-XL , are structurally similar to Bax but inhibit the membrane permeabilization process by an unknown mechanism . Two mutually exclusive models have been proposed to explain how the Bcl-2 family is operating: one states that Bcl-XL binds to tBid , thereby preventing Bax activation , while the second suggests that Bcl-XL binds directly to activated Bax . It has been difficult to sort out which interaction is important in cells , where multiple members of all three protein families are present simultaneously . Here , we describe an in vitro system containing the three recombinant proteins and the use of mutagenesis to selectively remove individual interactions . We show that Bcl-XL inhibits Bax by competing with it for binding to membranes , tBid , and activated Bax . Because Bcl-XL does not form pores , it inhibits apoptosis by acting as if it is a dominant-negative version of Bax . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"biochemistry"
] | 2008 | Bcl-XL Inhibits Membrane Permeabilization by Competing with Bax |
Structural classifications of proteins assume the existence of the fold , which is an intrinsic equivalence class of protein domains . Here , we test in which conditions such an equivalence class is compatible with objective similarity measures . We base our analysis on the transitive property of the equivalence relationship , requiring that similarity of A with B and B with C implies that A and C are also similar . Divergent gene evolution leads us to expect that the transitive property should approximately hold . However , if protein domains are a combination of recurrent short polypeptide fragments , as proposed by several authors , then similarity of partial fragments may violate the transitive property , favouring the continuous view of the protein structure space . We propose a measure to quantify the violations of the transitive property when a clustering algorithm joins elements into clusters , and we find out that such violations present a well defined and detectable cross-over point , from an approximately transitive regime at high structure similarity to a regime with large transitivity violations and large differences in length at low similarity . We argue that protein structure space is discrete and hierarchic classification is justified up to this cross-over point , whereas at lower similarities the structure space is continuous and it should be represented as a network . We have tested the qualitative behaviour of this measure , varying all the choices involved in the automatic classification procedure , i . e . , domain decomposition , alignment algorithm , similarity score , and clustering algorithm , and we have found out that this behaviour is quite robust . The final classification depends on the chosen algorithms . We used the values of the clustering coefficient and the transitivity violations to select the optimal choices among those that we tested . Interestingly , this criterion also favours the agreement between automatic and expert classifications . As a domain set , we have selected a consensus set of 2 , 890 domains decomposed very similarly in SCOP and CATH . As an alignment algorithm , we used a global version of MAMMOTH developed in our group , which is both rapid and accurate . As a similarity measure , we used the size-normalized contact overlap , and as a clustering algorithm , we used average linkage . The resulting automatic classification at the cross-over point was more consistent than expert ones with respect to the structure similarity measure , with 86% of the clusters corresponding to subsets of either SCOP or CATH superfamilies and fewer than 5% containing domains in distinct folds according to both SCOP and CATH . Almost 15% of SCOP superfamilies and 10% of CATH superfamilies were split , consistent with the notion of fold change in protein evolution . These results were qualitatively robust for all choices that we tested , although we did not try to use alignment algorithms developed by other groups . Folds defined in SCOP and CATH would be completely joined in the regime of large transitivity violations where clustering is more arbitrary . Consistently , the agreement between SCOP and CATH at fold level was lower than their agreement with the automatic classification obtained using as a clustering algorithm , respectively , average linkage ( for SCOP ) or single linkage ( for CATH ) . The networks representing significant evolutionary and structural relationships between clusters beyond the cross-over point may allow us to perform evolutionary , structural , or functional analyses beyond the limits of classification schemes . These networks and the underlying clusters are available at http://ub . cbm . uam . es/research/ProtNet . php
Some of the above difficulties are related with the very essence of protein classification schemes , which assume that it exists an intrinsic level of structure similarity for defining equivalence classes of protein structures . In SCOP , such an equivalence class is called fold [20] . Two proteins are defined to belong to the same fold if they share “the same major number and direction of secondary structures with a same connectivity” [4] . In CATH , the corresponding classification level is called topology , defined as “the overall shape and connectivity of secondary structures” [5] . These apparently clear definitions are in practice subject to substantial arbitrariety , first because it is not always clear which secondary structure elements belong to the structural core defining the fold and which ones are regarded as optional “embellishments” , and second because one has to allow a certain extent of structural divergence in the protein core . The difficulties presented above have led several authors to propose that the space of protein structures is continous [13] , [21] , [22] . This view is supported by the studies that underline the importance of substructures below the level of the globular domain , such as the autonomously folding units of Tsai et al [23] , the loops of standard size ( approximately 30 residues ) of Berezowski and Trifunov [24] , or the recurrent fragments of Tendulkar et al . [25] and Szustakowski et al . [26] . Expanding an old idea by Ohno [27] , Lupas et al . [28] proposed that the most ancient folds have arisen through an evolutionary process consisting in assembling polypeptide fragments together . These and similar ideas have suggest that the basic unit of protein classification should be substructures below the domain level , defined by Shindyalov and Bourne [22] as “highly repetitive near-contiguous pieces of polypeptide chain that occur frequently” in a set of non-redundant protein structures . If protein domains can be regarded as a combination of such substructures , the resulting structure space should be seen as continuous rather than discrete . A similar spirit is present in the approaches of Efimov [29] and in particular Taylor , who proposed to enumerate in a kind of periodic table all possible arrangements of secondary structure elements compatible with simple stability rules [30] , consistent with the view that evolution of protein structures proceeds by combining simpler modules , resulting in a continuous structure space . Another basic assumption of CATH and SCOP is that evolutionary relationships at the superfamily level imply structure similarity at the fold level . Although this assumption is most of the times correct , it was observed already in Ref . [31] that sequence divergence beyond ≈40% identity sometimes implies large structural variations . Grishin [32] , [33] has monitored several examples in which proteins belonging to the same superfamily diverged to the point where they do not share a common fold under the loose definition given above . Interestingly , many of these fold changes take place together with insertions or deletions of large polypeptide fragments , although an interesting example of secondary structure switching has been reported between two homologues regions of distant related proteins [34] , [35] . Viksna and Gilbert [36] recently quantified these fold changes in protein evolution , finding that some of them are relatively common . The occurrence of fold change implies that the classification level based on evolution , as the superfamily , and the classification based on structure , as the fold , should not be necessarily consistent , as already recognized by the group of Orengo [14] .
