pmid stringlengths 8 8 | pmcid stringlengths 8 11 ⌀ | source stringclasses 2
values | rank int64 1 9.78k | sections unknown | tokens int64 3 46.7k |
|---|---|---|---|---|---|
33814597 | null | s2 | 8,141 | {
"abstract": "There is a disconnect between the cutting-edge research done in academic labs, such as nanotechnology, and what is taught in undergraduate labs. In the current undergraduate curriculum, very few students get a chance to do hands-on experiments in nanotechnology-related experiments most of which are through selective undergraduate research programs. In most cases, complicated synthesis procedures, expensive reagents, and requirement of specific instrumentation prevent broad adaptation of nanotechnology-based experiments to laboratory courses. DNA, being a nanoscale molecule, has recently been used in bottom-up nanotechnology with applications in sensing, nano-robotics, and computing. In this article, we propose a simple experiment involving the synthesis of a DNA nanoswitch that can change its shape from a linear \"off\" state to a looped \"on\" state in the presence of a target DNA molecule. The experiment also demonstrates the programmable topology of the looped state of the nanoswitch and its effect on gel migration. The experiment is easy to adapt in an undergraduate laboratory, requires only agarose gel electrophoresis, a minimal set-up cost for materials, and can be completed in a 3-hour time frame."
} | 304 |
36133834 | PMC9419300 | pmc | 8,142 | {
"abstract": "Self-cleaning surfaces often make use of superhydrophobic coatings that repel water. Here, we report a hydrophobic Si nanospring surface that effectively suppresses wetting by repelling water droplets. The dynamic response of Si nanospring arrays fabricated by glancing-angle deposition is investigated. These hydrophobic arrays of vertically standing nanosprings (about 250 nm high and 60 nm apart) allow the droplets to rebound within a few milliseconds after contact. Amazingly, the morphology of the nanostructures influences the impact dynamics. The rebound time and coefficient of restitution are higher for Si nanosprings than for vertical Si columns. By considering the droplet/nanospring surface as a coupled-spring system, we argue that the restoring force of the nanosprings may be responsible for the water-droplet rebound. The bouncing phenomena studied here are essential in the design of self-cleaning surfaces and are also of fundamental importance for the study of wetting behavior on nanostructures.",
"conclusion": "4. Conclusion The dynamics of water droplets falling on vertical nanorods, tilted nanorods, and nanosprings of silicon were studied. After impact with the surface, the water droplet initially spreads and flattens. It then recoils in a way that is greatly influenced by the underlying morphology of the nanostructured surface. On slanted Si nanorods, no recoil was observed, whereas, on vertical nanorods and on nano-helices, the recoil was completed in approximately 16 ms. Interestingly, water droplets were observed to bounce on hydrophobic nanosprings with higher rebound time and COR than on vertical nanostructures. The elastic force arising from the difference between the equilibrium droplet shape and the deformed droplet shape drives the recoiling flow. The restoring force of the nanosprings may be responsible for the rebound of the water droplet; this may be seen by considering the droplet/nanospring surface as a coupled-spring system. As dust particles can be removed easily by the bouncing and rolling of droplets, there is enormous scope for multifunctional applications involving such nanospring arrays: self-cleaning windows, liquid-repellent exteriors, glass panels of solar cells, and antifouling agents for roof tiling.",
"introduction": "1. Introduction Self-cleaning surfaces are of increasing technological importance. The discovery of a new class of films that could render ordinary surfaces self-cleaning is of evident technological importance, and, if it proves economical, of great practical importance. Micro- and nanostructured surfaces with special wetting behaviors have received considerable attention in recent years. 1–6 Non-wettability is a crucial surface property that plays an important role in daily life, industry, and agriculture. The lotus effect is an example of self-cleaning in nature: superhydrophobic leaves protect the lotus plant against pathogens or fungi. 7 Depending on the surface energy and ruggedness of its microstructures, a surface can be hydrophilic, hydrophobic, or superhydrophobic. 8 Superhydrophobic surfaces can be fabricated by chemically modifying a surface with a low surface energy coating or by creating a surface from a hydrophobic material that exhibits roughness at the micro- or nanoscale. 9 For any practical applications, the superhydrophobicity and non-wetting behavior must be maintained under dynamic conditions when the droplet impacts the surface with a given velocity. On superhydrophobic surfaces, water will form almost spherical droplets with very high contact angles. When landing on such a surface, the water droplet may rebound; this is critical in situations where the impact of the water droplets on the surface is important, for example, in deicing applications. 10,11 The necessary conditions for droplet-bouncing have been considered in the literature. For example, bouncing can be easily achieved on superhydrophobic surfaces, as the interactions between the droplet and surface that might prevent the drop from bouncing are minimal. 12 When a droplet falls on such a surface, the rough structures of the surface and the air trapped in them can produce a significant capillary pressure that helps the droplet rebound from the surface. 13–16 Several studies have elucidated the dynamics of a bouncing droplet 16–22 as a function of the surface micro-and nanostructure 18,23,24 or of the impact velocity. 25 The shape-change in the droplet has also been shown to be a direct indicator of the contact angle and hydrophobicity. The bouncing of water droplets has been studied to determine the hydrophobicity of surfaces, and a relationship has been established between the contact angle of the water and the number of bounces. 26 It has also been reported that the surface must have a contact angle of at least 150° for a droplet to bounce ( i.e. , for the kinetic energy of the impinging droplet to be transformed into surface energy). 12,26,27 Other studies suggest that the hysteresis of the contact angle plays a crucial role in the bouncing behavior of impacting droplets. 28 Moreover, in addition to depending on the wetting properties of the surface, the rebound depends on such parameters as the surface tension, viscosity, and velocity of the droplet at impact. 12,18,20,22,25 There are many reports of the bouncing of a water droplet on high contact-angle (superhydrophobic) static surfaces, 22,29–31 but bouncing on a hydrophobic nanosprings structure, to the best of our knowledge, has not been reported. Here, we show that an ultrathin film of nanospring arrays can cause water droplets to rebound. We demonstrate that nanostructured surfaces that have comparable static contact angles exhibit remarkably different droplet-rebound dynamics. Even though millions of nanostructures interact simultaneously with a single water droplet, the underlying shape of the nanostructures can determine whether the droplet flies off the surface.",
"discussion": "3. Results and discussion \n Fig. 1 shows the top and cross-sectional SEM micrographs of the different Si nanostructures used in this study. The inset shows optical images of 10 μL water droplets on the corresponding surfaces. The average thickness, average diameter, and solid fraction of each sample are given in Table 1 . Fig. 1 Top and cross-sectional SEM images of thin film (a and b); vertically standing nanorods (c and d); slanted nanorods (e and f); and Si nanospring array (g and h), respectively. The inset shows the images of 10 μL water droplet on the corresponding surfaces. The contact angle was minimum for the TF and maximum for the VC. Altough the contact angles of the SC and NS were roughly equal, the water-droplet impact dynamics were found to be very different. The average thickness, average diameter, and solid fraction for the samples Sample Average thickness (nm) Average diameter (nm) Solid fraction (%) From contact angle From SEM image TF 553 ± 2 NA NA NA SC 265 ± 8 53 ± 15 35 43 VC 240 ± 2 48 ± 15 20 27 NS 256 ± 4 45 ± 10 35 45 Each nanostructured sample had a thickness of approximately 250 nm and an average diameter of approximately 50 nm. Generally, when a droplet is placed statically on a periodic nanostructured surface, the droplet shape is symmetric and is determined by the minimization of the total surface energy. The static apparent contact angle (APCA) values of water droplets on TF, SC, VC, and NS were observed to be 106.0°, 135.7°, 148.6°, and 138.6°, respectively ( Fig. 2 ). The contact angle was minimum for the TF and maximum for the VC. The contact angles of the SC and NS were approximately equal, although the water-droplet impact dynamics were found to be very different. Fig. 2 The bar graph showing static apparent contact angle, APCA, and scatterplot showing the advancing contact angle, ACA (blue points), receding contact angle, RCA (green points) and contact angle hysteresis, CAH (red points) values of water droplets on TF, SC, VC, and NS. The error bars represent the standard deviations of five identical measurements. Of all the samples, VC had the highest APCA and the smallest CAH. The chemical composition and morphology of a surface define its wetting properties. All the samples were made of Si and were coated with the same chemical (which resulted in a slightly higher contact angle). Given this compositional uniformity, the difference in contact angle must be due to the surface morphology of the samples. The nanocolumnar structure made the sample surface very rough and porous, resulting in an increase in contact angle as compared to the conventional thin film. The contact angle was found to increase from 106° for the conventional film to 148° for the vertical nanocolumnar sample. The increase in contact angle on the nanocolumnar samples can be attributed to the decrease in the solid fraction of the nanostructures, as per the Cassie–Baxter model. 8 In this model, the surface is a composite of air and Si, and the water droplet sits on the air trapped between the rough surfaces with apparent contact angle θ A . The solid fraction f is given by 1 where θ 0 is the contact angle on a conventional surface. The calculated solid fractions from eqn (1) and the SEM images are in good agreement. The slightly lower solid fraction calculated from the Cassie–Baxter model may be because of a change in contact angle due to the chemical modification. A similar increase in contact angle with the nanocolumnar structure has also been reported in previous studies. 44–46 The spreading dynamics of a water droplet on vertical Si nanocolumns have also been studied by other groups, but we are interested only in the bouncing behavior of water droplets on these nanostructures. 47 The droplet size, liquid viscosity μ , and impact velocity v 0 all influence the impact dynamics. A dimensionless variable—the Weber number W e (the ratio of the kinetic energy to the surface energy)—can be used to characterize the impact dynamics: 48 2 where D 0 is the droplet diameter, ρ is the density, and σ is the surface tension of the liquid. In this paper, ρ , D 0 , and σ are fixed, so W e only varies with ν 0 . Droplets can rebound for high values of W e (≥10). 49 For this reason, we performed experiments at W e ≈ 7, 10, and 14. We have performed additional experiments by varying the Weber number up to 70 for vertical and nanospring samples. The extended results are reported in ESI. † Water droplets of 10 μL volume (diameter ≈ 2.67 mm) were dropped onto the silanized Si nanostructured samples. The droplets were released from a height of 10 mm above the surface with impact velocity 44 cm s −1 and W e = 7.2. When the droplet was dropped from any height on the slanted nanorods, the droplet deformed to an ellipsoidal shape and then recoiled without detaching from the surface ( Fig. 3 ). A similar phenomenon was observed on thin-film for all impact velocities. The maximum spreading diameter ( d max ) depends on the impinging velocity of the droplet and the capillary and viscous forces, as well as the properties of the liquid and the solid surface. 50 When evaluating the spreading behavior of a droplet, the maximum spreading diameter is usually normalized with respect to the initial diameter ( d 0 ) as the dimensionless spreading factor, β m = d max / d 0. The maximum spreading factors for TF, SC, VC, and NS in this study were approximately 1.47, 1.36, 1.39, and 1.36, respectively. The β m was found to increase with increasing contact angle and also with increasing W e , Fig. 4 . The increase in the β m with an increase in the contact angle can be explained using the equation 51 3 where θ a is the advancing contact angle and R e is the Reynolds number. Fig. 3 Time evolution of 10 μL water droplets dropped from a height of 10 mm on different nanostructures (ESI). † Water droplet bouncing was observed on vertical columns and nanospring samples. The water droplet rebounded on VC and NS and left the surface in approximately 16 ms. Fig. 4 The dimensionless spreading factor, β m = d max / d 0 vs. W e for TF, SC, VC, and NS. The β m increases with increasing contact angle sand also with increasing W e . The evolution of the spreading factor is divided into four phases: the kinematic, spreading, relaxation, and wetting phases, respectively. Most of the spreading occurs during the spreading phase, which is dominated by inertia. The increase in the maximum spreading diameter can be due to the increase with increasing W e . The error bars represent the standard deviations of five identical measurements. The spreading mechanism of a drop onto a solid surface has been studied in detail in the past. 52 The evolution of the spreading factor is divided into four phases: the kinematic, spreading, relaxation, and wetting phases, respectively. Most of the spreading occurs during the second of these, which is dominated by inertia. 53 The increase in inertia can explain the increase in the maximum spreading diameter with increasing W e . The time to reach maximum deformation was also found to depend on the impacting surface: it was maximum for TF and minimum for NS ( Table 2 ). Sample NS took around 3.67 ms, on average, for maximum deformation when W e ≈ 7. For the VC sample, the droplet reached its maximum deformation at approximately t = 4.11 ms; after that, surface tension and viscous forces overcame inertia so that fluid accumulated at the leading edge of the splash and started pulling back. Droplets with higher velocity will have higher inertia and will take less time for maximum deformation. Hence, the time for maximum deformation decreased with an increase in impact velocity. Velocity, Weber number, maximum deformation, time for maximum deformation, rebound time, and coefficient of restitution for the four samples. The time to reach maximum deformation was also found to depend on the impacting surface and was maximum for TF and minimum for NS sample Sample Velocity (cm s −1 ) Weber no. W e maximum spreading diameter ( d max ) (mm) Time for max. deformation (ms) Rebound Time (ms) Time of flight (ms) Coefficient of restitution (COR) TF 44.3 7.2 3.92 4.67 — 54.2 10.8 4.04 4.11 — 62.6 14.3s 4.47 3.56 — SC 44.3 7.2 3.72 4.89 — 54.2 10.8 3.90 3.56 — 62.6 14.3 4.10 3.33 — VC 44.3 7.2 3.63 4.11 16.11 22.67 0.25 54.2 10.8 3.69 3.56 15.67 27.22 0.25 62.6 14.3 3.92 3.33 16.33 29 0.23 NS 44.3 7.2 3.70 3.87 15.33 14.68 0.16 54.2 10.8 3.82 3.67 17.42 19.33 0.18 62.6 14.3 4.12 3.44 18.58 26.33 0.21 The rebound time at which the droplet bounces off the surface is crucial because it determines the degree of energy transfer. When the droplet fell on vertically aligned nanorods from a height of 10 mm, it rebounded and left the surface in about 16.11 ms. However, when a droplet of the same volume fell with the same impact velocity of 44 cm s −1 on the Si nanosprings (APCA < 150°), instead of wetting the surface, the droplet bounced, leaving the surface in about 15.33 ms. The NS structure not only showed the bouncing of the droplet on the hydrophobic surface but also reduced the contact time (≈15.33 ms) and the time for maximum spreading (≈3.67 ms) relative to that of VC samples. 31 It is interesting to note that the rebound time for the VC sample was almost constant (≈16 ms) with increasing W e , but the rebound time for the NS sample increased from 15.3 to 18.5 ms. The spreading dynamics, in the case of VC, is consistent with the previous report by Fan et al. 47 The bouncing behavior on the VC was not unexpected, as it had a contact angle of ≈ 148.6° ± 4.0 and thus satisfied the first necessary condition for bouncing behavior. The contact angle for the NS was around 138.6° ± 3.0, but surprisingly this sample also showed the bouncing behavior. Some prior works have concluded that contact angle hysteresis (CAH) plays a significant role in bouncing from the surface. 28 The droplet impact process involves an interplay between the kinetic energy, surface energy, and viscous dissipation of the water droplet. The elastic force is due to the surface tension of the water droplet, whereas viscous dissipation is the cause of energy dissipation. Before the impact with the surface, the droplet possesses only kinetic energy. A drop impacting a solid is deformed, and a shock wave spreads radially outward towards the surface up to the point when the viscosity dissipates the kinetic energy (the dissipation due to heat can be neglected for water.) When the droplet reaches its maximum deformation, the restitution force due to surface tension comes into play, causing the droplet to recoil. Now the droplet shrinks and moves radially inward, gaining kinetic energy; a jet rises in the center (the Worthington jet), which may lead to the lift of the droplet ( Fig. 3 ). The droplet must do work to overcome the resistance force produced by CAH. The total work W done in the spreading and receding process of a droplet is given by 28 4 where cos θ r and cos θ a are the cosines of the receding and advancing contact angles (cos θ r − cos θ a ) is the CAH, and D 0 is the initial diameter of the droplet. Lower CAH values result in less work against the resistive force, and very little energy is required to overcome the work done, resulting in a rebound. The lower CAH value for the VC sample may be the other reason that it exhibits a rebound despite the longer contact time. SC and NS had mutually similar CAH values (25 and 27.6, respectively), but only NS showed the rebounding property. On superhydrophobic surfaces, the dynamics of a droplet impinging on a surface depending on the competition between the three wetting pressures: water-hammer pressure, P wh = ρC w v 0 , dynamic pressure, and anti-wetting capillary pressure, Here, ρ is the water density, C w is the speed of sound in water, V i is the droplet velocity, γ LV is the surface energy of the water at the water and vapor interface, θ a is the advancing contact angle, and B is the spacing between the nanostructures. 54,55 Capillary pressure is caused by the air trapped by the surface roughness. The air cushion trapped between the nanorods and the water droplet acts as an effective spring. For a droplet to rebound from the surface, the non-wetting condition P c > P wh > P d must be satisfied. In this study, for the three experimental heights in increasing order and considering ρ = 1000 kg m −3 and C w = 1482 m s −1 , P wh and P d were found to vary from 0.66–0.93 MPa and 0.1–0.2 kPa, respectively. P c was calculated as 2.44 MPa, 2.76 MPa, and 4.33 MPa for the SC, NS, and VC, respectively. The maximum capillary pressure was generated by the VC. For NS and SC nanostructured surfaces, the P c values were comparable to each other; however, the bouncing phenomenon was observed only on the nanosprings surface. Thus, capillary pressure alone cannot be the reason for the observed bouncing behavior of the droplet on the nanospring surface. The rebound of the droplet on the surface of the nanosprings is surprising: it is generally assumed that only superhydrophobic surfaces support bouncing, as only on them do the capillary forces allow the drop to leave the surface. A detailed model for the rebound of a water droplet on vertically aligned nanorods can be found in the ESI. † We propose the hypothesis that the elastic property of the nanosprings has a significant role in the bouncing of the water droplet. The rebound of an impinging droplet is only possible if its kinetic energy is larger than the surface energy dissipated during the retraction stage. Bouncing water droplets are vertically deformable, and, upon impact, some of the kinetic energy can be stored by the deformation of the droplet itself. 27 Thus, the droplet itself behaves like a spring, the stiffness of which is the surface tension. 27,56 The nanosprings can store sufficient energy to facilitate a rebound that causes the droplet to detach from the surface completely. We have modeled the elasticity of the droplet in contact with the elastic nanosprings as an effective mechanically coupled double-spring system. More specifically, we model the droplet by two identical masses m linked by a spring of stiffness k w and rest length L . The viscous effects are modeled by a mechanical damper with a dissipation parameter β . The coordinate y is taken vertically upward; the vertical positions of the upper and lower mass are y 1 and y 2 , respectively. A schematic representation of the spring system is given in Fig. 5 . The force of gravity acting on the two masses is F g1 = − mg = F g2 . The spring also exerts forces on each mass given by F s1 = − k w ( y 1 − y 2 − L ) and F s2 = k w ( y 1 − y 2 − L ). When the lower mass is in contact with the nanospring, it experiences a normal force F ns = − k ns y 3 , where k ns is the stiffness constant of the nanospring. F ns depends on the compression of the nanospring, which varies during contact; it is zero when the droplet is not in contact with the surface. The constant k w determines the undamped frequency of the spring, given by f 0 = √( k w / m ). Fig. 5 Representation of a mechanically coupled double-spring system by two identical masses m linked by a spring of stiffness k w and rest length L considering the water droplet as an elastic spring. Bouncing water droplets are vertically deformable, and, upon impact, some of the kinetic energy can be stored by the deformation of the droplet itself. Consequently, the droplet itself behaves like a spring, the stiffness of which is the surface tension. The loss of energy of two objects after a collision can be described in terms of the coefficient of restitution (COR), which itself depends on the elastic properties of the colliding objects. Since, in this study, one of the colliding objects is always a water droplet, the COR will change only with the elastic properties of the nanostructured surfaces. For an increase in W e , the COR was found to be almost constant for VC but to increase for NS (the decrease in the COR value of VC when W e = 14 may be due to the fact that the air trapped under the water droplet was forced out because of the higher impact velocity). We can explain the increase in the rebound time and COR for the NS sample if we consider the compression of the nanospring structure by droplet impact. The relation between the initial velocity v 0 , maximum compression y m of the nanospring, and spring constant k is given by 57 5 A higher velocity will lead to a higher compression, which may increase the rebound time on the nanospring sample. The potential energy of the nanosprings is also directly proportional to the square of the maximum compression and spring constant. The nanospring will absorb more energy for higher W e and return a higher fraction of energy on recoil. Kaneko et al. have already shown that a Si nanospring fabricated using the GLAD technique exhibits nonlinear elastic mechanical behavior. 58 They reported the load–displacement ( F – δ ) relationships obtained during the loading and unloading processes. The nanospring showed nonlinear reversible behavior; the relationship between load, F [nN], and displacement, δ [nm], was determined to be F = 4.1 δ + 0.0041 δ 2 . They also confirmed that this nonlinearity originated from the large deformation permitted by the spring shape. Therefore, if we consider the droplet/nanospring surface as a coupled-spring device, we can understand the bouncing behavior on the NS sample along with the increase in the rebound time and COR."