Given the above , one can ask whether protein classifications entirely based on a quantitative measure of structure similarity are possible at all , and if so to which extent . In formal terms , a protein fold is an equivalence class of protein structures . Mathematically , an equivalence relationship must possess the three property of symmetry , reflexivity and transitivity . Whereas symmetry and reflexivity are automatically fulfilled by any relationship based on a similarity measure , transitivity is not . For transitivity to hold , every time that is similar to and is similar to , then must also be similar to . In other words , you can not make a big step from to by making an intermediate small step through . Note that transitivity is not the same as the familiar triangular inequality , , which characterizes similarity measures obtained from a properly defined distance . Rather , transitivity is guaranteed by the much stronger property of ultrametricity [37] , , i . e . , the distance travelled in two steps can not be larger than the longer of the two steps . An ultrametric set can be uniquely classified in the form of a tree . If is similar to both and but and are not similar , there is no classification simultaneously compatible with all the pairwise similarity relationships . Borrowing a term from statistical physics , we can say that the classification problem is frustrated [38] when transitivity is violated . We expect that , if this situation is common for many triplets , there is an exponentially large number of substantially different classifications that are almost optimal , in the sense that they violate a small and similar number of pairwise relationships . Conversely , if the transitive property approximately holds , we expect that a well-defined unique globally optimal classification exists , and all sub-optimal classifications are very similar to it . We expect that the validity of the transitive property strongly depends on structure similarity . Domain pairs with high similarity share most of their structure , and we expect that transitivity approximately holds for them , so that at high similarity the structure space is made of discrete clusters . However , less stringent similarities may be due to partial substructures , and we expect that the transitive property will be violated , and the clustering will strongly depend on the algorithm used . We propose here a measure to quantify the violation of the transitive property at each step of a hierarchical clustering algorithm . In this way , we aim at detecting the minimum similarity at which transitivity still holds and clustering is justified . At lower similarity , the space should be regarded as continuous , and the significant similarities between clusters should be represented as a network rather than a tree . Let us consider three elements or clusters , with the convention that . Violation of the transitive property occurs if is large while is small , so that is an intermediate point between and . Therefore it is natural to define the transitivity violation of the triangle as . Such a quantity depends on the absolute scale and the offset of the similarity measure , i . e . , it is not invariant if we multiply all similarities times a scale factor or we add to them a constant . To remove this dependency , we divide times the difference between the largest and smallest similarities , , defining the transitivity violation associated to the triangle as ( 1 ) Notice that , by definition , Eq . ( 1 ) is comprised between zero and one because The maximum violation happens when while . Another way to interpret this formula is the following . Because of transitivity , only five clustering configurations of the elements , and are possible: all elements joined , all separated , two joined and the third one separated . For a threshold , we say that the link is violated if either and are joined despite ( overunification ) or and are separated despite ( oversplitting ) . For thresholds such that or there is one and only one configuration that satisfies all links . However , if , no one of the five possible configurations satisfies all links , since either and are incorrectly joined , or and are incorrectly separated . The volume in the space of the threshold parameter such that some links are violated quantifies the violation of transitivity as . On the other hand , if all elements are separated , and if all elements are joined , so that only values of such that correspond to non-trivial clustering . Therefore , Eq . ( 1 ) represents the ratio between the volume of parameter space for which transitivity is violated and the volume for which non-trivial clustering exist . Yet a third way to look at the above equation is the following . Most hierarchical clustering algorithms join at each step the two most similar clusters and and then recompute the similarity of the new cluster with any other one C . For the average linkage algorithm , we use the formula , where and are proportional to the number of elements in sets and . The error made by substituting the original similarities and with the combined one is , and it is proportional to Eq . ( 1 ) . Finally , also quantifies the violation of ultrametricity , since in an ultrametric set the two longest sides of any triangle must be equal [37] , which implies that . Eq . ( 1 ) is normalized in such a way that the value 1 corresponds to the maximum possible violation of ultrametricity , . Now let us consider the step of the clustering algorithm in which clusters and are joined . We define the transitivity violation at this step as the weighted sum of the transitivity violations for all triangles involving and : ( 2 ) where is proportional to the number of elements in cluster , and for each triangle we label as the element such that . The main result obtained in this study is the existence of a cross-over in the behavior of transitivity violations . This cross-over point determines an intrinsic condition for stopping the hierarchical clustering algorithm . We call the classification obtained at this point “automatic classification” . The results that we present here are based on a set of 2890 domains that are decomposed very similarly in the SCOP and CATH databases ( see Methods ) , so that the domain decompositions are more likely to be accurate and differences between CATH and SCOP on this set can not be attributed to their different ways of decomposing proteins into domains . We compute structure similarities using the Mammoth-mult algorithm [39] , which is one of the fastest algorithms for such a purpose and is comparable in accuracy to other state of the art algorithms [40] . The similarity measure that we use is based on the contact overlap , normalized in such a way as to eliminate the dependence on the domain size for pairs of unrelated domains , and for clustering we use the average linkage algorithm ( see Methods ) . These choices yielded the best results , as described below , and the results presented will refer to them unless otherwise stated . We plot in Figure 1 the transitivity violations as a function of the step of the clustering algorithm . For large the clusters joined are less similar and the transitivity violations increase . The plot can be divided into two regimes: an initial part with slow increase of transitivity violations at large similarity and a final part with faster increase and small similarity . The cross-over between these two regimes can be detected through a two-pieces fit ( see Methods ) . The normalized error of the fit , plotted in Figure 1 versus the trial cross-over point , allows us to detect at its minimum the optimal cross-over point , depicted as a vertical line . The classification obtained at this cross-over point is called here “automatic classification” , since the threshold similarity at which the clustering algorithm is stopped is automatically determined . We find , corresponding to joining two clusters with similarity . At the stopping point , the automatic classification has 779 clusters . In order to test the robustness of our method , we repeated the numerical experiments changing all the relevant choices: The alignment algorithm , the similarity measure and its normalization , the clustering algorithm and the set of domains . In all cases , we observed a clear cross-over in the behavior of the transitivity violations , and the cross-over point could be automatically located through our algorithm . Moreover , the cross-over point did not vary very much for different choices ( see Table 1 ) . In order to choose the best options , we measured the transitivity violations , the clustering coefficient , which is the network analogous of the transitive property ( see Methods ) , and the agreement of the automatic classification with SCOP and CATH as assessed through the weighted kappa measure , which is a normalized measure of consistency between two classifications ( see Methods ) . These measures tend to be consistent , i . e . , choices yielding larger clustering coefficient tend to yield smaller transitivity violations and larger weighted kappa as well . This justifies the use of the weighted kappa to assess the method , despite the problems that we will discuss in the following and that limit the best possible agreement between the automatic classification and SCOP or CATH . In particular , we considered the following options: 1 . As structure alignment method , we used either the multiple [39] or the pairwise [41] version of the MAMMOTH algorithm . As it has been recently assessed through an extensive test [40] , MAMMOTH multiple is of comparable accuracy to other state of the art structure alignment tools and faster than most of them , while its pairwise version is even faster , but at the expense of accuracy . Moreover , the two algorithms are based on different principles , since Mammoth pairwise optimizes the local superimpositions of heptamers whereas Mammoth-mult optimizes the global superimposition of the two structures . Nevertheless , we obtained very similar results with the two algorithms , which shows that the whole methodology is not very sensitive to the accuracy of the alignment . We used the more accurate MAMMOTH-mult algorithm as the standard option . 2 . We used several different measures of structure similarity . First , we used measures that require optimal rigid-body superimposition of the aligned residues . Such is the the percentage of structure identity ( PSI ) , which counts the percentage of aligned residues that superimpose within a given threshold after optimal rigid body superimposition . In order to examine the influence of this threshold , we used the standard value 4Å as used in the standard MAMMOTH score and the larger tolerance 6Å . We normalized the PSI either through the length of the shorter protein , Eq . ( 5 ) , which does not penalize matches that are only partial ( we refer to it as the Partial PSI ) or through the geometric mean length , Eq . ( 6 ) ( Total PSI ) . As an alternative to an arbitrary tolerance parameter we tested the TM score [42] , which uses a length dependent threshold that makes this score almost independent of the size of the aligned proteins . Second , we used the contact overlap , Eq . ( 7 ) , which does not depend neither on the optimal rigid body superimposition nor on a tolerance parameter , although it depends on the parameter used to define contacts , i . e . , interatomic interactions in the native structure . Most of the results presented here are obtained with the overlap as similarity score . In order to remove the dependence on protein length for unrelated proteins , we normalized the PSI and the overlap as in Eq . ( 8 ) . The parameters used in this expression were determined by fitting mean and standard deviation of the similarity of unrelated structures with respect to the length used to normalize the PSI , using either Gaussian statistics Eq . ( 9 ) , or extreme value statistics , Eq . ( 10 ) , as in the original Mammoth paper . The best similarity score was selected based on the value of transitivity violations and the clustering coefficient evaluated up to the automatic cross-over point ( see Methods ) . Using these criteria , the best score was the contact overlap ( see Figure S1 ) . The normalization with respect to domain size did not modify the clustering coefficient considerably . However , measures that omit the normalization yield much lower agreement with expert classifications , and their cross-over points are rather distinct , whereas all the normalized scores have almost the same cross-over points . Therefore , normalized scores were used as the standard . 3 . As clustering method , we considered average linkage ( AL ) , single linkage ( SL ) and complete linkage ( CL ) . We also used the neighbour joining algorithm ( NJ ) , finding results very similar to those with average linkage ( data not shown ) . For this comparison , we did not use the clustering coefficient , since it does not depend on the clustering algorithm . The plot of transitivity violations for the three algorithms is shown as Figure S2 , plot A . Not surprisingly , we found the best results with the average linkage algorithm , which can be interpreted as an algorithm trying to minimize the combination of oversplitting and overunification transitivity violations . The complete linkage only minimizes overunification errors , since it separates all structures that are below the similarity threshold . Its transitivity violations are only slightly larger than for the average linkage , but its weighted kappa is much smaller . The single linkage only minimizes oversplitting errors , since it joins all pairs above the similarity threshold . Correspondingly , it generates larger clusters . Its transitivity error is much larger than for complete and average linkage . Remarkably , single linkage clustering agrees much better than average linkage with the CATH classification at topology ( fold ) level . This is not surprising , since CATH uses single linkage clustering , but it is an interesting observation , since it illustrate that one basic difference between CATH and SCOP arises from their reliance on different clustering procedures . However , superfamilies agree much better with the average linkage classification for both CATH and SCOP . More important , the transitivity violation is an intrinsic criterion , not based on any reference classification , which clearly favors the average linkage algorithm ( see also the Discussion ) . 4 . As domain set , we used the consensus domains ( 2890 domains ) , the ASTRAL40 set of domains corresponding to SCOP release 1 . 63 ( 5041 domains ) , and the set of non-redundant domains at the 35 percent sequence identity threshold corresponding to CATH release 3 . 1 . 1 ( 7073 domains ) . The number of domains per fold as defined by SCOP ( 1 . 67 , 2 . 05 ) and CATH ( 1 . 64 , 2 . 30 ) increases with the size of the set , as we would expect from the fact that the cluster size is power law distributed , so that smaller samples are more likely to have smaller averages . The same happens at the level of superfamily . In contrast , the number of domains per cluster does not increase for larger samples , being 3 . 71 and 3 . 73 for SCOP domains and 3 . 71 and 3 . 