} | 5,917 |
32269706 | PMC7114960 | pmc | 8,143 | {
"abstract": "Production of bioethanol from brewers spent grains (BSG) using consolidated bioprocessing (CBP) is reported. Each CBP system consists of a primary filamentous fungal species, which secretes the enzymes required to deconstruct biomass, paired with a secondary yeast species to ferment liberated sugars to ethanol. Interestingly, although several pairings of fungi were investigated, the sake fermentation system ( A. oryzae and S. cerevisiae NCYC479) was found to yield the highest concentrations of ethanol (37 g/L of ethanol within 10 days). On this basis, 1 t of BSG (dry weight) would yield 94 kg of ethanol using 36 hL of water in the process. QRT-PCR analysis of selected carbohydrate degrading (CAZy) genes expressed by A. oryzae in the BSG sake system showed that hemicellulose was deconstructed first, followed by cellulose. One drawback of the CBP approach is lower ethanol productivity rates; however, it requires low energy and water inputs, and hence is worthy of further investigation and optimisation. Electronic supplementary material The online version of this article (doi:10.1007/s12155-016-9782-7) contains supplementary material, which is available to authorized users.",
"conclusion": "Conclusions The sake CBP system ( A. oryzae and S. cerevisiae NCYC479) was by far the most effective of all permutations tested for ethanol production from BSG, with maximal ethanol yields of ca. 37 g/L produced within 10 days. On this basis, 94 kg of pure ethanol could be produced from 1 t of BSG using 36 hL water. Whilst volumetric productivity was moderate (3.7 g/L/day), the process requires no pre-treatment and no exogenous enzymes. The final waste residue contained >22 % crude protein. Utilising this co-product stream efficiently (e.g. as an animal feed) would further improve overall process economics.",
"introduction": "Introduction The production of advanced biofuels (second generation biofuels) from lignocellulosic biomass has a number of technical difficulties associated with it due to the recalcitrant nature of the material [ 1 ]. A thermo-chemical pre-treatment of some form is usually required to enhance the subsequent enzymatic hydrolysis (or saccharification) yield [ 2 ]. This pre-treatment stage is considered to be the most energy intensive and expensive stage of conventional biofuel production [ 2 ]. The subsequent enzymatic saccharification stage incurs further costs in terms of commercial enzyme preparations containing cellulases and xylanases. Overall, lignocellulosic biofuel production is hindered by economic factors which currently limit its widespread, large-scale production [ 3 ]. As a consequence of these technical difficulties and the economic implications of this approach, researchers have investigated entirely biological alternatives, such as simultaneous saccharification and fermentation (SSF) and consolidated bioprocessing (CBP; [ 4 ]). CBP involves the conversion of lignocellulose into the required products in one step, without the addition of enzymes. Most attempts at CBP have utilised individual organisms, such as thermo-tolerant yeast strains [ 5 ] or bacteria (e.g. species of Clostridia; [ 6 ]). Frequently, a genetically modified (GM) approach has been used for expression or over-expression of lignocellulolytic enzymes. However, the usage of obligate anaerobic species such as Clostridia has numerous technical difficulties associated with it, not least due to its pathogenicity to humans. Also, the usage of GM organisms has additional implications and restrictions in many parts of the world. The production of ethanol by filamentous fungal species, already capable of secreting lignocellulolytic enzymes, has been reported [ 4 ]. In the absence of any thermo-chemical pre-treatment, the purely biological deconstruction and saccharification method would likely require a wide variety of carbohydrate degrading (CAZy) and associated enzymes [ 7 ]. Considering the natural propensity of many fungal species to deconstruct lignocellulosic material in the wild, the present research evaluated the potential to produce ethanol via CBP using fungal consortia. In this approach, a primary organism, such as a filamentous fungus, is used to deconstruct the lignocellulosic material through secretion of its native lignocellulolytic enzymes. Subsequently, a secondary fungal species, such as an industrial yeast strain, can ferment any liberated monomeric sugars into ethanol. Various candidates for the primary fungus are known, which secrete the required arsenal of lignocellulolytic enzymes, including Aspergillus niger , Aspergillus oryzae , Trichoderma reesei and Humicola insolens [ 8 ]. The production of the Japanese alcoholic beverage sake utilises a consortium of A. oryzae and S. cerevisiae (NCYC479) to produce high concentrations of ethanol (ca. 20 % ABV) from the starch component found within rice [ 9 ]. A. oryzae is responsible for the secretion of the enzymes (primarily α-amylases and endo-1,4-α- d -glucan glucohydrolase EC 3.2.1.1) that hydrolyse the starch into glucose, which S. cerevisiae then utilises for ethanol production [ 10 ]. However A. oryzae has also been shown to secrete lignocellulolytic enzymes (endocellulases and various xylanases) in significant quantities when cultured in media containing lignocellulosic biomass [ 11 ]. The sake fermentation system was thus of interest to us as a potential CBP approach for lignocellulosic bioethanol production, particularly since A. oryzae and S. cerevisiae are known to exhibit suitably high ethanol tolerance phenotypes [ 12 ]. Advanced biofuel production using agricultural or industrial co-product processing streams containing lignocellulosic material is of particular interest because it avoids the human ‘food versus fuel’ dilemma of first-generation biofuel substrates [ 4 ]. Brewers spent grain (BSG) is a co-product of the brewing process which is abundant with ca. 9.9 million tons wet weight annually from the EU alone (calculation based upon beer production in the EU for 2014 at approximately 522.8 million hL [ 13 ] with the following assumptions: approximate global average malt requirement of 11.8 kg/hL, 22 % dry weight of malt inwards generating 1.35 million metric tonnes [dry weight] with ca. 78 % moisture content producing 6.1 million tonnes fresh weight and 19 kg [fresh weight] per hL). BSG is also typically sold at a very low cost of ca. £38 per tonne wet weight [ 14 ]. Within BSG, the hemicellulose and cellulose contents typically range from 10 to 25 % and 15 to 30 %, respectively (depending on the barley cultivar and brewery technology employed), providing a significant pool of potentially fermentable sugars. In addition, BSG typically contains between 15 and 27 % protein. In contrast to lignocellulosic biofuel substrates such as wheat straw or switchgrass, this renders BSG an ideal substrate for microbial growth due to the significant nitrogenous component (which facilitates the production and secretion of various enzymes). BSG has been used as a growth substrate for the cultivation of the fungus, T. reesei , for the production of various cellulase enzymes (endo and exoglucanases; [ 15 ]). In addition, BSG has also been used as a substrate for the cultivation of A. oryzae for the production of α-amylases. A. oryzae has also been shown to secrete various proteases when grown on wheat straw [ 16 ] indicating its potential to utilise the significant nitrogenous component found within BSG. Here, we evaluate a CBP approach to ethanol production from BSG using consortia of various filamentous fungal species, each paired with a selected yeast strain. The sake fermentation system was included in the study for the reasons outlined above and is compared with alternative consortia in terms of ethanol productivity. In addition, the gene expression of a selection of key carbohydrate degrading (CAZy) enzymes and associated enzymes by A. oryzae was studied in the ‘sake fermentation’ of BSG, to try and identify which substrates were being utilised and at which stages of the fermentation.",
"discussion": "Results and Discussion CBP Using A. oryzae with S. cerevisiae NCYC479 (Sake-Based CBP System) Of all the pairings of primary filamentous fungi and secondary yeast strains which were tested, the sake pairing ( A. oryzae and S. cerevisiae NCYC479 ) was by far the most effective in terms of ethanol production (Table 2 ). Relatively high ethanol concentrations of up to 37 g/L (ca. 4 % ABV) were attained using this system with 10-days incubation at 15 °C, on a substrate of dried and ground BSG (Fig. 1 a). The data in Table 2 are ordered by increasing ethanol yield; hence, it can be readily noted that the eight best ethanol-yielding systems all contained both A. oryzae and S. cerevisiae NCYC479 . When BSG direct from a lauter tun was used as substrate, without prior drying and milling, ethanol yields were significantly lower (at best ca. 9.8 g/L after 5 days). This highlighted the requirement for some form of particle size reduction of the BSG in order to enhance the saccharification of the lignocellulosic material by the enzymes secreted from the A. oryzae . A reduction in particle size may also allow greater penetration of the substrate by fungal hyphae, as described in Lee [ 25 ], thus facilitating more successful interaction between the secreted enzymes and the substrate. However, conversely, the relatively high temperature drying process (105 °C) that occurred prior to the particle size reduction could have actually had a negative effect on sugar liberation from cellulose (and therefore also affected the ethanol yields). This is through the collapse of the cellulose pores which may have ultimately impeded enzymatic access somewhat. The superior ethanol yields from the sake-based CBP system (relative to other permutations of fungal species tested) were somewhat unexpected, since A. oryzae is not usually considered a cellulolytic fungal species, being more noted for starch hydrolysis [ 10 ]. Semi-quantitative analysis of secreted enzyme activities, using the congo red staining method, indicated significant cellulase and xylanase activity in the supernatants generated from 5-days incubation and onwards. Table 2 Maximal ethanol yields (g/L) and volumetric productivity (g/L/day) achieved from all CBP variants tested. Data are the mean of three replicate experiments CBP/SSF system Mean maximal ethanol yield (g/L) Volumetric productivity a (g/L/day) \n A. niger + S. cerevisiae NCYC2592 0.7 ± 0.1 0.1 \n A. niger + S. cerevisiae NCYC479 6.7 ± 0.4 2.3 \n H. insolens + S. cerevisiae NCYC2592 8.7 ± 0.5 0.9 \n A. oryzae + S. cerevisiae NCYC479 (BSG direct from lauter tun) 9.8 ± 2.2 1.0 \n A. oryzae + S. cerevisiae NCYC479 (1 % HCl 121 °C pre-treated BSG) 12.2 ± 0.8 1.2 \n A. oryzae + Kluyveromyces spp. NCYC546 14.8 ± 1.4 1.5 \n A. oryzae + Kluyveromyces spp. NCYC546 + Novozymes Cellic®CTec2 16.5 ± 3.4 1.7 \n A. oryzae + Kluyveromyces spp. NCYC179 18.2 ± 1.7 1.8 \n A. oryzae + Kluyveromyces spp. NCYC179 + Novozymes Cellic®CTec2 18.3 ± 3.7 1.8 \n H. insolens + S. cerevisiae NCYC479 18.5 ± 0.1 1.9 \n A. oryzae + S. cerevisiae NCYC2592 20.3 ± 0.6 2.0 \n A. oryzae + Kluyveromyces spp. NCYC1426 20.8 ± 1.0 2.1 \n A. oryzae + Kluyveromyces spp. NCYC1426 + Novozymes Cellic®CTec2 22.6 ± 2.0 2.3 \n A. oryzae + S. cerevisiae NCYC479 + Novozymes Cellic®CTec2 (30 °C) 24.6 ± 1.6 2.5 \n A. oryzae + S. cerevisiae NCYC479 + Kluyveromyces spp. NCYC1426 + Novozymes Cellic®CTec2 25.0 ± 2.0 2.5 \n A. oryzae + S. cerevisiae NCYC479 + Kluyveromyces spp. NCYC179 + Novozymes Cellic®CTec2 26.1 ± 4.0 2.6 \n A. oryzae + S. cerevisiae NCYC479 + Kluyveromyces spp. NCYC546 + Novozymes Cellic®CTec2 26.2 ± 4.4 2.6 \n A. oryzae + H. insolens + S. cerevisiae NCYC479 (30 °C) 30.0 ± 0.1 3.0 \n A. oryzae + S. cerevisiae NCYC479 + Novozymes Cellic®CTec2 (15 °C) 32.6 ± 1.6 3.3 \n A. oryzae + H. insolens + S. cerevisiae NCYC479 (15 °C) 33.0 ± 1.4 3.3 \n A. oryzae + S. cerevisiae NCYC479 36.8 ± 2.4 3.7 All BSG was dried and ground unless otherwise stated \n a Volumetric productivity calculations based on number of days taken to achieve highest mean ethanol yields for all CBP systems tested \n Fig. 1 Time course of ethanol production from BSG using consolidated bioprocessing with fungal consortia under the specified conditions (see data series legends). a At 15 °C. b At both 15 and 30 °C using the sake system but with the addition of Novozymes Cellic® CTec2 on day 10 (10 FPU/g biomass). c At 15 °C and inoculated initially with S. cerevisiae NCYC479 on day 0 and then subsequently inoculated with Kluyveromyces spp. and boosted with Novozymes Cellic® CTec2 on day 10 (dosed at 10 FPU/g biomass). d \n A. oryzae in partnership with three different strains of Kluyveromyces : marxianus NCYC1426, marxianus NCYC179 and wickerhamii NCYC546 at 15 °C \n No sugars (monosaccharides or oligosaccharides less than Dp5) were detected in any supernatant samples (when using either secondary yeast strain partner) indicating that any sugar liberated from the lignocellulosic material was likely occurring in very small trace amounts and was being utilised immediately by either the yeast species or the filamentous fungus. The rapid utilisation of any liberated sugars by the yeast species (the S. cerevisiae ) could possibly limit catabolite repression or product inhibition of any enzymes secreted by the primary deconstructive organism (the fungus A. oryzae in this case) and may also further induce enzyme secretion from the fungus due to the lack of immediately available carbon sources. This is similar to the phenomenon described in Suto and Tomita [ 26 ]. Although the IC analysis could not detect oligosaccharides greater than Dp5, it was also considered unlikely that any of the fungi used in these CBP systems would have been able to directly utilise these larger molecular weight sugars in the supernatant. Further hydrolysis of these cellodextrins to lower molecular weight sugars such as glucose would be a more probable mechanism of carbon source utilisation for both the filamentous fungi and yeast. A very small quantity of ethanol was produced in the koji controls (i.e. vessels containing no BSG; with just the koji A. oryzae inoculum and S. cerevisiae ) which could be attributed to hydrolysis of the very small quantity of starch that was present in the koji inoculum. However, this amount was subtracted from the final ethanol yields generated so as to allow for accurate evaluation of the degree of usage of the BSG carbon source. Therefore, the presence of a small quantity of starch contributed by both the koji and also in the BSG could have facilitated the initial growth of A. oryzae (as a rapidly utilisable carbon source favoured by the A. oryzae with its strong arsenal of amylose degrading enzymes) which then subsequently encouraged greater lignocellulosic enzyme secretion once the media was depleted of more readily available carbon sources. The small quantity of starch present in the BSG (ca. 1 % w / w ) was also only considered to potentially contribute a small quantity of the carbon flux to ethanol. This was calculated to potentially offer a maximum contribution of approximately 1.5 g/L ethanol (of the total produced). Therefore, when considering the high ethanol yields achieved (ca. 37 g/L optimally), it is clear that starch (as the carbon source alone) could not explain the ethanol produced and a significant contribution was likely provided by the lignocellulosic material. Ethanol yields from BSG using the sake-fermentation system were compared at 15 and 30 °C. Similar ethanol concentrations resulted in each case, with the higher temperature giving slightly higher ethanol concentrations at the 5-day time point (ca. 19 g/L ethanol compared to ca. 15 g/L at 15 °C: Supplementary Fig. 1 ). However, over longer timescales, from day 11 onwards, the 15 °C fermentations generated between 6 and 30 % more ethanol than fermentations at 30 °C. There could perhaps have been a slight increase in the rate and degree of evaporative loss of ethanol at 30 °C relative to 15 °C, which went some way to explaining this. Overall, these results indicated that the process has the potential to function effectively over a wide temperature range, without close regulation of temperature (potentially obviating the need for heating or cooling control). The process might therefore be carried out under ambient conditions in many countries. Interestingly, thermochemical pre-treatment of BSG (1 % HCl, 121 °C, 30 min) reduced ethanol yields in the sake-fermentation system (Fig. 1 a). Just 12 g/L ethanol was attained after 20-day incubation, representing around one third of the yield from non-pre-treated BSG. This system also took longer to achieve maximal ethanol concentrations (around 20 days, as opposed 10 days for the non-pre-treated BSG). Moderate cellulase activity was the only enzyme activity detected in the supernatants when using pre-treated BSG. The absence of any xylanase activity was not surprising, as the acid catalysed hydrothermal pre-treatment was likely to have solubilised the majority of hemicellulose present in the starting material, leaving cellulose as the only structural polysaccharide present in significant quantity, as suggested by Wilkinson et al. [ 19 ]. However, the degree of cellulase activity observed in the supernatants when using pre-treated BSG was lower than that observed with non-pre-treated BSG, which suggests that the original lignocellulosic material is a better activator of cellulase enzyme secretion from A. oryzae . Assumedly, the native, non-modified structure of the lignocellulose is more successfully recognised by A. oryzae , or results in more successful enzyme-substrate interactions, or both. In addition, the recognition of other lignocellulosic components such as hemicellulose might be crucial in triggering optimal secretion of cellulases through some form of signalling. This could be similar to the importance of the XLR-1 xylan degradation regulator (found in an alternative fungal species Neurospora crassa ) with regard to its requirement of induction of other cellulolytic encoding genes [ 5 ]. Retaining the A. oryzae fungal species but substituting the original S. cerevisiae NCYC479 yeast strain partner with the NCYC2592 strain resulted in maximal ethanol concentrations of only ca. 20 g/L after 15 days (Fig. 1 a). This equated to ca. 45 % lower maximal ethanol yields than when using the NCYC479 yeast. Whilst moderate cellulase and xylanase enzyme activities were detected in the supernatants at various time points, these were determined to be less than were observed with