09 for CATH domains . This indicates that our method tends to stop the clustering process relatively earlier for larger samples . In fact , larger samples are more likely to contain proteins that evidence transitivity violations . The plots of transitivity violations are qualitatively very similar , and are represented in Figure S2 , plot B . At each clustering step , we measure the difference between the average domain length of the two joined clusters and , ( 3 ) One can see from Figure 2 that the length difference is significantly larger after the cross-over point when transitivity violations increase faster . This observation is consistent with the intepretation that the regime of large transitivity violations takes place when the joined clusters are more likely to share only partial substructures . This behavior of the length difference is very robust with respect to changes in the clustering algorithm , similarity score , or set of domains . At the cross-over point , we find a broad distribution of the number of domains per cluster , with power-law probability density , . This result agrees with the distribution of the number of proteins predicted to belong to specific folds in various genomes , which follow power-laws [43] with exponents between 2 . 5 and 4 . 0 , approaching 2 . 5 for large genomes [44] . It also agrees very well with the automatic clustering by Dokholyan et al . [45] , who found an exponent of 2 . 5 using as similarity measure the Dali score [9] , with single linkage clustering and threshold derived from the statistical analysis of the domain similarity network . We also measured the cluster size distribution in the SCOP classification with 40 percent sequence similarity threshold to reduce redundancy , finding for folds and for superfamilies . Therefore , the exponent of the distribution of the number of domains per cluster agrees reasonably between the SCOP and the automatic classification . Nevertheless , this agreement is not an evidence of the consistency between classifications , since the same size distribution can be found also for clusters obtained from random networks with the same statistical properties [45] . The cross-over point of transitivity violations determines an intrinsic threshold beyond which protein similarity is better represented as a network rather than as a tree . Protein similarities have been previously represented as a network by other authors . Dokholyan et al . [45] generated the protein domain universe graph using as similarity measure the Z score of the structure alignment program Dali [9] . They found out that , for proper thresholds , the network is scale-free , i . e . , the number of links per node is power-law distributed . Performing single linkage clustering over this network , they obtained clusters whose size distribution is also a power-law , reminiscent of the distribution of protein domains per SCOP fold in a genome [43] , [44] . Krishnadev et al . [49] performed a similar study for the similarity graph of protein chains instead of protein domains . They also found scale-free behavior at large enough similarity threshold . They used spectral analysis of the adjacency matrix to partition the graph into clusters . In contrast to these previous approaches , the graph presented here is not a preliminary step for clustering , but it represents the significant similarity relationships for which clustering is not justified . These relationships not only allow to recover relationships present in expert classifications , such as splitted superfamilies and folds , but also allow to treat on the same ground the cross-fold relationships discussed by several authors , which go beyond expert classifications . We construct the similarity network by connecting the clusters of the automatic classification that have significant structural similarity . As the similarity threshold is decreased , more and more clusters are connected . Pairs of clusters containing structures from a superfamily splitted in the automatic classification get unified in the network . We measured the probability that a pair of domains is joined in the network as a function of the similarity threshold , distinguishing pairs of domains from the same superfamily , from the same fold , or from different folds . ( see Figure 9 ) . Only for similarities as low as , more than 90% of the domains in the same superfamily are joined . However , already for similarities the majority of the joined domains are from different folds . A reasonable threshold for significant structure similarity , mostly corresponding to pairs of different folds , seems to be between 3 and 4 . Results presented here are obtained using as threshold for significant structure similarity . A visual representation of such a network is shown in Figure 10B . One can see that almost all of the structure space is connected , but there is still some structure appearing . If we use a higher similarity threshold but still below the cross-over , such as , the resulting network contains several linear motifs clearly expressing transitivity violations , with connected to , to , to , and so on , but without direct connection between and or and . For comparison , we also show in Figure 10A the network constructed joining clusters at high similarity before the cross-over point ( ) using as threshold the cross-over similarity , . This network presents many regions with high density of links , representing clusters that have still to be joined , In the context of network analysis , the transitive property studied in this paper is analogous to the clustering coefficient ( see Methods ) . Clustering coefficient equal one means that the network is transitive , i . e . , if is connected with and is connected with , also is connected with . The high siilarity network obtained before the cross-over point has a high mean clustering coefficient equal to 0 . 69 , which decreases to 0 . 36 for the network after the cross-over . In general , as one could expect , the clustering coefficient increases with the similarity threshold ( see Figure S1 ) . However this increase is smooth , so that we can not use the clustering coefficient to detect the cross-over point . Interestingly , the network allows not only to recover similarity relationships at the superfamily and fold level that are below the threshold for clustering , but it may also help to discover new evolutionary or functional relationships that are not contained in SCOP or CATH . For instance , in a recent paper Xie and Bourne proposed a new method to detect remote evolutionary relationships based on the structure similarity of the active site [50] . Using this method , they confirm a previously proposed evolutionary relationship between SCOP superamily Phosphoenolpyruvate carboxykinase ( PCK ) and the P loop containing nucleotide triphosphate hydrolase ( NTH ) superfamily . The PCK domain 1ayl_1 used as a seed by Xie and Bourne is joined in the automatic classification with domains 1knxa2 and 1ko7a2 , which are classified in SCOP in the PCK superfamily but are classified in CATH in the NTH superfamily . The automatic classification supports the CATH classification . This cluster has a single significant structural link , with average similarity , with a cluster containing only domains classified in the NTH superfamily in both CATH and SCOP , and through this link another step connects it to many other clusters in the NTH superfamily or in the NTH fold . The relevant part of the network is represented in Figure S4 , from which it is clear that the structurally consistent clusters joined in a network give a richer evolutionary information than a unique fold . In order to complement structure information with sequence information , we constructed the network connecting clusters that have members belonging to the same superfamily . The networks based on sequence and structure similarity can be accessed at the url http://ub . cbm . uam . es/research/ProtNet . php
As for all problems for which hierarchical clustering algorithms are applied , for clustering protein structures it is of key importance to determine up to which point the clustering is justified . We propose to test the internal consistency of a clustering method based on a similarity measure by testing the transitive property , which requires that whenever is similar to and is similar to , then must be similar to . Only if the transitive property holds a hierarchical classification can be unambiguously built . If the transitive property is violated for an extensive number of triangles , hierarchical clustering is frustrated [38] , and we expect that there is a very large number of unrelated and almost optimal classifications , in each of which a similar number of similarity relationships are violated . We proposed here Eq . ( 1 ) to quantify the violations of transitivity of a group of three elements , and Eq . ( 2 ) to quantify the violation of transitivity when two clusters are joined . Transitivity violations as defined here occur either when a pair of domains is joined below the similarity threshold , or when a pair is separated above the same threshold . Another definition , common in the context of sequence comparisons , considers that transitivity is violated only when pairs are separated above threshold . This definition is motivated by the fact that significant sequence similarity demonstrates almost certainly an evolutionary relationship , whereas the lack of similarity does not exclude it . With this definition , the single linkage algorithm does not produce any transitivity violation , since it joins all pairs above threshold . In fact , the term transitivity is often used as a synonymous of single linkage clustering . Nevertheless , several reasons make the definition of transitivity adopted here more suitable in the context of structure classification . The first reason also applies to sequence comparisons , and it is based on protein modularity . If a domain is made of two fragments and , with similar to domain and similar to domain , single linkage will infer a non existing relationship between and . Indeed , for applying single linkage clustering to the triangle , one has to check whether the fragment overlap , Eq . ( 4 ) , is also significant . Secondly , single linkage joins many structures that are not significantly similar , producing clusters that are not structurally consistent . These clusters may lack a common core , as it is often found applying multiple structure alignment algorithms to SCOP and even more CATH superfamilies . For the goal of modelling , it may not be convenient to join structurally dissimilar domains in the same fold , since this would increase the likelihood of selecting wrong templates . The study of structure evolution is made more difficult when structural variation is hidden inside a very diverse cluster , whereas well defined clusters connected by links expressing evolutionary relationships may represent a better framework for the study of structure divergence . We have observed that the transitivity violations grow while the clustering algorithm joins protein domains into clusters . Interestingly , in all instances that we studied we have found a cross-over between two regimes of slow and fast increase of transitivity violations . We propose that the cross-over in transitivity violations is an intrinsic point to stop the automatic classification . Lower similarity relationships should be represented as a network rather than a tree . The method that we presented requires several arbitrary choices . In order to test its robustness , and the influence of the parameters , we have studied at least two alternatives for each of these choices . Qualitatively similar results were obtained for several similarity scores computed on two different alignments obtained with a local and a global version of the MAMMOTH algorithm . Both alignment algorithms were developed at our group . We did not test whether alignments obtained with algorithms developed by other groups , such as DALI , yield different conclusions , as they might do . In all cases that we tested , we have observed a cross-over in transitivity violations , finding that most of the clusters at the cross-over point correspond to subsets of SCOP or CATH superfamilies . However , the exact location of the cross-over point and the quality of the clustering , as assessed through the clustering coefficient and through the mean value of the transitivity violations , varies for different choices . Although we do not aim at reproducing SCOP or CATH , which we believe is impossible , we recognize that these expert classifications have important merits . It is therefore noteworthy that the highest clustering coefficients and lowest transitivity violations tend to be associated with scores that are better compatible with SCOP or CATH classifications . The first important choice is the structure alignment algorithm . Computationally , structure alignment is an NP-complete problem , and even if it were exactly solved different algorithms would differ , since they optimize different scores . We used two versions of the algorithm MAMMOTH that are quite different , since one optimizes local superimmposition of heptamers whereas the second one , MAMMOTH-mult , otpimizes the global structure superimposition , achieving alignments with better PSI and contact overlap . Despite this important difference , the results obtained with the two methods are rather similar . The similarity measure used is probably the most relevant choice , and we tried several of them . We obtained better results with the contact overlap than with measures that score the optimal spatial superimposition of the two structures , which are used in the standard MAMMOTH score . We conjecture that the contact overlap is a better measure than the PSI for clustering protein structures because of three reasons: ( 1 ) It does not assume that there is an optimal rigid body superimposition between the two structures . In doing so , it implicitly allows for flexible superimpositions , which might be better suited for detecting evolutionary relationships [51]–[54] . ( 2 ) It weights the residues in the core of the protein more than loop residues , since the former have a larger number of contacts . ( 3 ) The parameter it depends on , i . e . , the threshold at which two residues are considered in contact , has a physical meaning in terms of interatomic interactions , and it is therefore less arbitrary than the tolerance parameter of the PSI , i . e . , the threshold below which two residues are considered to be superimposed . Similarity scores based on structure superimposition typically need a tolerance threshold to decide whether two residues superimpose . We tested the TM score [42] , which uses a length dependent threshold that makes this score almost independent of the size of the aligned proteins . The results obtained with this score are very similar to those obtained with the contact overlap . In contrast , the percentage of structure identity ( PSI ) adopts a fixed tolerance threshold , usually chosen as 4Å . To study the effect of this parameter , we repeated our numerical experiments with a more tolerant threshold of 6Å . Not surprisingly , the more tolerant similarity measure makes the space more continuous , decreasing the clustering coefficient and increasing the transitivity violations . Therefore , the cross-over from the discrete to the continuous regime occurs at higher similarity , which means that protein domains are splitted into a larger number of clusters . In this case as well , the cross-over is clear and the clusters at the cross-over are mainly subsets of superfamilies . All measures , except the TM score , must be normalized in order to make them independent of the length of the aligned proteins . We implemented this through a length dependent Z score , as in the original MAMMOTH score . The drawback of the Z score is that not only it makes the similarity of unrelated proteins almost independent of length , but at the same time it reduces the similarity of related proteins with short length . In this way , the similarity of related proteins depend on their length and not on their evolutionary divergence , which makes the Z score an unsuitable measure for evolutionary analysis . This drawback does not occurr with the TM score , although this does not necessarily imply that it is a suitable