S. cerevisiae NCYC479, which might explain the lower observed ethanol yield. The impact of adding cellulolytic enzymes (Cellic® CTec2, 10 FPU/g biomass) to the Sake fermentation system in an attempt to increase ethanol yield and productivity rate was investigated (Fig. 1 b). However, no increase in ethanol yield was achieved with addition of supplementary enzymes (compare Fig. 1 a, b), which suggests that cellulolytic enzyme secretion by A. oryzae was not a rate-limiting factor in terms of ethanol productivity by S. cerevisiae . It may be that the actual activity of endogenous secreted enzymes is mass transfer limited, for example due to the relatively high viscosity of the media. Operation of the sake-based CBP system at a lower solids loading (lower solid to liquid ratio) or the use of stirred or mixed fermentation vessels could possibly result in higher ethanol yields, by increasing successful enzyme-substrate interactions. Paradoxically, however, enzyme secretion has been shown to be greater on solid-state cultures as compared with highly dilute liquid cultures [ 27 ]. This could perhaps relate to a starvation response, as seen in biofilm formation. Once again, lower ethanol yields were achieved from fermentations at 30 °C compared to 15 °C (even with the additional cellulolytic enzyme supplementation; Fig. 1 b). Cellic® CTec2 displays optimum activity in the region of 50 °C [ 28 ]; hence, the higher temperature fermentation might have been expected to perform better, from that perspective. That this was not the case once again suggests that the enzyme concentration was not the rate-limiting factor in these fermentations. It has previously been reported that temperature reduction can play a role in maintaining continued expression of extracellular hydrolases when using A. oryzae in a solid-state cultivation system for the traditional manufacture of products such as soy sauce [ 27 ]. The hypothesis, which may help to explain the results observed here, is that reduced mobility of the secreted enzymes in solid-state cultures at lower temperatures may significantly reduce product feedback inhibition of the expression of hydrolases, therefore resulting in higher secreted enzyme concentrations. Analysis of xylose in the supernatants from both the 15 and 30 °C enzyme-supplemented sake systems indicated a significantly higher concentration (up to 5 g/L was present), as compared to the equivalent non-enzyme supplemented fermentations (up to 1 g/L; supplementary data Fig. 2 ). Hence, the xylanases present in the Novozymes Cellic® CTec2 hydrolysed an additional proportion of the hemicellulose present in the BSG; however, this increased hemicellulose hydrolysis did not significantly improve the final ethanol yields achieved. Evaluation of Kluyveromyces Yeast Strains in CBP of BSG for Ethanol Production Whilst many different permutations were tested (Fig. 1 c, d; Table 2 ), the use of Kluyveromyces spp., either alone with A. oryzae (i.e. without S. cerevisiae ) or in a triple consortium with both the A. oryzae and S. cerevisiae NCYC479, resulted in all cases in lower ethanol yields than the standard sake-based system of A. oryzae with S. cerevisiae NCYC479 alone. All three Kluyveromyces species ( K . marxianus NCYC1426, K . marxianus NCYC179 and K . wickerhamii NCYC546) performed similarly in terms of ethanol yields (Figs. 1 c, d and Supplementary Fig. 3 ). In an attempt to further optimise the fermentation of C-5 sugars by Kluyveromyces spp., experiments were run using the standard sake-CBP system to ferment C-6 sugars up until day 10, after which Cellic CTec2 (10 FPU/g biomass) and Kluyveromyces species were added under aerobic conditions; this approach produced less than 35 g/L ethanol (Fig. 1 c). These yields were again inferior to the standard sake-based CBP system. Purely Kluyveromyces -based CBP combinations with A. oryzae produced ca. 15–20 g/L maximal ethanol concentrations (Fig. 1 d). Supplementation with Cellic® CTec 2 (using only Kluyveromyces yeast strains) once again failed to increase ethanol yields relative to non-supplemented systems, further supporting the earlier observation that enzyme secretion by A. oryzae did not appear to be rate-limiting (Supplementary Fig. 3 ). CBP Using a Consortia of H. insolens , A. oryzae and S. cerevisiae NCYC479 (Hybrid Sake-Based System) In these experiments, H. insolens was used in addition to the sake fermentation system, to see whether its enzyme secretion could boost ethanol production. However, ethanol yields were broadly similar at most time points (ca. 25–35 g/L ethanol at either 15 °C or 30 °C; Fig. 2 a) to those for the standard sake system, indicating that the presence of the H. insolens did not significantly improve ethanol productivity. However, there was evidence of faster initial production of ethanol in this system. At the 5-day time-point, there was a small increase in ethanol concentration in the presence of H. insolens (16 and 36 % higher ethanol yields for the 15 and 30 °C fermentations, respectively) compared to the standard sake system. This might be due to the thermophilic nature of H. insolens (45 °C optimal) and production of additional deconstructive enzymes. For any commercial application of this fermentation system, an evaluation would need to be made to establish whether the day 5 ethanol yield improvement (specifically the volumetric productivity) was sufficient to justify inclusion of the H. insolens . Fig. 2 Time course of ethanol production from BSG using consolidated bioprocessing with fungal consortia under the specified conditions (see data series legends) a at both 15 and 30 °C using a consortium of H. insolens , A. oryzae and S. cerevisiae NCYC479. b At 30 °C using a consortium of H. insolens and two individual strains of S. cerevisiae (NCYC2592 or NCYC479) and c qRT-PCR analysis of gene expression levels versus time (relative to ACT housekeeping gene) for seven A. oryzae target genes (CAZy and associated genes used to indicate carbon source utilisation) in the sake CBP system grown on BSG \n CBP Using A. niger with either S. cerevisiae NCYC2592 or NCYC479 \n A. niger is in theory an excellent CBP primary fungal candidate since it possesses a large arsenal of lignocellulose degrading enzymes [ 29 ] and is used in commercial biotechnology to produce enzymes due to its high capacity secretory system [ 30 ]. It was therefore expected to perform well in terms of the production of relatively high concentrations of ethanol when partnered with a suitable S. cerevisiae strain (i.e. either NCYC2592 or NCYC479). However, only partnership with NCYC479 generated significant amounts of ethanol (maximum yield 6.7 g/L after just 6 days; supplementary Fig. 4 ), and both systems were much lower yielding than the sake system at equivalent time-points. Only a moderate degree of activity of both cellulases and xylanases was present in the supernatant from the NCYC479 experiments. CBP Using H. insolens with either S. cerevisiae NCYC2592 or NCYC479 Both secondary yeast strains ( S. cerevisiae NCYC2592 & NCYC479) in partnership with H. insolens successfully produced ethanol directly from BSG and with strain NCYC479 again significantly out-performing NCYC2592 (maximal ethanol concentration of ca. 20 g/L by day 12 compared to only ca. 8.5 g/L at the equivalent stage with NCYC2592; Fig. 2 b). Semi-quantitative analysis of enzyme activity indicated a greater activity of both cellulase and in particular xylanase activity in the supernatant produced using NCYC479, as compared to NCYC2592. In addition to the higher ethanol yields which were achieved with the NCYC479 yeast strain, a significant increase in filamentous fungal biomass production was noted (5.6 g ± 1.2 g of H. insolens fungal biomass (dry weight) compared to only 1.3 g ± 0.4 g biomass in the NCYC2592 system). This large increase in fungal biomass, apparently purely in response to the inclusion of a different yeast partner, is not readily explained. Perhaps, the NCYC479 strain did not deplete (and therefore rate limit) a particular micronutrient that was essential for fungal biomass generation. QRT-PCR Analysis of CAZy Gene Expression in A. oryzae when Using the Sake CBP System for Ethanol Production from BSG Amongst the first genes monitored which were up-regulated was the xylanase (AO090005001210) which peaked in activity at around day 5, beyond which its expression began to decline (Fig. 2 c). This suggests that the A. oryzae targeted the hemicellulose (xylan) for degradation first, possibly due to its lower recalcitrance compared to that of crystalline cellulose or lignin. This is similar to the pattern (expression sequence of CAZy genes) seen by Delmas et al. [ 31 ], who used next-generation RNA sequencing technology (RNA-seq) in a study with A. niger cultured on wheat straw. It seems that in the absence of a preferred carbon source (glucose or starch), both A. oryza e and A. niger preferentially degrade hemicelluloses, perhaps as a pre-requisite of breaking down the lignocellulosic structure. Some degree of up-regulation of the ferulic acid esterase gene ( faeB : AO090001000066) was indicated at an early stage, peaking at ca. day 5. Either the liberated hydroxycinnamic acids (predominantly ferulic acid) were being used by A. oryzae as a carbon source, or alternatively, cleavage of the di- and tri-ferulic acid esterified cross linkages was a key step in breaking down the lignocellulosic matrix and thus improving the access or performance of other secreted CAZy enzymes. An increase in the expression of the endo-cellulase gene (AO090038000175) was then observed, peaking at ca. day 10. An endo-cellulase would likely be required in the early stages of cellulose degradation to cleave internal sites along the cellulose fibres and expose free reducing ends for exo-cellulases to attack, releasing lower molecular weight cello-oligomers (e.g. cellobiose) which in turn could be further depolymerised to glucose. This hypothesis was further supported by a similar degree of up-regulation of another gene with predicted cellulase activity (AO090005001553) from day 5 onwards as was seen with the endo-cellulase. From approximately day 5, a steady increase in the expression of a β-glucosidase gene (AO090113000148) was observed. This enzyme would act on the non-reducing ends of substrates created by endo-cellulase activities. Expression levels of the acetyl xylan esterase gene ( axeA : AO090011000745) were also noted to increase steadily from day 5 onwards. This could indicate the increased expression of ‘scoping’ enzymes with potential to act on substrates present within the BSG (e.g. the side chain decorations of hemicelluloses) and thus liberate an additional metabolic carbon source. Perhaps at around this time, the A. oryzae was beginning to deplete all of the readily accessible carbon sources and a true starvation response was commencing. This may suggest that at this point, gluconeogenesis was occurring through the breakdown of less energetically favourable substrates such as the hordein-prolamin glycoproteins or their constituent glucogenic amino acids such as proline [ 32 ] or lignin. However, gluconeogenesis alone was not considered likely to explain the high ethanol yields seen with the best performing CBP permutations. Expression levels of the glucoamylase gene ( glaA : AO090010000746) remained constant throughout the duration of the experiment suggesting that the A. oryzae was either not utilising the trace starch component of the BSG (as increased α-glycosidic bond hydrolysis would have indicated) or that it was occurring at a continually low, steady state. CBP Fermentation Optimisation Since all experimental runs were conducted using static, semi-solid-state bioreactors (with high initial viscosity of the media), additional mixing could be employed in the future in an attempt to enhance fungal growth (and thus achieve greater enzyme secretion). This in turn might facilitate better enzyme-substrate interactions, potentially yielding greater fermentable sugar yields. The low solubility of oxygen (in water) relative to other dissolved solutes can limit aerobic fungal growth through limited oxygen mass transfer (especially in high dissolved solids bioreactors such as those employed here) and thus limit the secretion of various lignocellulolytic enzymes [ 33 ]. In addition, the mycelial growth of any filamentous fungal species in a solid state or high solids CBP reactor may increase the viscosity of the supernatant which may further limit oxygen mass transfer. Thus, the anaerobic environmental conditions generated by the sake CBP system (with either of the yeast strain variants tested) could have limited maximal ethanol yields, by restricting fungal growth (and thus activity of saccharification enzymes). However, the issue is likely to be considerably more complex and dynamic than can be explained by one single factor. Furthermore, both fungal hyphae and cellulases are sensitive to shear stresses, rendering agitation-based improvements in oxygen mass transfer challenging [ 33 ]. Micro-bubble-based dispersion systems could possibly be used, as opposed to conventional mixing protocols. Ideally, a dissolved oxygen concentration (DO 2 ) of ≥20 % air saturation would be sufficient [ 34 ]. If one were to consider application of the sake CBP system for bioethanol production from BSG, it is of interest that other applications of A. oryzae might be developed. Since A. oryzae is able to decompose biodegradable plastic, such as poly-butylene succinate (PBS; [ 35 ]), it could perhaps be used in the recycling of biodegradable bottles. A. oryzae has both cutinase ( CutL1 ) and hydrophobin genes (such as rolA ) within its genome which have been shown to be responsible for the deconstructive mechanism. Cutinase facilitates the actual decomposition of the plastic, with hydrophobins acting as ‘scaffolding’ for the specific site recruitment of the cutinase onto the hydrophobic surfaces of biodegradable plastics [ 27 , 35 ]."
} | 8,237 |
24098303 | PMC3787543 | pmc | 8,145 | {
"abstract": "Nitrogen, particularly nitrate is an important yield determinant for crops. However, current agricultural practice with excessive fertilizer usage has detrimental effects on the environment. Therefore, legumes have been suggested as a sustainable alternative for replenishing soil nitrogen. Legumes can uniquely form nitrogen-fixing nodules through symbiotic interaction with specialized soil bacteria. Legumes possess a highly plastic root system which modulates its architecture according to the nitrogen availability in the soil. Understanding how legumes regulate root development in response to nitrogen availability is an important step to improving root architecture. The nitrogen-mediated root development pathway starts with sensing soil nitrogen level followed by subsequent signal transduction pathways involving phytohormones, microRNAs and regulatory peptides that collectively modulate the growth and shape of the root system. This review focuses on the current understanding of nitrogen-mediated legume root architecture including local and systemic regulations by different N-sources and the modulations by phytohormones and small regulatory molecules.",
"conclusion": "CONCLUSION Nitrogen is an essential nutrient for plant productivity and its environmental availability strongly regulates root architecture. To optimize N acquisition, nitrate and ammonium transporters/transceptors provide the sensory components of N-mediated root development in legumes. The signal for N-availability is translated into an array of phytohormone pathways which regulates root development. Small regulatory molecules such as microRNAs and peptides provide further fine-tuning of these phytohormone signals to produce highly dynamic and plastic root responses to N-levels. Hence these regulatory pathways, which integrate environmental sensory signals with the modulation of phytohormones and small regulatory molecules could be exploited to improve legume root architecture for better NUE.",
"introduction": "INTRODUCTION Understanding how plants grow and develop under diverse environmental conditions is crucial for improving crop productivity. As plants are sessile, they are highly sensitive to the environment and respond accordingly for growth and survival. Of particular importance is nitrogen (N) which provides the building blocks for protein production in plants and dictates crop yield and productivity. The root system adapts to soil N-levels by modifying its architecture ( Hodge, 2006 ). In legumes, during N-limitation, specialized root organs called nodules, can form through symbiotic interaction with rhizobia which are specialized nitrogen-fixing bacteria. Rhizobia convert atmospheric N 2 to ammonium to provide legumes with N for growth. Some of this fixed N is recycled back into the soil to sustain subsequent plant growth. Due to this ability, legumes are used as rotational or cover crops to replenish soil N ( Collette et al., 2011 ). As legume root architecture is strongly regulated by N, understanding N-regulation of root development has great agricultural importance. The recent discovery of small regulatory molecules such as microRNAs and regulatory peptides provide additional facets to the classic phytohormone mediated pathways of root development. Therefore this review aims to give a brief perspective on the current knowledge of the signaling components involved in N-mediated root architecture with emphasis on the legume root system."
} | 859 |
24656471 | null | s2 | 8,146 | {
"abstract": "To characterize the microstructure and determine some mechanical properties of a polymer-infiltrated ceramic-network (PICN) material (Vita Enamic, Vita Zahnfabrik) available for CAD-CAM systems. Specimens were fabricated to perform quantitative and qualitative analyses of the material's microstructure and to determine the fracture toughness (KIc), density (ρ), Poisson's ratio (ν) and Young's modulus (E). KIc was determined using V-notched specimens and the short beam toughness method, where bar-shaped specimens were notched and 3-point loaded to fracture. ρ was calculated using Archimedes principle, and ν and E were measured using an ultrasonic thickness gauge with a combination of a pulse generator and an oscilloscope. Microstructural analyses showed a ceramic- and a polymer-based interpenetrating network. Mean and standard deviation values for the properties evaluated were: KIc=1.09±0.05MPam(1/2), ρ=2.09±0.01g/cm(3), ν=0.23±0.002 and E=37.95±0.34GPa. The PICN material showed mechanical properties between porcelains and resin-based composites, reflecting its microstructural components."