measure for evolutionary analysis . Last , we have to decide which clustering algorithm we use . If we adopt the definition of transitivity proposed in the present work , the average linkage algorithm has to be preferred over both single linkage and complete linkage . In fact , average linkage reduces the combination of splitting and overunification errors , whereas single linkage only eliminates splitting errors , since it joins all pairs above the similarity threshold , and the complete linkage eliminates overunification errors , since it separates all structures that are below the similarity threshold . Interestingly , from our analysis it turns out that the main difference between SCOP and CATH is that the latter uses single linkage , while the former uses some procedure effectively similar to average linkage . As a last remark , we note that there is some analogy between our method , which uses transitivity violations to detect the point at which hierarchical clustering is not justified , and the bootstrap method that scores the significance of each cluster in a tree . Nevertheless , there are also important differences . Besides the fact that bootstrap is computationally much more cumbersome than our method , for obtaining a classification with the bootstrap method we would have to fix a threshold bootstrap probability to accept one cluster , whereas the cross-over that we obtain with our method arises in a natural way without fixing an arbitrary threshold . The existence of two regimes of transitivity violations , and the fact that the automatic classification at the cross-over point mostly consists of sets of SCOP or CATH superfamilies are the main results of this work . They are robust with respect to changes in the clustering algorithm , the similarity measure , the set of protein domains that we automatically classify , and the accuracy of the alignment algorithm . These results suggest that it is possible to automatically and objectively define an equivalence class for protein domains up to the similarity corresponding to the cross-over point . Clusters in the automatic classification are structurally more consistent than SCOP folds or CATH topologies , mainly because of two reasons . ( 1 ) In the automatic classification , almost 15 percent of superfamilies are split into structurally divergent clusters , indicating that there can be important structural changes in protein evolution [32] , [33] , [36] . Interestingly , domains in split superfamilies tend to have larger size difference between each other , suggesting that insertions and deletions play an important role for structural divergence , consistent with recent analysis [55] , [56] . ( 2 ) Only 44 percent of the pairs of domains in different SCOP superfamilies and the same SCOP fold are joined in the automatic classification . This percentage becomes much smaller for CATH ( less than 11 percent ) , whereas 68 and 66 percent of the pairs in the same SCOP or CATH superfamily are joined in the automatic classification The similarity between most of the pairs that are not joined is significant , but it is at the level where transitivity violations are large and a network fits the data better than a classification . Our analysis thus suggests that CATH and SCOP classify proteins up to similarities that are below the cross-over of transitivity violations . The same is possibly true for the automatic FSSP classification as well , where proteins are classified in the same fold if the Z score of their similarity is above 2 . This is the smallest threshold at which the structures compared are significantly related . Here we also use a Z score , but we find that the cross-over point is at implying that the transitive property is severely violated at the similarity level . An indication that the fold defined in expert classification may not correspond to an intrinsic similarity level is that CATH and SCOP neatly agree at the level of superfamily , as assessed through the weighted kappa measure , but they disagree between each other at the level of fold even more than they disagree with the automatic classification , when the proper clustering algorithm is used . Indeed , the main difference between SCOP and CATH at fold level is that SCOP uses a procedure effectively similar to the average linkage algorithm , whereas CATH uses the single linkage algorithm , which does not penalize the joining of structurally distinct domains , resulting in clusters that are structurally very diverse . Furthermore , we have shown that the structural diversity within a SCOP fold is larger if the fold was defined since longer time , suggesting that the criteria underlying the definition of fold may change through time . Classifications are very useful , but the present analysis supports the view that the low similarities at the fold level are better represented as a network rather than as a tree . The comparison between the automatic and the expert classifications also indicates that the automatic classification can be improved along three lines . First , in the present study we considered protein domains as defined in the SCOP and CATH classifications . However , proteins are split into domains in the two schemes in a rather different way . In particular , some domains defined in the SCOP classification appear by visual inspection to consist of more than one domain . An incomplete domain partition can be an important source of transitivity violations and consequent errors in an automatic classification of protein structures . We are developing a new automatic method for decomposing proteins into domains based on their recurrence in a database of unrelated structures , similar to the method proposed by Holm and Sanders [57] . The domains obtained in this way will be subject to further decomposition based on their structure , to obtain a set of domains to which we will apply our clustering procedure . Secondly , our method tends to split superfamilies constituted of short domains . Some of these splitting appear to be due to the dependency of the similarity score on the protein length . The raw similarity score , either PSI or contact overlap , is transformed into a Z score in order to reduce as much as possible the dependency of the score of unrelated structures on their size . Our results show that the classification deteriorates if this normalization is not properly performed . However , due to this normalization the similarity score corresponding to identical structures decreases for decreasing domain size , which makes it more difficult to cluster together short proteins . In order to overcome this problem , it would be very helpful to define a similarity score that is independent of domain size both for unrelated and for closely related structures . This will be presented in a forthcoming work . Third , we found 63 over 779 clusters that contain protein domains defined by SCOP curators as different folds ( although 27 of these clusters are homogeneous in terms of CATH topologies ) . The distribution of structure similarity suggests that several of the foreign domains appearing in clusters that are mostly from another fold are characterized by low mean similarity , and that it could be possible to “clean” the clusters of the automatic classification . Preliminary results indicates that this strategy is promising . Significant sequence or structure similarity below the threshold for clustering [14] , [15] constitutes a very valuable information for evolutionary or functional studies . In the CASP and SCOP database , these significant cross-fold similarities are not available . We present this information in the form of two networks with structure-based and sequence-based links between the clusters of the automatic classification . In this way , we can recover not only superfamily and fold relationships that are not present in the automatic classification , but also new relationships that are not reported in expert classifications . As a concluding remark , we note that the two regimes of transitivity violations that we found can be related with two modes of protein domain evolution . In the regime of large structure similarity , transitivity violations are small , related domains are similar in size , and 95 percent of them contain domains from a single CATH or SCOP fold , whereas 86 percent contain evolutionarily related domains from the same superfamily . These results indicate that most of the domains with structure similarity above the cross-over are evolutionarily related through gene duplication and divergent evolution . Moreover , domains in different superfamilies but same fold can not be excluded to be evolutionarily related , and some careful studies have been able to demonstrate this common origin also in the absence of a clear signal from sequence similarity , as in the case of the study of TIM-barrels conducted by Nagano et al . [58] . This view also agrees with the results by Deeds et al . [59] , who tested models of convergent and divergent evolution using statistical properties of protein structural clusters , finding that the data support divergent evolution [60] . We summarize these findings saying that , for large similarity , protein domain evolution is mostly uniparental . On the other hand , similarities below the cross-over of transitivity violations are often due to partial substructures , and the typical size difference between related domains raises from 20 to 40 residues , indicating the occurrence of large insertions and deletions when the related domains belong to the same superfamily . These are clues of multi-parental evolution , proceeding through the assembly of new polypeptide fragments . This hypothetical mechanism has been proposed by Lupas et al . for the evolution of early protein domains through assembly of small peptide fragments [28] . Our findings suggest that it can also be extended to more recent evolution , consistent with another recent study [15] . In this regime the domain structure space should be regarded as continuous , and significant structure similarity should be described as a network rather than a tree . These considerations parallel recent considerations about the classification of organisms on the tree of life [61] . Speciation and evolutionary divergence generate a tree of species , which can be reconstructed by estimating the time of divergence from the molecular sequences of their genes . In order to do this , one has to use a proper sequence distance , approximately ultrametric , which makes species classification possible on a rigorous basis . Nevertheless , this view of the tree of life has been recently challenged by the discovery of the high rate of horizontal gene transfer in genome evolution . Due to horizontal gene transfer , genome evolution is multiparental , and genes that have been subject to gene transfer can not be used to reconstruct the phylogenetic tree . The extensive presence of horizontal gene transfer in evolution has led Doolittle to propose that the evolutionary relationships between organisms should be regarded as a net of life rather than a tree [61] . The present work suggests that , in the context of protein domain evolution , a tree scenario of uniparental divergent evolution is suitable to represent high similarity relationships , but a pluriparental network emerges for more remote relationships .
We have used two non redundant sets of protein domains . The first set was obtained from the ASTRAL 40 database , in which no pair has sequence similarity larger than 40% . We used the SCOP version 1 . 65 and selected only domains from the four main SCOP classes , all , all , and . The second set is the non redundant set of domains from the CATH classification , with sequence similarity smaller than 35% . Also in this case we excluded domains outside the four main classes . The final number of domains was 5041 for the SCOP set and 7073 for the CATH set . In order to select a set of domains consistently defined in SCOP and CATH , we aligned with BLAST [62] the sequences of domains in the non redundant ASTRAL40 database against domains in the non redundant CATH database at 35% sequence identity . We identified two domains to be equivalent if their BLAST evalue was smaller than 10−3 , with sequence identity larger than 75% , and their size differed by less than 10% . In this way we have obtained a set of 2890 non redundant domains classified in 779 SCOP superfamilies , 466 SCOP folds , 885 CATH superfamilies and 473 CATH topologies . We performed pairwise structure alignments using either the program MAMMOTH [41] , which is the fastest program of protein structure alignment that we know , or its multiple alignment version MAMMOTHmult [39] , which is a bit slower but much more accurate . The MAMMOTH similarity score is based on the number of aligned residues that are closer than 4Å after optimal spatial superimposition of structures and , . This is transformed into a percentage of structure identity ( PSI ) dividing it by the length of the shortest structure , ( 5 ) equals one if the two structures coincide over the length of the shorter one . There is no penalization for additional residues in the longer structure , i . e . , the score is sensitive to good partial matches and we call it partial PSI . However , the fact that the score does not penalize inserted regions may lead to join domains with very large length difference . To tackle this problem , we also defined the total similarity score , which penalizes regions in the larger structure that are not matched by the short one: ( 6 ) equals one only if the match completely covers the longer protein . Third , we adopted the contact overlap , which counts the fraction of contacts in common between two aligned structures and . Also this score is normalized in such a way to penalize partial matches . We defined the contact matrix of protein such that equals one if two heavy atoms of residues and are closer than 4 . 5Å and , and zero otherwise . We considered two cases , and . In this last case , intrahelical contacts are not considered . Denoting by the residue in structure aligned with residue in structure , the contact overlap can be written as ( 7 ) The main qualitative difference between the contact overlap and the PSI is that in the contact overlap superimposed residues in the core of the protein , which form many contacts , receive a larger weight . It is crucial for protein structure classification that the distribution of the similarity score used is almost independent of the length for comparisons of unrelated proteins . The MAMMOTH score takes care of this by normalizing the PSI in such a way that the distribution of the normalized PSI is almost independent of size for unrelated pairs: ( 8 ) where in the case of the partial PSI , and in the case of the total PSI . In the case of the overlap , we also used as a normalization . The exponents and depend on the raw similarity score and on the alignment algorithm used , and they were determined by fitting the mean and standard deviation of the PSI of unrelated structures having in some given interval , using the best fit between a Gaussian fit or an Extreme Value statistics fit ( see Table 4 ) . Using Gaussian statistics , we fit ( 9 ) and using Extreme Value statistics , we fit ( 10 ) The domain similarity score of domain in cluster A is defined as the average pairwise similarity between domain and all other domains in the cluster , ( 11 ) We programmed and tested three hierarchical clustering algorithms: average linkage [63] , single linkage and complete linkage . Starting from each element being a separate cluster , at each step all algorithms join the two most similar clusters and , and compute the similarity between the new combined cluster and all other clusters in a way that depends on the clustering algorithm . With average linkage , the