} | 275 |
17123434 | PMC1664590 | pmc | 8,147 | {
"abstract": "Background The availability of genome sequences for many organisms enabled the reconstruction of several genome-scale metabolic network models. Currently, significant efforts are put into the automated reconstruction of such models. For this, several computational tools have been developed that particularly assist in identifying and compiling the organism-specific lists of metabolic reactions. In contrast, the last step of the model reconstruction process, which is the definition of the thermodynamic constraints in terms of reaction directionalities, still needs to be done manually. No computational method exists that allows for an automated and systematic assignment of reaction directions in genome-scale models. Results We present an algorithm that – based on thermodynamics, network topology and heuristic rules – automatically assigns reaction directions in metabolic models such that the reaction network is thermodynamically feasible with respect to the production of energy equivalents. It first exploits all available experimentally derived Gibbs energies of formation to identify irreversible reactions. As these thermodynamic data are not available for all metabolites, in a next step, further reaction directions are assigned on the basis of network topology considerations and thermodynamics-based heuristic rules. Briefly, the algorithm identifies reaction subsets from the metabolic network that are able to convert low-energy co-substrates into their high-energy counterparts and thus net produce energy. Our algorithm aims at disabling such thermodynamically infeasible cyclic operation of reaction subnetworks by assigning reaction directions based on a set of thermodynamics-derived heuristic rules. We demonstrate our algorithm on a genome-scale metabolic model of E. coli . The introduced systematic direction assignment yielded 130 irreversible reactions (out of 920 total reactions), which corresponds to about 70% of all irreversible reactions that are required to disable thermodynamically infeasible energy production. Conclusion Although not being fully comprehensive, our algorithm for systematic reaction direction assignment could define a significant number of irreversible reactions automatically with low computational effort. We envision that the presented algorithm is a valuable part of a computational framework that assists the automated reconstruction of genome-scale metabolic models.",
"conclusion": "Conclusion This paper reports on a computational framework that – based on thermodynamic principles – systematically assigns reaction directionalities in genome-scale stoichiometric metabolic models. We demonstrated its application on a metabolic reconstruction of E. coli . After having exploited all available thermodynamic data to define irreversible reactions, we drew on network topology and thermodynamic heuristics to assign further reaction directions: Energy producing cycles were extracted from the reaction network and thermodynamically infeasible reaction steps that produce high-energy from low-energy co-substrates were disabled. The proposed direction assignment procedure has several advantages over other approaches. The group contribution method to computationally estimate the Gibbs energies of formation is associated with such large uncertainties that only five reactions could be identified as irreversible in a genome-scale model [ 18 ]. The method developed by Beard and co-workers for ab initio prediction of reaction directions [ 20 ] relies on the availability of all possible cycles in the metabolic network. Currently, these cannot be calculated with reasonable computational effort for genome-scale models and the method also does not completely disable thermodynamically infeasible cycling. In contrast, using our algorithm, we demonstrated that a large number of assignments could be made without laborious calculations: A total of 130 directions could be assigned automatically, which constitutes a large fraction of the direction assignments necessary to exclude thermodynamically infeasible energy production. Along with the development of mathematical methods that employ genome-scale metabolic models, these models became valuable tools in systems biology and metabolic engineering. Here, our systematic assignment procedure can be used in the reconstruction of new models or in the revision of existing ones. Currently, large efforts are put into the automated reconstruction of such models [ 10 , 29 ] and several computational tools exist that support the first steps of the reconstruction process [ 11 , 30 ]. On the contrary, the following steps towards finalizing the model – which include the definition of reaction directionalities – are still done manually. We envision that the here proposed algorithm could be a valuable part of a computational framework that assists the automated reconstruction process for genome-scale metabolic models.",
"discussion": "Discussion Achieved direction assignment Table 2 summarizes all assignments that were made by our systematic procedure. While the thermodynamic facts-based assignment yielded 43 irreversible reactions, 87 further reaction directions were assigned based on network topology and thermodynamic heuristics. Altogether, 130 reactions were restricted to one direction, which disabled the operation of 129 of the 145 energy producing cycles present in the employed null space matrix. Table 2 Overview over the number of direction assignments made in each step assignment step analysis of ... number of assigned directions in the respective step in total thermodynamic facts 43 43 thermodynamic heuristics pair cycles 42 85 remaining energy producing cycles 21 106 bypasses 24 130 Our algorithm did not completely disable thermodynamically infeasible energy production: The heuristics failed in blocking all energy producing cycles and the bypass analysis was not able to identify all possible energy producing cycles. In order to assess the completeness achieved with our approach, we estimated how many additional direction assignments had to be made to completely prohibit infeasible co-substrate conversion. For this, an iterative procedure was applied: A possible energy producing cycle was identified using flux balance analysis, and then, reaction directions were assigned manually to block this cycle (cf. Methods section). When no further energy producing cycles were found, the reactions' directionalities were assumed to reflect thermodynamic feasibility with respect to energy generation. At this point, the direction assignment was considered to be complete. With this procedure, 59 additional assignments of reaction directions were required until infeasible energy production was excluded. Simulating aerobic growth on glucose by flux balance analysis, ATP was then produced via the respiratory chain. Importantly, the production of energy equivalents such as ATP by metabolic reactions was not generally rendered impossible by our linear constraints as our algorithm only selectively disables the generation of highly energetic co-substrates. In summary, the 189 irreversible reactions (of which 130 were assigned by our algorithm) were sufficient to yield a thermodynamically reasonable model with respect to the production of energy equivalents. At this point, we checked whether the application of general biochemical rules such as defining all kinase reactions as irreversible would have been a much simpler and also valid alternative to our approach. A close inspection of the 74 kinase reactions in the model revealed that this would not had resulted in a correct model: For instance, the phosphoglycerate kinase reaction is known to operate in both directions and it is correctly defined as reversible in our assignment. This demonstrates that employing heuristic rules in combination with analyzing co-substrate converting cycles is superior to simple general biochemical rules. With the model analyzed here, the calculation time required for the assignment procedure was roughly two minutes on a Pentium 3 GHz PC, if the calculation of the null space matrix and generation of a Excel file for output documentation is included. The assignment algorithm itself required about 30 to 40 s. Such, the computational effort is small and the algorithm can be efficiently executed on a usual PC. Comparison to original model The introduced systematic direction assignment yielded 130 reactions that were restricted in one direction. Together with the 59 manual assignments that eventually eliminated any thermodynamically infeasible cycling, we obtained 189 reactions that are irreversible in our model. In comparison to the 676 irreversible reactions in the original model from Palsson and co-workers [ 4 ], this is a rather small number and indicates a much less constrained model. From a constraint-based modeling viewpoint, a direct comparison of the number of irreversible reactions, however, is misleading as one assigned reaction direction can practically render impossible the reversible operation for a set of other reactions. For example, one irreversible reaction that is part of an unbranched linear pathway restricts the operation of the whole pathway to one direction. Hence, in effect it is no difference if the direction of only one or all reactions of the pathway are defined as irreversible. To allow for assessment of model flexibility due to different direction assignments, we had to identify correlated sets of reactions (cf. Methods). Using the identified correlated sets, the number of de facto irreversible reactions was assessed. We found that the stoichiometric network of E. coli comprises 175 sets of correlated reactions. If one reaction in such a set is defined as irreversible, mass balance constraints rule out one particular direction for each of the other reactions in the set. In the original model, 749 reactions are practically irreversible. In comparison, our direction assignment eventually resulted in 292 reactions that practically can operate only in one direction. We found that only in one case – namely the UTP-glucose-1-phosphate uridylyltransferase reaction – our algorithm defines a reaction as irreversible which is reversible in the original model. Remarkably, our assignment is in agreement with the EcoCyc database [ 27 ] which also states that this reaction is irreversible. As the predicted maximal biomass yield on glucose is increased by about 20% using our reaction directions in comparison to the original, the model with our reaction directions is much less constrained and there are more possibilities to distribute the mass flux through the reaction network. Therefore, it is envisioned that it covers a larger range of metabolic scenarios, e.g. knockout mutants or different environmental conditions. As an example, a frdA deletion mutant ( in vivo viable when grown anaerobically on glucose [ 28 ]) is in silico nonviable with the original reaction directions while it is viable with our reaction directions. Extension of heuristic rules Next, we evaluated whether we could complement the employed heuristic assignment rules to increase the number of reactions that are automatically defined as irreversible. Additional or modified heuristic rules should eliminate the energy producing cycles that were not yet disabled by our algorithm. First, we closely inspected the additional manual direction assigments that were required to eliminate all the remaining energy producing cycles (cf. Additional file 1 ). In this reaction set, we found reactions, which potentially could have been made irreversible by the heuristics already used in the algorithm, i.e. reactions that produce/consume high-energy/low-energy co-substrates, but for several reasons (as outlined above), the respective directions were not assigned. There are, however, groups of reactions (e.g. quinone pool reducing/oxidizing reactions) whose common attributes could be exploited by new heuristics that specifically assign directions to such sets of reactions (cf. Table 3 ). Table 3 Number of additional direction assignments required to eliminate remaining thermodynamically infeasible energy production common attributes standard procedure standard procedure with consideration of final electron acceptors quinone pool reductions 15 - transporters 9 8 NTP production 14 12 NADH/NADPH production 5 4 O 2 production 2 1 CO 2 consumption 4 4 NMP synthesis 7 7 other 3 3 sum 59 39 As an example for such an extension of the heuristic rules, the quinone pool converting reactions were set as irreversible such that the electrons are transferred from the reduced metabolites to the final electron acceptors. Having defined the final electron acceptors, it was possible to assign 43 reaction directions in the E. coli model. When we incorporated this rule into the assignment algorithm, in total 26 more reactions were restricted in one direction (cf. Fig. 4 ). Fourteen out of the 43 reactions had been already assigned by the thermodynamic facts, and the bypass analysis assigned three reactions less. In summary, 156 instead of 130 reactions could then be defined as irreversible by our systematic assignment procedure. Figure 4 Comparison of the assignment where final electron acceptors are considered to the default assignment . The numbers of made direction assignments of the standard assignment procedure (A) and the assignment procedure, which additionally drew on the direction of electron transfer within the respiratory chain (B) are compared. The numbers (1 – 2c) refer to the assignment steps depicted in Fig. 1, while step 0 represents the reaction directions that were assigned by the additional heuristic rule based on final electron acceptors. The extension of heuristic rules by organism-specific knowledge obviously is an effective and effortless approach to increase the number of assigned directions. Similarily, one could define the directions of the transporters according to their function, which often can be identified from stoichiometry alone (e.g. sugars are taken up by PTS systems)."
} | 3,522 |
23918042 | PMC3734524 | pmc | 8,149 | {
"abstract": "Materials with both high strength and toughness are in great demand for a wide range of applications, requiring strict design of ingredients and hierarchically ordered architecture from nano- to macro-scale. Nacre achieves such a target in the long natural evolution by alternative alignment of inorganic nanoplatelets and biomacromolecules. To mimic nacre, various strategies were developed, approaching nacre-comparable performance in limited size. How to remarkably exceed nacre in both property and size is a key issue to further the advancement of composites. Here we present liquid crystal self-templating methodology to make the next generation of ultrastrong and tough nacre-mimics continuously. The hierarchically assembled composites show the highest tensile strength (652 MPa) among nacre mimics, five to eight times as high as that of nacre (80–135 MPa), and excellent ductility with toughness of 18 MJ m −3 , one to two orders of magnitude greater than that of nacre (0.1 ~ 1.8 MJ m −3 ).",
"discussion": "Discussion Compared with previous artificial nacre composites, GGO-HPG fibres show great enhancement in both σ and ε, mainly because the chosen GGO building blocks have extraordinary properties that can be translated into the macroscopic GGO-HPG fibres by effective assembly. The large size of GGO minimizes the defects in our composites. The combination of LCST strategy and wet-spinning process provides effective assembly and ordered alignment of building blocks. The flexible feature of GGO generates wrinkles and waviness during the spinning process, which are favorable to improve the strength. The formation of strong and adaptive hydrogen bonding networks between GGO and HPG also contributes to the super properties. As the adaptive character of hydrogen bonding network, the destructed hydrogen bondings can be recovered for a number of cycles, providing our composites good ductility 54 . Moreover, the uniform monolayer of HPG molecules maximizes the quantity of hydrogen bondings, which can repair some defects and further improve the mechanical strength of our composites. For the neat GGO system, three kinds of dominant interactions between individual GGO sheets, van der Waals interactions, hydrogen bondings, and ionic interactions ( Supplementary Fig. S16 ), contribute to its mechanical strength (σ 345 MPa). Considering that hydrogen bondings need oxygen containing groups, HPG with abundant hydroxyl groups were introduced to increase the density of hydrogen bondings at both interlayer and intralayer ( Supplementary Fig. S6 ), further improving σ to 555 MPa. The supramolecular interactions were successfully converted into stronger covalent linkages by cross-linking with GA, resulting in higher mechanical properties (σ 652 MPa, E 20.9 GPa). After chemical reduction, the RGG-HPG composites mostly maintained good performance from the original GGO-HPG fibres ( Table 2 ) due to the increased van der Waals interactions between graphene sheets ( Supplementary Fig. S11 ), slightly decreased hydrogen bondings and ionic interactions ( Supplementary Fig. S6 ), as well as the hierarchical structure of composites. In GGO-HPG composites, the large size and waviness of GGO sheets and the dendritic binder structure of HPG give the primary structure of fibres; the high density of hydrogen bondings between HPG and GGO provide the secondary structure; the regular and compact cross-section derived from the LC structures of GGO-HPG aqueous solution and effective assembly by wet-spinning form the tertiary structure; the alignment and wrinkles (formed in the coagulation procedure) of GGO sheets along the fibre axis make up the quarternary structure. These four hierarchical levels of GGO-HPG fibres not only provide high strength and toughness of original composites, but also are responsible for the comparable mechanical performance of RGG-HPG composites since such hierarchical structures remained after reduction. Although most of RGG-HPG fibres obtained by different methods remain excellent properties, the stress-strain curves are distinct. Thermal treatment reserved compact cross-section structure of GGO-HPG ( Supplementary Fig. S13 ), resulting in comparable mechanical performance. The chemical reduction of GGO-HPG fibres may create new products and change the compositions of RGG-HPG fibres, which differ from each other according to the reducing agents 46 47 48 , thus the stress-strain curves are sensitive to the reduction methods. Notably, the hydrazine reduction process may generate gas, which results in porous structure of RGG-HPG fibres ( Supplementary Fig. S13 ), seriously decreasing the mechanical properties. A typical shear lag model has been proposed to illustrate the strength of nacre. Based on this model, two fracture types of layered composites are found, according to the aspect ratio (S) of platelets and the critical one (S c ). Generally, S c is equal to the ratio of tensile strength of platelets (σ p ) and shear strength of organic matrix (τ y ) (S c = σ p /τ y ). For S < S c , the relatively weak organic matrix and the interface of platelet-organic matrix yield the tensile strength, and thus the composites break by pull-out of platelets. In the case of S > S c , the composites fail due to the fracture of platelets. The later case usually gives stronger but brittler materials. In our composites, S c is around 10 4 , close to the aspect ratio of part GGO sheets (S ~ 1.6 × 10 4 ). Therefore, we observed the fracture of GGO sheets in GGO-HPG composites ( Fig. 4b,c ). As the GGO sheets have a wide distribution of width, some GGO platelets pulled out accompanying the rupture of composites ( Fig. 4d,e ). Moreover, the incorporation of adaptive hydrogen bondings between GGO and HPG further provided the composites ductile attribute, resulting in high ε and toughness. In summary, we designed a green, simple, general, fast and efficient LCST strategy to fabricate the next generation of continuous, ultrastrong and tough bio-mimetic composites. Despite of the mechanical blending process, phase separation between nanofillers and polymer was avoided, even at the case of high content of nanoparticles (~ 77 wt%). The uniform, regular arrangement and the high aspect ratio of GGO sheets together with the hierarchical structures of our composites led to a large improvement of mechanical performance for nanocomposites. The biomimetic composites set a new record σ (~ 0.65 GPa) and toughness (~ 18 MJ m −3 ) among nacre mimics. In addition, the composites exhibited high electrical conductivity (5261 S m −1 ) that was comparable to neat graphene papers. Such multifunctional composites have wide applications in functional textiles as well as flexible and wearable devices. The LCST strategy can be readily extended to prepare other hierarchically structured composites that can be hardly accessed by previous protocols, opening the avenue to multifunctional, highly ordered and tailor-made materials."