combined similarity is computed as the average similarity with the two joined clusters , ( 12 ) where labels the step of the algorithm , A and B are the clusters that are joined , and is the number of elements they contain , AB denotes the new composite cluster , and C is any other cluster . Note that this updating rule is equivalent to computing the new similarity score as the average between the similarity between all pairs of elements from the cluster C and the cluster AB . With single linkage , the combined similarity is the largest similarity in the set , so that two sets are joined if at least one pair of elements is above threshold ( 13 ) With complete linkage , the combined similarity is the smallest similarity in the set , so that two sets are joined if all pairs of elements are above threshold ( 14 ) An ultrametric set is a set with an associated distance measure where every triplet of points , and fulfils a property stronger than the ordinary triangular inequality: each side of a triangle is smaller than the larger between the other two sides , i . e . , . This implies that the two longer sides must be equal . In particular , for an ultrametric set and for every threshold , it holds that if and , then . Consider now the cluster containing all elements within a distance from element , . It is easy to see that , for every pair of points and , either and coincide , or they do not share any point . Therefore , is an equivalence relationship , since if then it must also be , and the set of points can be considered discrete . A concept related to transitivity in the context of networks is the clustering coefficient , which can be computed through the formula ( 15 ) where is the number of nodes in the network , labelled as , and , is the adjacency matrix ( one if and are joined , zero otherwise ) , is the number of neighbors of node , and the clustering coefficient of node is the fraction of pairs of its neighbors and that are neighbors between each other . If the clustering coefficient is one for all nodes , connections on the network define an equivalence relationship . We have computed the clustering coefficient for the network obtained by joining domains with similarity , for various values of . To compare different similarity measures , we have plotted the clustering coefficient versus the number of clusters obtained through single linkage clustering with the same threshold . For detecting the cross-over point of transitivity violations ( TV ) , we first measure TV at each step of the clustering algorithm using Eq . ( 2 ) . We then perform two-pieces exponential fits of TV versus the step , as , where is zero for negative and one otherwise . Fits are performed for all possible cross-over points , and their quadratic error is measured as ( 16 ) where is the mean value of TV . To find the optimum in a robust way , we perform a cubic fit of the error function in an interval centered around the step yielding the minimum error , and such that for all . The analytic minimum of this cubic fitting is then selected as the best first estimate of the cross-over point . The last points in the curve , where the transitivity violations approach the maximum possible value , are very badly fitted through the two-pieces fit . Therefore , we refined the estimate of the cross-over point by removing the outliers of the optimal fit , with the conditions that a point is removed if its residual with respect to the optimal fit is more than three times larger than the median , which is the condition used to define type-1 outliers . We then apply the procedure described above to the reduced set of points , and we determine the cross-over point at which the clustering is stopped . We assessed the agreement of two classifications through the weighted kappa measure [64] , which uses as reference the expected agreement for two independent classifications with the same number of relationships . We define ( ) the number of related pairs in classification ( ) of the same objects , with pairs in total . If and are independent , the number of pairs that are either related or unrelated in both and is given by ( 17 ) We compare this number to the observed number of pairs that agree , ( 18 ) where is the number of pairs that are related in both classifications . >From this number , the weighted kappa is computed as ( 19 ) A value of zero means that two classifications are as related as independent classifications , one means that the two classifications coincide . Using the weighted kappa , we have compared the classification obtained at every step of the clustering algorithm with the manual classifications of CATH and SCOP at the superfamily and the fold level . Notice that the weighted kappa can be decomposed into the contributions of related and unrelated pairs as follows: ( 20 ) where is the number of pairs related in both classifications expected by random , , and the weights are and for related and unrelated pairs , respectively . For the sake of illustration , we have represented two domain similarity networks obtained before and beyond the stopping point of the automatic classification . Two networks were constructed by considering each cluster as a node , and connecting nodes with . In the first case , we used clusters obtained before the cross-over point of the average linkage algorithm using a high similarity threshold , and we connected them if , which is the similarity at the cross-over point . In the second case we used clusters generated at the cross-over point and we connected them with . The networks have been visualized using the Pajek software [65] . To visualize spatial superimpositions , we used the multiple structure allignments program MAMMOTHmult [39] in combination with the Pymol software . | Making order of the fast-growing information on proteins is essential for gaining evolutionary and functional knowledge . The most successful approaches to this task are based on classifications of protein structures , such as SCOP and CATH , which assume a discrete view of the protein structure space as a collection of separated equivalence classes ( folds ) . However , several authors proposed that protein domains should be regarded as assemblies of polypeptide fragments , which implies that the protein–structure space is continuous . Here , we assess these views of domain space through the concept of transitivity; i . e . , we test whether structure similarity of A with B and B with C implies that A and C are similar , as required for consistent classification . We find that the domain space is approximately transitive and discrete at high similarity and continuous at low similarity , where transitivity is severely violated . Comparing our classification at the cross-over similarity with CATH and SCOP , we find that they join proteins at low similarity where classification is inconsistent . Part of this discrepancy is due to structural divergence of homologous domains , which are forced to be in a single cluster in CATH and SCOP . Structural and evolutionary relationships between consistent clusters are represented as a network in our approach , going beyond current protein classification schemes . We conjecture that our results are related to a change of evolutionary regime , from uniparental divergent evolution for highly related domains to assembly of large fragments for which the classical tree representation is unsuitable . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [
"computational",
"biology/macromolecular",
"structure",
"analysis",
"biophysics/structural",
"genomics",
"biophysics/protein",
"folding",
"molecular",
"biology/molecular",
"evolution",
"molecular",
"biology/bioinformatics",
"evolutionary",
"biology/bioinformatics"
] | 2009 | Cross-Over between Discrete and Continuous Protein Structure Space: Insights into Automatic Classification and Networks of Protein Structures |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.