} | 1,744 |
34831448 | PMC8621022 | pmc | 8,150 | {
"abstract": "Antibiotics are well known drugs which, when present above certain concentrations, are able to inhibit the growth of certain bacteria. However, a growing body of evidence shows that even when present at lower doses (subMIC, for sub-minimal inhibitory concentration), unable to inhibit or affect microbial growth, antibiotics work as signaling molecules, affect gene expression and trigger important bacterial stress responses. However, how subMIC antibiotic signaling interplays with other well-known signaling networks in bacteria (and the consequences of such interplay) is not well understood. In this work, through transcriptomic and genetic approaches, we have explored how quorum-sensing (QS) proficiency of V. cholerae affects this pathogen’s response to subMIC doses of the aminoglycoside tobramycin (TOB). We show that the transcriptomic signature of V. cholerae in response to subMIC TOB depends highly on the presence of QS master regulator HapR. In parallel, we show that subMIC doses of TOB are able to negatively interfere with the AI-2/LuxS QS network of V. cholerae , which seems critical for survival to aminoglycoside treatment and TOB-mediated induction of SOS response in this species. This interplay between QS and aminoglycosides suggests that targeting QS signaling may be a strategy to enhance aminoglycoside efficacy in V. cholerae .",
"introduction": "1. Introduction Many bacterial species secrete small diffusible signaling molecules to synchronize multicellular behaviors which allow them to adapt and survive in natural environments [ 1 , 2 ]. The most studied intercellular communication mechanism is quorum-sensing (QS), which monitors local population density [ 3 , 4 ]. QS is achieved via the production and detection of extracellular small molecules called autoinducers. At low cell density, autoinducers diffuse away, but at high cell density their concentration increases and triggers synchronization of gene expression in bacterial populations. Gram-negative bacteria are able to produce and detect several classes of autoinducers. Autoinducer 1 (AI-1) is a species-specific signaling molecule, while autoinducer 2 (AI-2), which is produced by Gram-negative and Gram-positive bacteria, is able to mediate both intra and interspecies QS communication [ 5 , 6 ]. Vibrio cholerae , the causative agent of cholera disease, produces both autoinducers. AI-1, called cholera AI-1 (CAI-1), is produced by the CqsA protein and sensed by CqsS, while AI-2 is produced by LuxS and sensed by LuxQ, via LuxP periplasmic protein. The QS regulatory network of V. cholerae relies on a well described phosphorylation cascade ( Figure 1 ). At low cell density, CqsS and LuxPQ work as kinases and phosphorylate LuxU which will then transfer the phosphate to the regulator LuxO. Phosphorylated LuxO will then trigger the expression of four small RNAs (qrr1–4) which in turn allow for translation of AphA (master regulator of low cell density), while inhibiting that of HapR, the LuxR family master regulator of high cell density in V. cholerae . By contrast, at high cell numbers, both CAI-1 and AI-2 accumulate and bind the cognate receptors CqsS and LuxPQ, which will now act as phosphatases and inhibit the phosphorylation cascade described above. This leads to an absence of qrr1–4 sRNAs and, consequently, the absence of AphA. Concomitantly, HapR is produced, inducing the expression of several genes involved in group behavior [ 7 , 8 ]. Escherichia coli can also detect autoinducers produced by other bacteria and react to them via SdiA, a LuxR protein homologue [ 9 , 10 ]. E. coli AI-2 is produced by the LuxS protein and sensed by the proteins encoded by the lsr operon [ 11 , 12 ]. Interestingly, bacterial communication through small molecule signaling can induce antibiotic tolerant phenotypes [ 13 , 14 , 15 ]. In parallel, it is also known that antibiotics at low doses can work as signaling molecules [ 16 ]. While studying the bacterial response to antibiotics, we showed that antibiotics from different families induce stress responses in Gram-negative bacteria, at concentrations below the minimal inhibitory concentration (subMIC), namely the SOS response [ 17 , 18 ]. SOS induction reflects the presence of a genotoxic stress to which the bacterial cell responds by triggering mutagenic DNA repair and recombination pathways, as well as rearrangements in the Superintegron carried by the V. cholerae ’s second chromosome, which carries antibiotic resistance and adaptation genes [ 19 , 20 ]. We have pursued with the study of the response to aminoglycosides (AGs), which is a class of antibiotics that target the ribosome and induce mistranslation [ 21 ]. The AG-mediated SOS induction that we observed in V. cholerae is conserved among distantly related Gram-negative pathogens, such as Photorhabdus luminescens and Klebsiella pneumonia [ 18 ]. This observation was puzzling, because AGs do not directly target DNA synthesis or DNA molecules. Strikingly, we observed that the induction of SOS by low doses of AGs appeared to be dependent on HapR [ 17 ], because SOS induction by the aminoglycoside tobramycin was prevented in the V. cholerae strain lacking hapR . This observation suggested that QS could play a significant role in the evolution of antibiotic resistance. We thus decided to study the impact of quorum sensing on the effect of sub-inhibitory concentrations of AGs in V . cholerae . We constructed mutants deriving from the N16961 HapR+ strain (referred to as wild-type), deleted for the genes cqsA (deficient for CAI intra-species signaling), luxS (deficient for AI-2 inter-species signaling), luxPQ (deficient for AI-2 sensing), luxO (“locked” in high cell density state) and aphA (the master regulator of low cell density). We asked which QS pathway(s) are involved in the response to sub-inhibitory concentrations of aminoglycosides, and how QS is involved in modulation of gene expression patterns by treatment with sub-inhibitory concentrations of AGs. RNA-seq performed on both QS proficient (HapR+) and QS deficient (HapR−) V. cholerae strains points to major differences on global gene expression in response to subMIC tobramycin (TOB) treatment. Moreover, transcriptomic data suggest that subMIC AG treatment may interfere with the quorum sensing pathways and lead to the activation of the AphA low cell density regulon. We find that supplementation of growth media with AI-2 alleviates SOS induction by subMIC TOB. We further show that deletion of luxS (and to a lesser degree cqsA ) is strongly detrimental for growth in presence of sublethal AGs concentrations and also for survival to lethal doses of this antibiotic family. These observations strongly suggest that QS signaling plays an important role in the response to antibiotics.",
"discussion": "4. Discussion We based this study on our previous findings that SOS is induced by subMICs of aminoglycosides when QS proficient ( hapR +) but not when QS deficient ( hapR −) V. cholerae ([ 17 ] and Supplementary Figure S2 ). The observation that the V. cholerae strain lacking HapR fails to trigger aminoglycoside-mediated SOS induction prompted us to investigate how the responses to subMIC aminoglycosides vary in different QS contexts in this species. The transcriptomic analysis of both HapR− or HapR+ cells treated with 2% MIC of tobramycin revealed substantially different gene expression profiles between the two strains, specifically regarding the expression of genes involved in translation, cell energy and sugar transport processes ( Supplementary Figure S1 ), which are known to modulate the physiological activity of aminoglycosides in bacteria [ 14 , 31 , 32 , 33 , 34 ]. Aminoglycosides are antibiotics known to target the ribosome, generating mistranslation and protein stress [ 35 ]. In agreement with this, we observed the induction of several members of the heat-shock regulon by subMIC TOB in both strains, showing that even very low concentrations of these drugs are able to generate protein stress in V. cholerae . However, the extent of this induction seems to be dependent on the QS state of the cells, as we noticed a greater induction of the heat-shock regulon in hapR+ cells, thus suggesting a link between QS and response to aminoglycoside treatment in V. cholerae . Moreover, when we treated the hapR + strain of V. cholerae with subMIC TOB we observed the upregulation of several genes whose expression is positively controlled by AphA, and the downregulation of genes known to be repressed by AphA [ 27 , 28 ] ( Figure 3 A). This suggests that subMIC tobramycin treatment mimics a state of low cell density, which is characterized by the absence (or low concentration) of autoinducers and the lack of activation of the respective QS systems. How subMIC TOB leads to the activation of the low cell density regulon is not clear, but it is possible that subMIC tobramycin interferes with one or several of these QS systems. In fact, we show that subMIC TOB seems to interfere with AI-2 QS signaling, as we observed that AI-2-dependent bioluminescence production in Photorhabdus luminescens is halted by subMIC TOB. Thus, by interfering with LuxS/AI-2 system, subMIC TOB could partially inhibit the LuxS–LuxO phosphorylation cascade and lead to an increase of AphA protein levels in the cell with the activation of AphA regulon. However, it has been suggested that the QS network of V. cholerae is quite robust, is resilient to signal perturbations by relying on four functionally redundant QS circuits [ 8 ], and that full QS network activation requires the concerted action of AI-2, CAI-1 and DPO molecules, which act together to fully repress AphA [ 30 ]. Thus, it is possible that some other additional factors, together with low AI-2 signaling, can be involved in the AphA regulon activation by subMIC TOB. Given the lack of AG-mediated SOS induction in V. cholerae hapR − ( Supplementary Figure S3 ), and the observation that AI-2 signaling seems to be affected by tobramycin ( Figure 3 B), we also sought to determine whether AG-mediated SOS induction in V. cholerae relies on this interspecies QS system. We found that subMIC TOB generates higher levels of genotoxic stress in absence of AI-2 signaling, as we observed a greater induction of SOS response in the luxS mutant ( Figure 4 ). Moreover, deficiency of LuxS is highly detrimental for V. cholerae growth in subMIC aminoglycosides ( Figure 5 ) and survival to lethal doses of these antibiotics ( Figure 6 ). The results described here suggest an interplay between aminoglycoside activity and QS in V. cholerae : on one hand, we show that subMIC TOB affects AI-2 signaling. On the other, we demonstrate the QS state of the cells (specially mediated by the luxS/AI-2 system) seems to dictate the response of V. cholerae to aminoglycosides. Several studies have demonstrated that QS signaling in bacteria often controls a multitude of processes that promote tolerance and resistance to several antibiotics. For example, expression of the MexAB-OprM efflux pump is positively controlled by QS and promotes resistance to beta-lactams in Pseudomonas aeruginosa [ 36 ]. Additionally, biofilm formation, which is critical to aminoglycoside susceptibility [ 37 ], is known to be controlled by QS [ 38 , 39 ], and it has been shown that the susceptibility of P. aeruginosa biofilms to aminoglycosides increases in presence of QS inhibitors of the LasI/LasR and RhlI/RlhR systems [ 40 ]. In addition, the AI-2/LuxS interspecies QS system has also been shown to modulate antibiotic resistance mechanisms in several species. Examples include the AI-2/LuxS—dependent upregulation of MDR efflux pumps, which promotes fluoroquinolone resistance in E. coli [ 41 ] and Streptococcus suis [ 42 ], or the AI-2/LuxS-dependent upregulation of a two-component system responsible for increasing vancomycin resistance in Staphylococcus aureus [ 43 ]. Other examples linking AI-2/LuxS QS system and drug resistance are reviewed in [ 44 ]. In V. cholerae , the molecular mechanisms behind AI-2-mediated protection against aminoglycosides remain to be elucidated. Nonetheless, such protection raises the interesting possibility that even small populations of V. cholerae , when in a high-cell-density multi-species context, can be less susceptible to aminoglycoside action. This may be of particular importance in the context of infections in the human gut, where AI-2-producing communities may help low loads of V. cholerae to survive aminoglycoside treatment. In parallel, the fact that subMIC TOB interferes with QS signaling may have important consequences in the context of infection. In fact, AI-2 seems to be the necessary signal to repress biofilm formation and induce dispersal in V. cholerae [ 45 ]. Thus, by interfering with this QS system, low doses of aminoglycosides may enhance biofilm formation and virulence of V. cholerae . Further work is necessary to uncover the mechanism by which low doses of tobramycin (and potentially aminoglycosides in general) disrupt the interspecies QS system. Given that we do not observe any effect of TOB on the transcription of QS genes in our RNAseq data, one hypothesis may be that aminoglycosides affect the correct synthesis of specific proteins involved in this system. Alternatively, aminoglycoside molecules may directly interfere with AI-2 receptors. In fact, subMIC aminoglycosides attenuate QS-mediated virulence phenotypes in P. aeruginosa , and they have been found to possess strong binding properties to the QS receptor of P. aeruginosa , LasR [ 46 ]. Overall, the results obtained here contribute to the notion that QS communication and antibiotic resistance/tolerance mechanisms are linked. A link between bacterial signaling and antibiotic tolerance was also previously shown for a different signaling system, through indole secretion [ 13 , 15 ]. Manipulation of cell-to-cell signaling may thus be a potential way to fight antimicrobial resistance."
} | 3,511 |
21414040 | null | s2 | 8,151 | {
"abstract": "Magnetotactic bacteria contain nanometre-sized, membrane-bound organelles, called magnetosomes, which are tasked with the biomineralization of small crystals of the iron oxide magnetite allowing the organism to use geomagnetic field lines for navigation. A key player in this process is the HtrA/DegP family protease MamE. In its absence, Magnetospirillum magneticum str AMB-1 is able to form magnetosome membranes but not magnetite crystals, a defect previously linked to the mislocalization of magnetosome proteins. In this work we use a directed genetic approach to find that MamE, and another predicted magnetosome-associated protease, MamO, likely function as proteases in vivo. However, as opposed to the complete loss of mamE where no biomineralization is observed, the protease-deficient variant of this protein still supports the initiation and formation of small, 20 nm-sized crystals of magnetite, too small to hold a permanent magnetic dipole moment. This analysis also reveals that MamE is a bifunctional protein with a protease-independent role in magnetosome protein localization and a protease-dependent role in maturation of small magnetite crystals. Together, these results imply the existence of a previously unrecognized 'checkpoint' in biomineralization where MamE moderates the completion of magnetite formation and thus committal to magneto-aerotaxis as the organism's dominant mode of navigating the environment."
} | 359 |
32927059 | PMC8015268 | pmc | 8,153 | {
"abstract": "Metabolic engineering has allowed the production of a diverse number of valuable chemicals using microbial organisms. Many biological challenges for improving bio-production exist which limit performance and slow the commercialization of metabolically engineered systems. Dynamic metabolic engineering is a rapidly developing field that seeks to address these challenges through the design of genetically encoded metabolic control systems which allow cells to autonomously adjust their flux in response to their external and internal metabolic state. This review first discusses theoretical works which provide mechanistic insights and design choices for dynamic control systems including two-stage, continuous, and population behavior control strategies. Next, we summarize molecular mechanisms for various sensors and actuators which enable dynamic metabolic control in microbial systems. Finally, important applications of dynamic control to the production of several metabolite products are highlighted, including fatty acids, aromatics, and terpene compounds. Altogether, this review provides a comprehensive overview of the progress, advances, and prospects in the design of dynamic control systems for improved titer, rate, and yield metrics in metabolic engineering.",
"conclusion": "5. Conclusions Dynamic control has been used for overcoming numerous challenges in metabolic engineering, including toxic metabolite accumulation, unbalanced pathway flux, and production heterogeneity, with new applications still being developed. Theoretical work has elucidated the fundamentals on the dynamics and robustness of controlled metabolic systems and has illustrated the choice of control valves and topology. Synthetic biology has provided delicate tools to construct sensors for detecting changing environments and actuators for precise control of cellular metabolism. Altogether, these models and tools have established a strong foundation for future research in dynamic metabolic engineering. In the near future, we expect this field to continue to grow, especially along the lines of solving several unaddressed challenges. Firstly, current dynamic metabolic control often relies on extensive trial-and-error for choosing topology and tuning of the control parameters, which is labor-intensive. Although several control topologies may help to improve production, the most optimal topology can be difficult to predict. In addition, improved TRY performance is only seen within a narrow range of parameter settings. Addressing these challenges requires precise control design algorithms, which connect the dynamics of control systems with overall metabolite production metrics to quickly identify the desirable control topology and parameters for experimental implementation. Second, for control systems that are not performing adequately, there is a lack of diagnostic tools for assessing dynamic problems, for example long delays, overshoot, slow dynamics, and oscillations. This lack of diagnostic tools exacerbates the trial-and-error search for good control systems. Third, although a key advantage of implementing dynamic controls is to respond to heterogeneous microenvironments and metabolite heterogeneity, models of control system often do not account for these heterogeneities. These phenomena may be important in the choice of topologies and parameters. Additionally, methods for rapidly assessing the performance of control systems when subjected to these heterogeneities are lacking. Development of models and methods to account for heterogeneity could be critical for the scale-up and commercialization of many bioprocesses, where large heterogeneities exist. Finally, as the most important component of dynamic control, sensors are still lacking for many metabolites of interest, which limits the application of dynamic control. General methods to engineer sensors specific to a metabolite of interest are strongly needed. New design strategies beyond known MRTFs, responsive promoters, and riboswitches, are particularly interesting. Solving these grand challenges will provide exciting avenues for research at the nexus of control theory, systems and synthetic biology, and metabolic engineering.",
"introduction": "1. Introduction In research laboratory settings, metabolic engineering has enabled the production of a vast array of metabolite products including fuels ( Jiang et al., 2018 , 2017 ; Yan and Pfleger, 2020 ), chemicals ( Bai et al., 2019 ; Bowen et al., 2016 ; Reed and Alper, 2018 ), medicines ( Cao et al., 2020 ), and polymer precursors ( Zheng et al., 2020 ), using dozens of microbial species from bacteria ( Becker et al., 2018 ; Pontrelli et al., 2018 ) to eukarya ( Abdel-Mawgoud et al., 2018 ; Lian et al., 2018 ) and archaea domains ( Crosby et al., 2019 ). However, commercial production of these compounds in industrial scales has been lagging, largely due to the inability of the engineered strains to maintain stable performance at large scales while meeting stringent titer, rate, and yield (TRY) requirements. Metabolic engineers have faced a myriad of challenges in forcing engineered microbes to over-produce metabolite products. Engineered metabolic pathways utilize shared host machinery, including RNA polymerases, ribosomes, ATP, cofactors, and other native metabolites. Competition on cellular resources is sensitive to fermentation conditions and can cause metabolic burden ( Kurland and Dong, 1996 ), improper cofactor balance ( Bentley et al., 2016 ; Charusanti et al., 2010 ), or accumulation of metabolites to toxic levels ( Kizer et al., 2008 ), all of which can interfere with the growth and desired metabolic objective of the engineered microbes. Additionally, cells growing in large-scale bioreactors often experience different and changing microenvironments, leading to heterogeneity in their performance ( Delvigne et al., 2014 ; Pigou and Morchain, 2015 ). These issues constrain metabolite production and give advantage to fast-growing, yet non-productive isogenic cells or mutant strains ( Rugbjerg et al., 2018a ), which ultimately lowers overall TRY performance ( Zhuang et al., 2013 ). Optimizing strain performance through design-build-test cycles is lengthy and costly. The cost of commercializing a metabolite product was estimated to range from $100 million to $1 billion ( Crater and Lievense, 2018 ; Wehrs et al., 2019 ). Dynamic metabolic engineering seeks to address these challenges through the development of genetically encoded control systems which allow microbes to autonomously adjust their metabolic flux in response to the external environment and/or internal metabolic state. The concept is inspired by natural metabolic control systems which microbes use to maintain homeostasis, coordinate metabolic flux ( Chubukov et al., 2014 ; Kochanowski et al., 2017 ), and adapt metabolism to changing environments ( Chin et al., 2008 ) and stresses ( Jozefczuk et al., 2010 ). These dynamic control systems are in contrast to static control systems traditionally used in metabolic engineering, where metabolic pathways are expressed constitutively and are tuned by the choice of promoters, ribosome binding sites, and gene copy numbers ( Holtz and Keasling, 2010 ). Since the first demonstration of enhanced lycopene production two decades ago ( Farmer and Liao, 2000 ), dynamic metabolic engineering has become a popular strategy. Recent advances in synthetic biology and systems biology have provided the tools for dynamic metabolic engineering while control theory has provided new design principles for expanding beyond natural control systems. There are multiple examples where dynamic metabolic control has provided microbes remarkable robustness in different fermentation conditions and improved TRY performance ( Anesiadis et al., 2008 ; Cress et al., 2015 ; Liu et al., 2018 ). In this work, we seek to provide a comprehensive review of the theory and practice of dynamic metabolic engineering, focusing on its improvement to TRY metrics. In the first section, we will present theoretical works that discuss the benefits of incorporating dynamic metabolic control as well as works that provide guidance on design choices for the construction of control systems. In the second section, we will review the construction and engineering of sensors and control mechanisms, referred to as actuators, the two fundamental components of dynamic metabolic control systems. Finally, we will review metabolic pathways to which dynamic metabolic engineering has been applied with the focus on control topologies."
} | 2,142 |
36516242 | PMC9750142 | pmc | 8,154 | {
"abstract": "Physical reservoir computing has recently been attracting attention for its ability to substantially reduce the computational resources required to process time series data. However, the physical reservoirs that have been reported to date have had insufficient computational capacity, and most of them have a large volume, which makes their practical application difficult. Here, we describe the development of a Li + electrolyte–based ion-gating reservoir (IGR), with ion-electron–coupled dynamics, for use in high-performance physical reservoir computing. A variety of synaptic responses were obtained in response to past experience, which were stored as transient charge density patterns in an electric double layer, at the Li + electrolyte/diamond interface. Performance for a second-order nonlinear dynamical equation task is one order of magnitude higher than memristor-based reservoirs. The edge-of-chaos state of the IGR enabled the best computational capacity. The IGR described here opens the way for high-performance and integrated neural network devices.",
"introduction": "INTRODUCTION Artificial neural network (ANN)–based information processing is becoming more and more important as a way to deal with the vast amount of information currently in existence ( 1 , 2 ). ANN computing [e.g., deep learning with a multilayer neural network (NN)] can provide excellent learning, classification, and inference characteristics that are close to, and in some cases beyond, those found in natural intelligence (i.e., the human brain), whereas the enormous amounts of power required by ANN (as in a typical multilayer NN) are far higher than that required by human beings ( 2 ). The low-energy efficiency of ANN computing is a serious drawback in the realization of ubiquitous and versatile artificial intelligence (AI) but is inherent in the structure of the ANN, which requires the weights of millions of virtual synaptic nodes to be stored and updated (i.e., large network size) and requires the consumption of tremendous amounts of energy. Reservoir computing (RC) has recently been attracting attention because of its ability to substantially reduce the computational resources required to process time series data, which it is able to do because of its utilization of the nonlinear responses of a “reservoir” to input signals. While simulated recurrent NNs have been used as reservoirs to perform fully simulated RC ( 3 – 5 ), materials or devices with nonlinearity, high dimensionality, and short-term memory have been explored as possible “physical reservoirs” that can process information without heavy computational burdens for complicated simulations of the dynamical states of a reservoir ( 6 ). To date, the nonlinear dynamics of various materials and devices (e.g., soft bodies, optical devices, spin torque oscillators, and memristors) have been reported as providing nonlinear dynamics that are sufficient to perform physical reservoir–based RC with various time series tasks, including image recognition, spoken digit classification, and combinatorial optimization ( 6 – 24 ). However, to date, the performance of physical reservoir–based RC has been far from satisfactory because of the low expression power of physical reservoirs in comparison to the RC performance of simulated reservoirs. Furthermore, most of the high-performance physical reservoirs have large volumes, more than several cubic centimeters, which are not realistic choices for practical application to integrated AI devices ( 7 – 9 ). Therefore, achieving compatibility between (i) the high expression power of a physical reservoir and (ii) small reservoir volume is a great challenge in nanotechnology research leading toward the physical implementation of RC at practical levels. Here, we report the achievement of high-performance physical RC using an ion-gating reservoir (IGR), in which ion-electron–coupled dynamics at a lithium ion electrolyte/diamond interface generate an “edge-of-chaos” state, which is empirically known to exhibit high computational performance ( 25 ). Various synaptic responses, with asymmetric relaxation and spikes, were obtained with respect to the input history of a single IGR transistor (IGRT), which operates in an electric double layer (EDL) mechanism ( 26 – 33 ), to achieve excellent expression power in a physical reservoir–based RC. Furthermore, a strong dependence of the synaptic response on channel length, which is a feature of ion-electron–coupled dynamics, is used to realize high dimensionality in a single IGRT. The IGRT exhibited small errors in some RC tasks, including 0.020 of normalized mean squared error (NMSE) in a nonlinear autoregressive moving average (NARMA) task, which is a typical benchmark for RC ( 7 , 8 , 13 , 14 , 34 – 37 ), and achieved 88.8% accuracy in a handwritten-digit recognition task. The underlying mechanism of the characteristic synaptic response was investigated on the basis of multiphysics simulation, and it was found that complexed charge density patterns form and change from moment to moment in an extremely thin EDL region (<2 nm) during storage and processing of input signals. We further performed a Lyapunov analysis to investigate a possible origin of the high performance from a nonlinear dynamics viewpoint. The calculated value of the maximum Lyapunov exponent was −6.3 × 10 −3 , close to an edge-of-chaos state, which is located between order (λ < 0) and chaos (λ > 0) and is empirically known to derive high expression power from a reservoir in RC ( 14 , 22 , 24 , 38 – 40 ). Because of the many advantages observed, including (i) the very thin nature of EDL (e.g., nanometer order) and the spontaneous formation at interfaces and (ii) the strong nonlinear response based on ion-electron–coupled dynamics, our approach is useful for realizing high-performance, highly integrated, and low–power consumption AI devices by harnessing the inherent physical and chemical characteristics of materials.",
"discussion": "RESULTS AND DISCUSSION Electrical responses of IGRT and its application to image recognition In biological neuronal network systems, synaptic responses show characteristic variations in waveform, intensity, and frequency, with respect to environmental inputs in various forms due to the chaotic dynamics of the NN, as illustrated in Fig. 1A ( 41 , 42 ). The synapses belonging to different dendritic structures are strongly correlated, and their electrical behavior is affected by this correlation. The wide variations in synaptic responses are used to achieve high expression power for efficient information processing. In this study, we use an all-solid-state IGRT, operating in an EDL mechanism, to obtain wide variation in electrical response for efficient information processing. While the versatile electrical behavior in biological neuronal network is due to the correlation between synapses, our IGRT achieves similar versatile electrical responses on the basis of ion-electron–coupled dynamics at a lithium ion electrolyte/diamond interface, as discussed below. Fig. 1. Electrical response of IGRT based on the EDL effect and its application to image recognition. ( A ) Illustrations of synaptic responses in biological NNs and our IGR operating in an EDL mechanism. ( B ) An example of the handwritten digit 6 from the MNIST database ( 44 ). ( C ) Drain current ( I D ) responses of the IGRT to 16 different pulse streams. ( D ) Image recognition accuracy achieved by IGR as a function of the number of trained images. The dotted line shows the accuracy of a typical, full-simulation, three-layer NN. The size of the IGR and NNs are given in parentheses. The recognition accuracies of other physical reservoirs, such as memristors [magnetic skyrmion memristor (MSM)] ( 13 ), WO x ( 16 ), SiO x -Ag ( 18 ), and ionic liquid (IL) ( 43 ), are shown for comparison. As shown in Fig. 1A , the transistor consists of a lithium ion–conducting solid electrolyte [Li-Si-Zr-O (LSZO)], a hydrogen-terminated diamond (100) single crystal with a homoepitaxial layer, and LiCoO 2 /Pt gate electrode, which works in the manner of an EDL transistor (EDLT) ( 33 ). LiCoO 2 is a Li + -electron (hole) mixed conductor, and it serves as the gate electrode by a reversible Li + ion insertion/desertion property. It can supply (remove) Li + ions to (from) the diamond channel/LSZO interface to modulate accumulated charge at the EDL under gate voltage ( V G )–applied conditions. By applying negative V G to the EDLT, Li + ions are removed from the diamond/LSZO interface to form the EDL with negatively charged Li vacancy in the LSZO and positively charged hole in the diamond, resulting in notable conductance enhancement in the channel [a low-resistance state (LRS)], as depicted in Fig. 1A . In the opposite manner, positive V G application causes Li + ion accumulation at the interface, which is accompanied by hole depletion and the resultant insulating [a high-resistance state (HRS)] in the channel. While a transition process from LRS to HRS is fast, the one from HRS to LRS is slow because the Li + ion motion in the electrolyte is affected by channel resistance with HRS, that is, the ion-electron–coupled dynamics. The I D (drain current)– V G and gate current– V G curves for IGRT are shown in fig. S1. Details of the dynamics will be discussed later with multiphysics simulation. In addition to the asymmetric relaxation behavior, spikes are observed in the I D response as shown in Fig. 1A . The spikes are originated from the EDL charging current and useful to enhance diversity of the I D response, as discussed later. The IGRT has eight channels with different channel length ( L ). Because all the channels of the device are connected by the electrolyte layer and share common gate and source electrodes, input signals are also shared by the channels. Therefore, the different channels are coupled to each other by three-dimensional Li + transport through the electrolyte to some extent. Said transistor is used as an IGR, which is a novel class of physical reservoirs. The IGR can map time series data in high-dimension feature space by using I D response, with asymmetric relaxation and spikes, the characteristics of which are widely modified by the input history. To investigate a function of IGR as a physical reservoir, we performed a handwritten-digit recognition task ( 13 , 16 , 18 , 43 ). Figure 1B is an example of the 28 × 28 pixel input digit “6” from the Modified National Institute of Standards and Technology (MNIST) database ( 44 ). Said image was converted into binary time series data and input to the IGRT. The reservoir states were obtained from I D . Figure 1C shows the 16 different reservoir states, which were well separated from each other so that all 16 different pixel combinations could be expressed by unique reservoir states. These values were used as the reservoir output to train and test the readout network. Similar methods have been used elsewhere ( 13 , 16 , 18 , 43 ). The details of the procedure used are given in Materials and Methods. Figure 1D shows image recognition accuracy versus the number of trained images. The recognition accuracy improved from 70.4 to 88.8% as the number of trained images increased from 100 to 60 thousand, supporting understanding that the recognition task is suitable for the IGR. While the performance was not as good as that achieved by a typical three-layer NN, the size of the network in the present study (1960) is far smaller than in a three-layer NN (784,000). Compared to the recognition accuracies of other physical reservoirs (83 to 90.2%) ( 13 , 16 , 18 , 43 ), that of IGR is similar or slightly better. However, while the advantage of IGR is minor for such a relatively easy task, IGR showed very good computational performance on more difficult time series data analysis tasks that require superior properties, such as reservoir diversity, which is discussed below. Solving a second-order nonlinear dynamic equation by IGR RC is suitable for time series data analysis because it has features such as short-term memory, nonlinearity, and high dimensionality for input data. We took advantage of such suitability by using the IGR to solve a second-order nonlinear dynamical equation task ( 13 , 16 ), a schematic of which is shown in Fig. 2A . The target y t ( k ) is obtained from following equation y t ( k ) = 0.4 y t ( k − 1 ) + 0.4 y t ( k − 1 ) y t ( k − 2 ) + u 3 ( k ) + 0.1 (1) where k and u ( k ) = [0,0.5] are a discrete time and a random input that were applied to IGRT as V G pulse streams, respectively. The reservoir states X i ( k ) were obtained from the I D response, and the reservoir output y ( k ) is the linear combination of X i ( k ) and read out weights w i trained by ridge regression as follows y ( k ) = ∑ i = 1 N w i X i + b (2) where N and b are the reservoir size and bias, respectively. Details of the procedure are given in Materials and Methods. Fig. 2. Solving a second-order nonlinear dynamic equation task. ( A ) Schematic of task calculated by IGR. ( B ) Various I D responses of the IGRT at different channel lengths. ( C ) The method for obtaining virtual nodes and ( D ) various reservoir states ( I D streams) from 10 virtual nodes. ( E ) Target and prediction waveforms of second-order nonlinear dynamic equation at the test phase. ( F ) Prediction error compared to other physical reservoirs. To obtain high-dimensional reservoir states from one-dimensional input, IGRT with an eight-channel (drain)–one-gate–one-source structure and eight different channel lengths (20 to 1000 μm) were fabricated as shown in Fig. 1A . As shown in Fig. 2A , the eight different channels provide eight physical nodes in the RC. By applying V G pulse streams to one common gate and one common source, we can measure eight different I D responses from the eight drains (drains 1 to 8 in Fig. 2A ). Constant drain-source voltages ( V D = −500 mV) are applied to between drains 1 and 8 and common source. Concerning V G pulse streams, a random input u ( k ) is converted to V G pulse streams over a range from 0 to 0.5 V (please refer to Materials and Methods for the details). We can get 10 virtual nodes from each of eight different I D responses. So, we have 80 nodes in total. Target waveform is reproduced from a linear combination of the 80 nodes (reservoir states) with 80 weights. In the training phase, 80 weights are stored in the operating computer, and error was minimized by the ridge regression. Eight I D responses from common gate inputs were obtained as shown in Fig. 2B . The intensity of the spikes observed at the edges of V G pulses differ depending on the channel length. The spikes are due to a gate current induced by the ion current, which depends on the differential of charge in the EDL and are more significant with small I D compared to the gate current. Therefore, short channels with low resistance (≤100 μm) do not exhibit spikes, while long channels with high resistance (≥200 μm) exhibit large spikes. For further higher dimensionality of reservoir states, multiple reservoir states were obtained as virtual nodes, as shown in Fig. 2C ( 35 ). The former nodes 1 to 5 and the latter nodes 6 to 10 correspond to I D responses measured during the application of write pulses and during the pulse intervals ( V G = 0 V), respectively. The former nodes use the fast relaxation process of channels from an LRS to an HRS, which is dominated by Li + ion accumulation at the electrolyte/channel interface, while the latter nodes use a relaxation process from an HRS to an LRS of the channel, which is a relatively slow relaxation process because the Li + ion motion in the electrolyte is affected by channel resistance with HRS. Because of the ion-electron–coupled dynamics, in which the ions of the electrolyte and the electrons of the channel interact, the IGRT exhibits asymmetric relaxation behavior. In addition, nodes 1 and 6 are characteristic virtual nodes located at the peak of the spike-like I D . Thus, by using the virtual nodes, we could effectively extract features such as asymmetric relaxation and spike behavior, whichare unique features of the EDL. These unique IGR features, induced by ion-electron–coupled dynamics, will be discussed in Fig. 4 . Figure 2D shows the reservoir state obtained at each virtual node (1000-μm channel) for a random wave u ( k ) input. Furthermore, Fig. 2D shows how virtual nodes are taken from the I D responses measured in the eight physical nodes. One V G pulse, which corresponds to u ( k ), gives one I D response in a physical node just as the one shown in Fig. 2C (80 to 100 ms). It includes 10 virtual nodes as shown in Fig. 2 (C and D) . Thus, 10 different I D streams are reproduced as a function of time step from the 10 virtual nodes as shown in Fig. 2D . Because the device has 8 physical nodes, each of which give 10 virtual nodes, we have 8 (physical nodes) × 10 (virtual nodes) = 80 nodes in total as shown in Fig. 2A . It can be seen that the IGR has good diversity, with each virtual node showing various behaviors as reflections of its own characteristics. The combination of physical and virtual nodes resulted in a reservoir size of 80. Figure 2E shows the target for the test data and the predicted output by IGR. The predicted output is in excellent agreement with the target, that is, Eq. 1 was successfully solved by the IGR. As shown in Fig. 2F , the predicted error was 1.62 × 10 −4 (training data) and 2.08 × 10 −4 (test data), respectively. Compared to other physical reservoirs ( 13 , 16 ), the prediction error was extremely low, indicating that the IGR performs well on time series data analysis tasks. Such good computational performance of IGR is due to its ability to effectively exploit the complex and diverse features inherent in the ion-electron–coupled dynamics of IGR as reservoir states. An additional important factor was the stable reproduction of the good expressivity of the IGR. This means that the nonlinear mapping to higher-dimensional spaces by the IGR performed on the training data was exactly the same as for the test data without altering the IGR condition during operation. This indicates that the IGR satisfies the echo state property, which is one of the important properties required for reservoirs ( 3 ). NARMA2 task We performed predictions on time series data generated by a NARMA2 system ( 45 ), as shown in Eq. 3 , as a more challenging time series data analysis task. This is known as a NARMA2 task and is commonly used as a typical RC benchmark task ( 7 , 34 , 36 , 37 ). y t ( k + 1 ) = 0.4 y t ( k ) + 0.4 y t ( k ) y t ( k − 1 ) + 0.6 u 3 ( k ) + 0.1 (3) where u ( k ) = [0,0.5] is a random input. To evaluate the computational performance of IGR in the NARMA2 task, we used the NMSE for an index of RC performance, an explanation of which is given in Materials and Methods. Figure 3A shows the relationship between the IGRT operating conditions and the NMSEs (test phase) of the NARMA2 task. Good prediction performance was observed in the operation region with an input pulse period of 20 ms or longer and a duty ratio of 75% or higher. In particular, the best prediction performance (NMSE = 0.020 in the test phase) was achieved at a pulse period of 50 ms and a duty ratio of 75%. The target and the predicted output by IGR (test phase) under these conditions are shown in Fig. 3B . Both waveforms are in excellent agreement, evidencing that IGR successfully predicted the time series generated by the NARMA2 system (please refer to fig. S2 and section S2 for details). Figure 3C shows the relationship between the NMSE of the NARMA2 task in the test phase and the volume of the physical reservoirs reported so far ( 7 , 34 , 36 , 37 ). Although there are not many reports of physical reservoirs that experimentally demonstrate the NARMA task, IGR showed the best results in the prediction performance despite its extremely small volume compared to other physical reservoirs. That is, the IGR showed both extremely good computational performance on a single device and its suitability for integration. We also evaluated the effect of device geometry on IGR performance. Please refer to fig. S3 and section S3 for details. Fig. 3. NARMA2 task. ( A ) The relationship between IGRT operating conditions and NMSEs in the test phase of the NARMA2 task. ( B ) Target and prediction waveforms of the NARMA2 task. ( C ) NMSEs of the NARMA2 task and reservoir volumes of various physical reservoirs, which experimentally demonstrated the NARMA2 task. The reservoir volume of IGR was calculated as the product of the total channel area and the thickness of the electrolyte. Simulation of ion-electron–coupled dynamics in IGR The ion and electron dynamics in our IGR were simulated by using COMSOL multiphysics simulation software (COMSOL Inc.) to clarify the underlying mechanism in the unique I - V characteristics of our device. As shown in Fig. 4A , the EDLT model, which is composed of a Li + electrolyte, a channel, and electrodes, was constructed by assuming the physical properties of LSZO, EDL, and the device structure with an adjustment of for reduction of computation burden. The I D - V G and I D - V D characteristics of the simulated model agreed well with the experimental result (please refer to fig. S4, A and B, and to Materials and Methods). Figure 4B shows the I D response of the device model under four sequential gate pulse applied conditions. As seen in the rise and fall behavior of I D , the model reproduces asymmetric I D responses with spikes that are signatures of our device, supporting our understanding that the simulation reproduces the actual electrochemical transport phenomena in the device. To grasp the ion and electron (hole) dynamics in the model in the operation, we capture snapshots of the ion and hole density distribution at specific points. In the snapshot at the initial state, shown in Fig. 4A , substantial in-plane carrier distribution is found, in which densities of positively charged holes and negatively charged Li vacancies are higher near the source electrode than near the drain electrode. This corresponds to formation of EDL, which is differently charged by voltage distribution due to application of V D (= − 500 mV) between the source and drain electrodes. It is noted that the out-of-plane distribution of excess Li + (and Li vacancies) accumulates within 0.3 nm from the interface. The extremely thin nature of the EDL is consistent with the in situ hard x-ray photoelectron spectroscopy observation ( 33 ). Besides, comparison between the four conditions shown in Fig. 4C evidences that repetition of input makes a variety of charge density patterns in the channel. For example, at t = 50 ms, not only the low–hole density region proceeds from the drain side to the source side; an island-like pattern also appears within 0.3 nm from the channel/electrolyte interface. This is because such proceeding of the low–hole density region occurs not only from the drain side but also from the source side, resulting in a variety of transient charge density patterns. Please refer to movie S1 and fig. S5 for a movie of the charge density variation and a detailed discussion on the charge distribution change for the first pulse input, respectively. Fig. 4. Simulation of ion-electron–coupled dynamics. ( A ) The simulated EDLT, modeled by COMSOL Multiphysics, and the ion and hole distribution at the electrolyte/channel interface at the initial state. The ion distribution of the electrolyte shows the amount of change from 10 22 cm −3 of Li + concentration. ( B ) The I D response of the simulated model under sequential gate pulse applied condition. The dotted line shows the experimental result. ( C ) Snapshots of the ion and hole distribution, which are captured at each of the four pulses shown in (B). ( D ) Schematic illustration of the I D path in the IGR, consisting of two I D s: one corresponding to state variable X ( t ) and the other corresponding to state variable Y ( t ). ( E ) The I D response as a mixed reservoir of X ( t ) and Y ( t ). Basically, such behavior can be understood in the framework of transmission line model, in which electrical resistance is dependent on the location due to the different length of the current path ( 46 ). However, in the present case, local hole resistance in the channel is strongly dependent on the charging history of EDL. This gives the I D response of the IGR asymmetric relaxation. Furthermore, spikes add another feature to the response. High performance of the IGR is discussed below as two contributions to the total I D : I D to the source and I D to the gate. Figure 4D is an illustration of the I D path in the IGR. A partial I D (from the drain to source) corresponds to state variable X ( t ) defined with an integral of local resistance R ( x , t ) at each channel position x (0 ≤ x ≤ L ), $ X ( t ) = V D ∫ x = 0 L R ( x , t ) d x $ , in which L , x , and V D are channel length, position in the channel, and drain voltage (constant), respectively. By introducing a local voltage applied to EDL V EDL ( x , t ) and EDL capacitance (constant) C , X ( t ) can be further transformed to X ( t ) = ∫ x = 0 L V D q μ C V EDL ( x , t ) d x (4) in which q and μ are the elementary charge and hole mobility, respectively. On the other hand, the rest of the I D with a spike appearance (from the drain to the gate) corresponds to state variable Y ( t ), defined with an integral of EDL charging current I EDL ( x , t ) at each channel position x (0 ≤ x ≤ L ) Y ( t ) = ∫ x = 0 L I EDL ( x , t ) d x = ∫ x = 0 L C d V EDL ( x , t ) d t d x (5) in which V EDL ( x , t ) and C are a local voltage applied to EDL and EDL capacitance (constant), respectively. As seen from Eqs. 4 and 5 , while both X ( t ) and Y ( t ) include V EDL ( x , t ), it is only expressed in a derivative form. Therefore, although both X ( t ) and Y ( t ) are related to V EDL ( x , t ), they function as two different types of reservoir. Because the I D observed is the sum of I D to source and to gate, the I D response is a mixed reservoir of X ( t ) and Y ( t ), as shown in Fig. 4E . Recently, such mixed reservoirs have been theoretically predicted to show high performance by overcoming a trade-off relationship between short-term memory and nonlinearity due to the coexistence or mixture of linear dynamics and nonlinear dynamics in a reservoir ( 47 ). The mixed reservoir property can be a reasonable explanation for the high performance discussed in time series data analysis tasks shown in Figs. 2 and 3 . In addition, we evaluated the virtual node dependence of NMSE for NARMA2 task to analyze the mixed reservoir effect in IGR. The best prediction performance was obtained for virtual nodes 1 and 6, which correspond to the spike behavior of the I D ( Y ( t ) dominant regions in Fig. 4E ), as shown in fig. S6A. This indicates that such spikes not only provide reservoir diversity shown in fig. S6B but also contribute significantly to the computation. Please refer to section S5 for detailed discussion. Lyapunov analysis To evaluate the high performance of the IGR in terms of nonlinear dynamics, we calculated the Lyapunov exponent, which quantifies the trajectory stability of the dynamical system by the Jacobi matrix method for direct analysis of time series data based on unknown dynamical systems ( 39 , 48 ). Figure 5A shows the nonlinear I D response used in the chaos time series analysis for the 20-, 700-, and 1000-μm channels when a triangular wave is input to the IGRT. These channels exhibit completely different nonlinear responses, including the presence of spikes. While I D response for L = 20 μm shows monotonical increase (blue arrow) and decrease (red arrow) with respect to the input triangle waves, those for L = 700 μm and L = 1000 μm show much complex behavior with negative differential resistance–like nonlinear input-output characteristics. Specifically, the I D responses for L = 700 μm and L = 1000 μm show a decrease in I D output as indicated by blue arrows in regions where input V G increases (colored in red in the inset) and an increase in I D output as indicated by red arrows in regions where input V G decreases (colored in blue in the inset). Such a negative differential resistance–like nonlinear input-output characteristic is a highly nonlinear behavior that has been reported for memristors in edge-of-chaos states ( 49 ). To analyze the nonlinearity of IGR in detail, we generated 40 reservoir states X by obtaining five virtual nodes, corresponding to nodes 1 to 5 in Fig. 2C , for I D obtained from eight channels. Fig. 5. Lyapunov analysis. ( A ) Nonlinear I D response of the IGR. Triangular wave input (top) and I D response of IGR obtained from 20-μm length channel, 700-μm length channel, and 1000-μm length channel. ( B ) The return maps of the reservoir correspond to a 20-μm length channel (left), a 700-μm length channel (middle), and a 1000-μm length channel (right) with nodes 1 and 5. ( C ) The three-dimensional cross section of the 41D reservoir state spaces of the IGR. ( D ) Lyapunov spectrum of the IGR calculated by the Jacobi matrix method. Figure 5B shows the return map [ X ( k ) versus X ( k + 1)] obtained from the reservoir states of nodes 1 and 5 with L = 20 μm, L = 700 μm, and L = 1000 μm ( X 20μm,Node1 , X 20μm,Node5 , X 700μm,Node1 , X 700μm,Node5 , X 1000μm,Node1 , and X 1000μm,Node5 ). The return maps are completely different for each virtual node and for each channel length (physical node), which indicates that IGR achieves good diversity as a result of higher dimensioning by introducing virtual node and channel length. The return map at L = 20 μm, shown on the left of Fig. 5B , has a narrow trajectory width, indicating an almost completely periodic response to the triangular wave input. On the other hand, the return map at L = 700 μm and L = 1000 μm, shown in the middle and right of Fig. 5B , has a wide trajectory, indicating that the reservoir state has a relatively unstable response that varies slightly from period to period. Similar unstable characteristics have been reported for memristors ( 49 ) and nanowire networks ( 22 ) in chaos and edge-of-chaos states. We calculated Lyapunov exponents λ, an index of order-chaotic dynamics, of the IGR using the Jacobi matrix method ( 39 , 48 ). Figure 5C shows the attractor in phase space created by selecting the axes in the X 20μm,Node1 , X 700μm,Node1 , and X 700μm,Node5 directions as one of the cross sections of the 41-dimensional phase space. The calculated Lyapunov spectrum is also shown in Fig. 5D . The Lyapunov exponents show values ranging from a minimum of −3.0 to a maximum of −6.3 × 10 −3 : The maximum Lyapunov exponents λ Max of the IGR is −6.3 × 10 −3 . Dynamical systems with maximum Lyapunov exponents λ Max near zero are called edges of chaos, and it has been reported that high computational performance is achieved at these edges of chaos in computing for physical reservoirs ( 14 , 22 , 24 ), full simulation reservoirs ( 38 – 40 ), and recurrent NN ( 50 ) because of their robustness in the processing of information ( 25 ). The high computational performance of IGR can also be attributed to edge of chaos, which was achieved by nonlinearity and high dimensionality. In other words, the high expressivity is realized by the asymmetric relaxation and spiking of the I D , which originates from the ion-electron–coupled dynamics at the electrolyte/semiconductor interface. During application of pulse V G streams to the IGRT, diverse and complex charge density distribution is formed by reflecting the past experience (hysteresis) as discussed in Fig. 4 . Because of ion-electron–coupled dynamics, ionic and electronic carrier density distributions at a certain time step k strongly affect on the carrier density distributions formed by subsequential applications of V G pulse streams later than time step k + 1. Therefore, the carrier density distribution becomes very sensitive to the past experience, and it gives instability to the system, leading to the edge-of-chaos state. To achieve high-performance RC, we developed an IGR on the basis of ion-electron–coupled dynamics in the vicinity of a lithium ion solid electrolyte/diamond interface. In the study, various synaptic responses, with asymmetric relaxation and spikes, are effective in achieving excellent expression power for mapping time series data to higher-dimensional feature space. Good RC performance of the IGR was demonstrated in handwritten-digit recognition, nonlinear transformations, and NARMA2 tasks. Multiphysics simulation revealed that during operation, transient charge density patterns form and change from moment to moment in an extremely thin EDL region. Asymmetric relaxation and spikes in the I D response enables high expression power by realizing a mixed reservoir comprising different nonlinear dynamics. Lyapunov analysis was performed to inspect the dynamical features of the IGR, which analysis revealed that the maximum Lyapunov exponents of the carrier dynamics is −6.3 × 10 −3 , supporting the understanding that the IGR operates in edge-of-chaos states under certain conditions. Furthermore, the concept of an IGR can be extended to various information carriers (e.g., electrons, ions, light, and spin) as long as their dynamics or transport interact with each other. While the present EDL system, with its ion-gating transistor structure, is a typical case, various physical or chemical systems can be used for achieving IGR with diverse information carriers. Various materials and interfaces present exciting frontiers for exploring high-performance, versatile, and integrated physical RC based on the coupled dynamics inherent in IGR."
} | 8,539 |
34830199 | PMC8621035 | pmc | 8,155 | {
"abstract": "Surface antimicrobial materials are of interest as they can combat the critical threat of microbial contamination without contributing to issues of environmental contamination and the development drug resistance. Most nanostructured surfaces are prepared by post fabrication modifications and actively release antimicrobial agents. These properties limit the potential applications of nanostructured materials on flexible surfaces. Here, we report on an easily synthesized plastic material with inherent antimicrobial activity, demonstrating excellent microbicidal properties against common bacteria and fungus. The plastic material did not release antimicrobial components as they were anchored to the polymer chains via strong covalent bonds. Time-kill kinetics studies have shown that bactericidal effects take place when bacteria come into contact with a material for a prolonged period, resulting in the deformation and rupture of bacteria cells. A scanning probe microscopy analysis revealed soft nanostructures on the submicron scale, for which the formation is thought to occur via surface phase separation. These soft nanostructures allow for polyionic antimicrobial components to be present on the surface, where they freely interact with and kill microbes. Overall, the new green and sustainable plastic is easily synthesized and demonstrates inherent and long-lasting activity without toxic chemical leaching.",
"conclusion": "4. Conclusions A method of fabricating the inherent antimicrobial polystyrene, by decorating polymer chains with polyionic antimicrobial components, thereby leading to surface phase separation, was developed. The functionalized PS* exhibits a very rough surface with submicron scale structures of active components and presents enhanced hydrophilicity, demonstrating excellent antimicrobial efficiency. The interaction of bacteria cells with semi-free polyionic components on PS* surfaces via weak forces resulted in the high antimicrobial performance of PS*. Considering the relatively long shelf life and biocompatibility, the PS* can potentially be used for a wide range of applications, including in healthcare, medicine, and food packaging as green and sustainable self-disinfection materials.",
"introduction": "1. Introduction Microbial infection remains one of the most serious complications and has received much attention in recent years [ 1 , 2 ]. Contamination by microorganisms is of great concern for numerous applications, including in healthcare products [ 3 ], medical devices [ 4 , 5 ], food packaging [ 6 , 7 ], water purification systems [ 8 ] and manufacturing industries [ 9 , 10 ]. Polystyrene (PS), for example, is a commodity plastics that is widely used for packaging materials or common utilities, however, the contamination on PS surfaces is yet to be resolved. Considerable effort has been devoted to developing self-disinfection polystyrene to avoid using a large amount of disinfectants and for enhanced sustainability [ 11 ]. Most strategies focus on formulating PS with other materials, such as chitosan [ 12 ] and inorganic materials [ 13 , 14 , 15 , 16 ]. Considering the immense scale of PS applications in packaging, especially in the food packaging industry, inherently antimicrobial polystyrene that has a self-disinfecting surface without the drawback of leaching toxic chemicals is highly desired [ 17 , 18 ]. Phase separation in polymer blends or block co-polymers is a widely studied process, and has been used in lithography for various surface modifications [ 19 ]. The phase separation of polymer blends has also been used to create an antimicrobial nanostructure by casting-replication processes [ 20 ]. Herein, we describe a new method to fabricate submicron soft nanostructures on a polystyrene film surface through a bottom-up approach via the copolymerization of 1% of polyionic components into hydrophobic polystyrene chains. The hydrophilic components self-assemble into different soft sub-micron structures in different solvent systems. The surface nanostructures allow polyionic components, situated at the tip of nanostructure, to achieve greater flexibility and the freedom to interact with the surrounding microbes, thereby granting inherent antimicrobial properties to the polystyrene surface. The resultant antimicrobial plastic (PS*) attracts and kills bacteria via the semi-free polyionic components, functioning in a similar manner to Drosera plants, which capture and digest insects upon contact with the plant’s hairy leaf surfaces ( Scheme 1 ).",
"discussion": "2. Results and discussions 2.1. Synthesis and Characterization of Polystyrene Modified with Polyimidazoliums on Surface To induce phase separation, a polyimidazolium (PIM) compound (PIM-45, Mn~2500 kDa) [ 21 ] was selected as the active component to be incorporated into the polystyrene chain. PIM-45 is a main-chain antimicrobial polymer with high selectivity and broad spectrum activity [ 22 , 23 ]. It was modified with styrenyl functional groups at both ends for further copolymerization ( Figure 1 and Figure S1 ). The synthetic approach is illustrated in Figure 1 a, where the resultant polymer PIM-Vinyl is subsequently polymerized with styrene and integrated within the polystyrene chain. A series of functional polystyrene film samples (PS*-X, X = wt.% of PIM-Vinyl in polymer) were synthesized via the copolymerization of styrene with different loadings of PIM-Vinyl. PS-PIM 45 was also synthesized by simply mixing styrene with PIM-45 (with no styrenyl end group) as a control. The free PIM-45 can be entirely washed away during the sample preparation process. Signals corresponding to the PIM-45 component within the synthesized polystyrene material can be observed from the 1 H and 13 C NMR spectra of PS*-1 ( Figure S2 ). The average molecular weight of PS*-1 is around 49.9 kDa, which is slightly lower than the PS control of 52.2 kDa ( Table S1 ). To determine the potential phase separation of PIM incorporated polymers, all PS films were prepared in two solvents, toluene (T) and chloroform (C), with different polarities. Firstly, the surface wettability of synthesized PS was determined by water contact angle measurement. The contact angles of the PS*-1 films were significantly lower than PS controls ( Figure 1 b,c; Table S2 ). The PIM compound has high polarity with positively charged imidazolium rings within its structure. A reduction in the contact angle of PS*-1 suggested that the hydrophilic imidazolium group could be mostly presented on the material surface, resulting in a small contact angle. 2.2. Surface Nanostructures of PS-Based Materials The atomic force microscopy (AFM) images exhibit significantly different surface structures between PS-control and PS* samples, Figure 2 a–d [ 24 , 25 ]. All PS-control sample surfaces are generally smooth with an average roughness (Ra) in the range of 1 nm < Ra < 20 nm, for which there is no significant difference between both control samples prepared in different solvents [ 26 ]. However, PS*-1 samples revealed very interesting submicron scale surface topographic structures. AFM topography images showed that the overall average surface roughness of PS*-1 prepared with chloroform is Ra = 55.1 nm and with toluene it is Ra = 102.9 nm ( Table S3 ). AFM height images showed that PS*-1 sample prepared in toluene has an aligned hair structure, with the height of hairs of around 500–600 nm. In contrast, a PS*-1 sample prepared in chloroform exhibits a honeycomb-like surface structure with a drop from peak to valley of around 800 nm. It is proposed that the surface nanostructures of PS*-1 in toluene or chloroform are PS-PIM co-polymer induced phase separation. The PIM-Vinyl loading in the PS*-1 sample is 1 wt.%, where the number average molecular weight (Mn) of PIM-45 and PS*-1 are about 2.5 kDa and 50 kDa, respectively. It is believed that around 20% of the polymer chains were decorated with a PIM-45 block and that the highly hydrophilic PIM-45 block successfully induced phase separation on the surface to create submicron scale structures. To further verify the relationship between surface nanostructures and the PIM component in functionalized polystyrene, PS* with different PIM-Vinyl loadings were investigated. The surface roughness and the height of the hair structure of PS* increased as the PIM-Vinyl loading increased from 0.2 to 1 wt.% ( Figure 2 g, Table S4 ). When the PIM-Vinyl loading further increased, the PS* surface roughness reached a plateau. The density of the nanostructures (amount of hair structures) on the surface were found to be well correlated with the PIM-Vinyl loading. In addition, the peak or bright areas in the AFM height images are correlated to the dark areas reflected in the phase images, which indicates that these areas are relatively more hydrophilic ( Figure S3 ) [ 27 ]. These results clearly indicate that the PIM components in the copolymer can spontaneously assemble on the surface, induce phase separation, and accumulates the surface nanostructures. Given these findings, this method of inducing nanostructures on PS surfaces was validated using other antimicrobial compounds, a main-chain DABCO-imidazolium copolymer (Mn~2.2 kDa) [ 28 ] and small molecules of benzalkonium chloride (BZK), by introducing styrenyl functional moiety to the compound for subsequent polymerization with styrene. Functionalized plastic PS#-1 (with DABCO-imidazolium copolymer) and PS-BZK with a 1 wt.% compound loading were fabricated respectively ( Scheme S1 ). As expected, PS#-1 exhibited similar surface nanostructures and average roughness (Ra = 64.9 nm) to PS*-1, while PS-BZK showed a very smooth surface (Ra = 6.7 nm) ( Figure 2 e,f, Table S3 ). This result further suggests that the submicron scale soft nanostructures are induced by the surface phase separation of polystyrene chains, which are decorated with hydrophilic chain blocks. In contrast, polystyrene chains that are decorated with a small molecular compound will not experience a similar surface phase separation, leading to the absence of nanostructures on the material’s surface. 2.3. Antimicrobial Property Evaluation Functionalized PS films (PS*-1, PS#-1) displaying submicron soft nanostructures on their surfaces were then evaluated for their respective antimicrobial properties. All active compounds, before and after the addition of styrenyl functional groups, were first evaluated and proven to have good microbial inhibitory activity, i.e., minimum inhibitory concentrations (MICs) against E. coli and S. aureus of 4 µg/mL and 2 µg/mL respectively ( Tables S5 and S6 ). The functionalized PS plastic films were subsequently evaluated according to the Japanese Industrial Standard JIS Z 2801/ISO 22196 test method. As shown in Table 1 , PS-PIM 45 did not exhibit antimicrobial activity. In contrast, both PS*-1 (T) and PS*-1 (C) exhibited excellent bactericidal properties against E. coli and S. aureus with a log reduction in the colony forming units of more than 6 after 24 h incubation. The good bactericidal activity indicates that PS*-1 can effectively kill bacteria even when the antimicrobial PIM-Vinyl compound was integrated within the polymer chains. As for fungus C. albicans , PS*-1 did not exhibit antifungal activity, but strong fungicidal activity was demonstrated for the PS*-3 surface where the loading of PIM-Vinyl was increased to 3 wt.% ( Figure S4 ). The time-kill kinetics of PS*-1 against E. coli with different incubation periods were compared ( Figure 3 a). PS*-1 exhibited a strong bactericidal property against E. coli when the incubation time was more than 8 h. Similarly, PS#-1 exhibited a good bactericidal property against E. coli ( Figure 3 b), while PS-BZK did not exhibit antibacterial activity against E. coli for a log reduction <2 ( Figure 3 c). This result is highly aligned with the AFM observations, whereby PS#-1/PS*-1 exhibited similar surface structures and roughness, while PS-BZK showed a very smooth surface ( Figure 2 e,f, Table S3 ). Therefore, we propose that the formation of soft and flexible surface nanostructures in PS films, prepared using co-polymerizing styrene, with main-chain antimicrobial compounds is responsible for the high antibacterial activity observed. In contrast to rigid nano-structured surfaces, most of them do not exhibit a good bacteria killing efficacy using the JIS Z 2801 method, as they are only able to kill the attached bacterial cells [ 29 , 30 ]. To determine the microbial killing mechanism of copolymer PS*-1, a leaching assay was used to assess the leaching properties of PS*-1 following the GB 15979-2002 method with minor modifications made ( Figure S5 ). The results are shown in Figure S6a . The leached solution did not present any bactericidal activity against E. coli , which indicates the absence of antimicrobial compounds in solution. In contrast, a bacterial reduction of more than 80% was observed ( Figure S6b ) when PS*1 was introduced to the bacterial solution. This suggested that PIM-Vinyl was well incorporated in PS*-1 rather than simply embedded within the PS film, thus it did not leach out from the material to cause antibacterial activity. PS*-1, with covalently bonded PIM-Vinyl, inhibited bacterial growth while in contact with bacteria in the solution. In addition, PS-PIM 45 did not reveal any bactericidal activity, indicating that the compound was not entrapped within the polymer matrix. As the PIM-45 compound with no reactive end groups was washed off during the repeated precipitation steps, the resultant PS-PIM 45 material did not contain any remaining antimicrobial materials that may leach out to the test solution to cause bactericidal activity. It should be noted that the length or height of the PS* surface nanostructures (hairs or peaks) are above 500 nm, which provides great flexibility and freedom for the polyionic components in the nanostructures to act as free antimicrobial compounds to kill bacteria [ 21 , 22 , 23 ]. The morphological changes of bacteria on PS-control and PS*-1 surfaces were examined under a scanning electron microscope (SEM) ( Figure 4 ). No changes to the morphology of E. coli in the control samples were observed at varied time points. However, a small number of deformed bacteria cells were observed on both PS*-1 films prepared from chloroform and toluene after 6 h exposure to PS*-1 surfaces, where the rupture of bacteria can be observed on the end of E. coli cells. After 24 h, all bacteria cells on PS*-1 surfaces were deformed and ruptured, leading to cell death. This further confirmed that the killing of bacteria occurred due to the contact of bacteria with polystyrene surfaces that contained an imidazolium-based antimicrobial compound, which eventually caused membrane lysis [ 21 , 22 , 23 ]. Therefore, it is suggested that the presentation of polyionic components on PS*-1/PS#-1 surfaces creates soft nanostructures (submicron scale) that can interact with bacteria cells via electrostatic and other forces. The contact of bacteria cells with semi-free polyionic components resulted in the high antibacterial performance of functionalized PS, as demonstrated in Scheme 1 . The soft nanostructures and the positively charged antimicrobial component on the surface function as micro- Drosera plants that capture and digest insects using stalk hair and the sticky liquid covering their leaf surfaces. 2.4. Durability and Biocompatibility of PS-Based Antimicrobial Plastics The inherent antimicrobial property or the shelf life of the disinfection effect of PS*-1 was evaluated using an accelerated aging test according to the ASTM F1980-16 test method [ 31 ]. PS-control and PS*-1 (T) were stored in a closed environment with a constant humidity of >95% at 60 °C. After two months of aging at 60 °C which approximates to two years in real-time, PS*-1 exhibited excellent bactericidal activity against E. coli ( Figure S7a ). None of the aged PS samples showed obvious physical changes. This suggests that the PS*-1 samples may have an average shelf life of more than two years for different applications. As for safety concerns regarding PS* materials, a cell viability assay was conducted using mouse fibroblast L929 cells. There is no significant viability difference between PS*-1, PS-PIM 45, and PS-control, with cell viability close to 100% ( Figure S7b ). This confirmed that PS*-1 did not have a cytotoxicity effect on mouse fibroblast cells. Separately, a hemolysis assay was performed using red blood cells. No significant hemolysis activity was observed on PS*-1 as the average hemolysis percentage at both 5 mg and 10 mg were comparable to the negative control ( Table S7 ). Hence, it can be concluded that PS*-1 is biocompatible, as illustrated in both the hemo- and cyto-compatibility assessments."
} | 4,208 |
38204420 | PMC11023807 | pmc | 8,156 | {
"abstract": "Microbial cells must continually adapt their physiology in the face of changing environmental conditions. Archaea living in extreme conditions, such as saturated salinity, represent important examples of such resilience. The model salt-loving organism Haloferax volcanii exhibits remarkable plasticity in its morphology, biofilm formation, and motility in response to variations in nutrients and cell density. However, the mechanisms regulating these lifestyle transitions remain unclear. In prior research, we showed that the transcriptional regulator, TrmB, maintains the rod shape in the related species Halobacterium salinarum by activating the expression of enzyme-coding genes in the gluconeogenesis metabolic pathway. In Hbt. salinarum , TrmB-dependent production of glucose moieties is required for cell surface glycoprotein biogenesis. Here, we use a combination of genetics and quantitative phenotyping assays to demonstrate that TrmB is essential for growth under gluconeogenic conditions in Hfx. volcanii . The Δ trmB strain rapidly accumulated suppressor mutations in a gene encoding a novel transcriptional regulator, which we name t rm B s u p pressor, or TbsP (a.k.a. “tablespoon”). TbsP is required for adhesion to abiotic surfaces (i.e., biofilm formation) and maintains wild-type cell morphology and motility. We use functional genomics and promoter fusion assays to characterize the regulons controlled by each of TrmB and TbsP, including joint regulation of the glucose-dependent transcription of gapII , which encodes an important gluconeogenic enzyme. We conclude that TrmB and TbsP coregulate gluconeogenesis, with downstream impacts on lifestyle transitions in response to nutrients in Hfx. volcanii .",
"introduction": "1 | INTRODUCTION Across bacteria and archaea, cells transition between different lifestyles in response to environmental cues. For example, many bacteria transition from a motile planktonic lifestyle to a sessile adherent biofilm upon sensing a rigid surface ( Laventie & Jenal, 2020 ). Other bacteria undergo dramatic morphological changes when shifting from gliding motility to swarming during nutrient starvation ( Little et al., 2018 ). In contrast to bacteria, such lifestyle transitions are not well understood in archaea. Haloferax volcanii , a model organism for archaea, has been shown to undergo a lifestyle transition across the growth curve, shifting from rod-shaped cells during early exponential phase to disk-shaped cells during late-exponential and stationary phases ( de Silva et al., 2021 ; Li et al., 2019 ). Rod-shaped cells predominate at the motile leading edge of colonies swimming in soft agar, whereas disk-shaped cells are found in the less motile center of motility colonies ( Duggin et al., 2015 ). Genetic analyses have demonstrated that disk shape is associated with increased adhesion to abiotic surfaces, a proxy for biofilm formation ( Legerme et al., 2016 , Schiller et al., 2023 ). Conversely, rod shape is typically associated with increased motility. These data together suggest that cells transition from the rod-shaped motile cell state to the adherent biofilm-forming state as the growth curve progresses. However, the mechanisms underlying lifestyle changes in archaea, and in Hfx. volcanii in particular, remain understudied relative to those in bacteria ( Laventie & Jenal, 2020 ). Biofilm formation of Hfx. volcanii has been shown to require type IV pili-like structures as in bacteria. Adhesion pili homologous to bacterial type IV secretion systems are encoded by the genes pilA [1–6] in Hfx. volcanii ( Esquivel et al., 2013 ). Pilin proteins are thought to inhibit the synthesis of archaella through a post-translational mechanism ( Esquivel & Pohlschroder, 2014 ). Consistent with this observation, motility using archaella, the archaeal analog of flagella, is not required for adhesion unlike in bacteria ( Tripepi et al., 2010 ). The proteins of both cell surface structures are decorated with glycosyl chains. The process of N-glycosylation of pilin and archaellin subunits is required for biogenesis and function of the corresponding structures ( Esquivel et al., 2016 ; Zaretsky et al., 2019 ). However, more research is needed to understand the regulatory mechanisms governing this lifestyle transition in Hfx. volcanii given the complexity and multifactorial phenotypes involved. In general, Hfx. volcanii is an excellent model system for investigating how microbes respond dynamically to their environment. Resident in hypersaline evaporation ponds and waters such as the Dead Sea, this organism experiences dramatic fluctuations in nutrient availability, desiccation/rehydration cycles, and oxidative stress ( Oren, 2008 ). Haloferax volcanii is a member of the hypersaline-adapted archaea, or haloarchaea, of the phylum Euryarchaeota. These organisms require high concentrations of NaCl (~1.5 to 2.5 M) for viability. Transcriptional regulation is an important mechanism enabling organisms to modulate gene expression in response to changing environmental conditions. Systems biology and transcription regulatory network analyses have shown that complex regulatory networks have been selected for in haloarchaea that enable them to respond quickly and efficiently to environmental signals ( Brooks et al., 2014 ; Martinez-Pastor et al., 2017 ). Transcription initiation in archaea utilizes both eukaryotic-like and bacterial-like proteins ( Grohmann & Werner, 2011 ; Martinez-Pastor et al., 2017 ). The archaeal basal transcription machinery resembles that found in eukaryotes, including a TATA box promoter sequence, which binds a preinitiation complex comprised of TATA binding protein and TFIIB homologs. This complex then recruits a eukaryote-like RNA polymerase containing between eight and 13 subunits ( Tenorio-Salgado et al., 2011 ). Archaeal genes, in contrast, are often organized in operons and can be cotranscribed in polycistronic mRNAs, similar to bacterial mRNAs ( Tenorio-Salgado et al., 2011 ). Furthermore, most transcription factors identified in archaea are homologous to bacterial activators and repressors rather than eukaryotic transcription factors ( Martinez-Pastor et al., 2017 ; Perez-Rueda & Janga, 2010 ; Tenorio-Salgado et al., 2011 ). Despite the importance of transcription networks in archaeal response to the environment, the function of many transcription factor proteins remains to be determined. TrmB is a highly conserved archaeal transcription regulatory protein that plays a key role in the regulation of genes encoding nutrient uptake transporters and enzymes in central carbon metabolism ( Kim et al., 2016 ). Originally characterized as a repressor of the operon encoding an ABC transporter for trehalose and maltose in thermophiles ( Lee et al., 2003 ), TrmB has since been recognized as a global regulator of carbon metabolism in thermophiles ( Kanai et al., 2007 ) and haloarchaea ( Schmid et al., 2009 ; Todor et al., 2013 , 2014 , 2015 ). Our previous work in the model haloarchaeal species Hbt. salinarum NRC-1 demonstrated that TrmB is required for activation of genes encoding enzymes in gluconeogenesis and repression of those in glycolysis ( Schmid et al., 2009 ; Todor et al., 2013 ). This transcriptional activity has profound and widespread consequences for metabolite pools ( Todor et al., 2015 ). In particular, TrmB activation of gluconeogenic enzyme-coding genes is required for supplying sugar precursors that feed into N-glycosylation of the S-layer in Hbt. salinarum ( Todor et al., 2014 ). Consequently, cells deleted for trmB are round rather than the characteristic constitutive rod shape of this species throughout growth; furthermore, Δ trmB was shown to be a glucose auxotroph ( Schmid et al., 2009 ). Given the importance of N-glycosylation for the biogenesis and function of cell surface structures in Hfx. volcanii , TrmB is a prime candidate for transcriptional regulation of cell shape and lifestyle transitions in this organism. However, TrmB function has not yet been investigated in Hfx. volcanii . In this work, we identify and characterize the Hfx. volcanii ortholog of Hbt. salinarum TrmB. We used bioinformatic analysis to identify HVO_2688 in Hfx. volcanii as the closest homolog (out of 5 paralogs encoded in the genome) to the previously characterized Hbt. salinarum TrmB (VNG1451C). As in Hbt. salinarum NRC-1, the Δ trmB mutant of Hfx. volcanii is shown to be a glucose auxotroph, with growth defects rescued by the addition of glucose. Prolonged incubation of Δ trmB results in heritable phenotypic reversion, and whole genome resequencing mapped all secondary-site suppressor mutations to the gene HVO_2861. This suggests that the trmB gene is essential under gluconeogenic conditions in Hfx. volcanii . Bioinformatics, genetics, gene expression, chromatin immunoprecipitation, and quantitative phenotypic analysis characterized HVO_2861, renamed here as TbsP ( T rm B s u p pressor, or “tablespoon”), as a novel putative transcriptional regulator of adhesion, motility, and cell morphology in response to glucose in Hfx. volcanii . Moreover, TrmB and TbsP jointly regulate the transcription of gapII , encoding the gluconeogenic GAPDHII enzyme ( Tastensen & Schonheit, 2018 ), potentially linking lifestyle transitions in Hfx. volcanii to central metabolism.",
"discussion": "4 | DISCUSSION The molecular mechanisms underlying cell developmental processes such as morphology and lifestyle transitions are understudied in archaea. Here, we identify and characterize two transcriptional regulators important for these transitions in response to sugar availability. We demonstrate that TrmB is an essential regulator of genes involved in carbon metabolism in gluconeogenic conditions in Hfx. volcanii ( Figures 2 and 6b , Figure S6 ). Strains deleted for trmB accumulate suppressor mutations in tbsP , which encodes a novel, archaeal-specific transcriptional regulator implicated in cell shape, motility, and adhesion ( Figures 3 , 5 , and 9 ). In the conditions studied here, we find that TbsP binds nine locations throughout the genome to directly regulate expression of seven genes involved in phosphate transport, gluconeogenesis, and lipid synthesis ( Table 3 ). Both TbsP and TrmB are important for maintaining wild type morphology ( Figure 5 ); and expression of genes implicated in proper maintenance of cell shape is altered in both Δ trmB and Δ tbsP deletion strain backgrounds ( Figure S5 ). However, only TbsP impacts adhesion and motility ( Figure 9 ). Surprisingly, these phenotypes are sensitive to glucose, contrasting with TbsP genome-wide DNA-binding behavior ( Figure 7 ). Taken together, these bioinformatics predictions, reverse genetics experiments, genome-wide expression and binding assays, and quantitative phenotyping data suggest that TbsP is dispensable for cell growth but necessary for normal adhesion, motility, and cell morphology ( Figures 5 , 6 , and 9 ). In terms of explaining the mechanism of suppression of Δ trmB lethality by deletion of tbsP , only three genes, all involved in gluconeogenesis, were distinguished as potential shared targets of both TrmB and TbsP ( Figures 6 and 7 ). Of those, only gapII is a direct target of TbsP ( Figure 7 , Table 3 ), suggesting that gapII may be the sole locus underlying suppression. Consistent with epistatic suppression of trmB by tbsP , we find that both TFs oppositely regulate gapII in response to sugar ( Figures 6 and 8 ). Furthermore, in the absence of both regulators, basal expression is upregulated 5.8-fold relative to the parent strain and likely sufficient to restore gluconeogenic activity ( Figure 8c , Table S4 ). Although the majority of TbsP-binding sites do not exhibit glucose dependent affinity, ChIP-seq is semi-quantitative and there is slight evidence that TbsP is enriched at the gapII promoter in the presence of glucose ( Figure 7a ). Competition for partially overlapping binding sites could explain glucose-specific expression of gapII ( Figure 8d ). While promoter activity data show that both TrmB and TbsP interact with the gapII promoter ( Figure 8e ), it remains unclear how changes in TF abundance or binding affinity impact regulation. TrmB expression is unexpectedly higher in the presence of glucose, and even further elevated when tbsP is deleted, suggesting feedback between the two regulators that awaits characterization ( Figure S5b ). No other characterized TrmB homologs in haloarchaea exhibit glucose-dependent differential expression ( Hackley et al., 2023 , Schmid et al., 2009 ), suggesting that TrmB in Hfx. volcanii is regulated at the level of transcription and that this regulation is species-specific. Gene regulatory network interactions between TrmB and other regulators of central metabolism is also of interest based on results presented here. For example, the glycolytic enzyme-coding genes gdh , gad , and kdg are predicted to be repressed by TrmB in the absence of glucose ( Figure 6 ). Previous studies demonstrated that these genes are also activated by the novel TF GcfR in the presence of glucose ( Johnsen et al., 2023 ). Nevertheless, our data are consistent with the interpretation that tbsP is epistatic to trmB in terms of growth, motility, and adhesion. Given our current data, it is unlikely that TbsP regulates morphology, adhesion, and motility via direct transcriptional control. No known genes involved in pili or archaella synthesis were nearby TbsP-binding sites, TbsP-binding motifs, or differentially expressed in the conditions tested. algJ (HVO_1517) is upregulated when tbsP is deleted, though the expression change did not pass our significance cutoff (LFC 1.68, adjusted p = 0.052). AlgJ is responsible for adding the first hexose to the cell wall surface S-layer protein and archaellin via the N-glycosylation pathway ( Kaminski et al., 2010 ). Deletion of algJ results in reduced N-glycosylation and compromised s-layer integrity ( Kaminski et al., 2010 ), as well as loss of motility ( Tripepi et al., 2012 ). Because algJ expression is not glucose dependent, it is unlikely the sole gene underlying Δ tbsP hypermotility. Comparison with proteomics data enabled the identification of several candidates, namely acs2 and acs4 , involved in long-chain fatty acid synthesis ( Kuprat et al., 2021 ). However, none exhibit expression patterns concordant with the observed phenotypes ( Figure S5 ). Thus, the genetic determinants underlying adhesion and motility regulation by TbsP are likely to be complex and require further investigation. Rather, the results presented here are consistent with the hypothesis that TbsP regulates cell morphology, adhesion, and motility indirectly. For instance, it is possible that the phenotypes obtained are a downstream effect of its repression of gapII ( Figure 8 ), or of transcriptional feedback regulation between TrmB and TbsP ( Figure S6c ). In Hbt. salinarum , TrmB is required for activation of gluconeogenic enzyme-coding genes, which limits the supply of sugar precursors needed to assemble N- and O-linked glycans that decorate the cell surface protein ( Todor et al., 2014 ). N-glycosylation is required for several linked cellular processes in hypersaline adapted archaea, including biosynthesis and stability of archaella and type IV adhesion pili in Hfx. volcanii ( Esquivel et al., 2013 ; Pohlschroder & Esquivel, 2015 ; Tripepi et al., 2012 ; Zaretsky et al., 2019 ), and for maintaining rod shape in Hbt. salinarum ( Todor et al., 2014 ). The cell shape defect in the Hbt. salinarum Δ trmB mutant is an indirect effect of TrmB transcriptional regulation of gluconeogenesis because N-glycosylation of the S-layer protein is necessary for the characteristic rod shape ( Mescher et al., 1974 ; Mescher & Strominger, 1976 ; Todor et al., 2014 ). In the current study, quantitative techniques did not detect a shape defect of Δ trmB in Hfx. volcanii . However, we observed by visual inspection of micrographs that Δ trmB cell shape appeared more uniformly round as cells were deprived of glucose than that of the more pleiomorphic Δ pyrE parent strain ( Figure 5c , Figure S4 ). TrmB appeared unrelated to motility and adhesion under the conditions tested. In summary, data presented here are consistent with the hypothesis that TbsP and TrmB play an important role in regulating nutrient-mediated cell state transitions in Hfx. volcanii , perhaps via joint regulation of central metabolic functions. Our data further suggest that the positive synergistic interaction observed between trmB and tbsP genes arises, at least in part, via the co-regulation of gapII by TrmB and TbsP. Their regulatory effects are expected to reinforce a strong and sudden induction of gapII as glucose supply is exhausted (i.e., TbsP-mediated derepression together with TrmB-mediated induction). The precise molecular mechanisms await future experiments with high resolution time courses. The lifestyle switch between foraging and adhesion is likely highly nuanced, indicated by the large cast of protein actors identified here and in previous studies in Hfx. volcanii , including transcriptional regulators such as TrmB family member TrmBL1 ( Schiller et al., 2023 ). Further, glucose and other hexose sugars are precursors for N-glycosylation of pili involved in adhesion and subsequent biofilm formation ( Esquivel et al., 2016 ; She et al., 2019 ), implicating central carbon metabolism in this cellular switch. Adhesion and motility in Hfx. volcanii are therefore expected to be controlled at multiple regulatory levels, with TbsP as an important component of the regulatory network."
} | 4,449 |
23871659 | null | s2 | 8,158 | {
"abstract": "Plants can develop an enhanced defensive capacity in response to infection by arbuscular mycorrhizal fungi (AMF). This 'mycorrhiza-induced resistance' (MIR) provides systemic protection against a wide range of attackers and shares characteristics with systemic acquired resistance (SAR) after pathogen infection and induced systemic resistance (ISR) following root colonisation by non-pathogenic rhizobacteria. It is commonly assumed that fungal stimulation of the plant immune system is solely responsible for MIR. In this opinion article, we present a novel model of MIR that integrates different aspects of the induced resistance phenomenon. We propose that MIR is a cumulative effect of direct plant responses to mycorrhizal infection and indirect immune responses to ISR-eliciting rhizobacteria in the mycorrhizosphere."
} | 206 |
35274289 | PMC9313621 | pmc | 8,161 | {
"abstract": "Abstract Biofilms are communities of bacterial cells encased in a self‐produced polymeric matrix and exhibit high tolerance towards environmental stress. Despite the plethora of research on biofilms, most biofilm models are produced using mono‐interface culture in static flow conditions, and knowledge of the effects of interfaces and mechanical forces on biofilm development remains fragmentary. This study elucidated the effects of air–liquid (ALI) or liquid–liquid (LLI) interfaces and mechanical shear forces induced by airflow and hydrodynamic flow on biofilm growing using a custom‐designed dual‐channel microfluidic platform. Results from this study showed that comparing biofilms developed under continuous nutrient supply and shear stresses free condition to those developed with limited nutrient supply, ALI biofilms were four times thicker, 60% less permeable, and 100 times more resistant to antibiotics, while LLI biofilms were two times thicker, 20% less permeable, and 100 times more resistant to antibiotics. Subjecting the biofilms to mechanical shear stresses affected the biofilm structure across the biofilm thickness significantly, resulting in generally thinner and denser biofilm compared to their controlled biofilm cultured in the absence of shear stresses, and the ALI and LLI biofilm's morphology was vastly different. Biofilms developed under hydrodynamic shear stress also showed increased antibiotic resistance. These findings highlight the importance of investigating biofilm growth and its mechanisms in realistic environmental conditions and demonstrate a feasible approach to undertake this study using a novel platform.",
"conclusion": "5 CONCLUSIONS The dual‐channel microfluidic platform developed to investigate biofilm development in this study has been demonstrated as a successful in vitro tool to emulate biofilm growth in a realistic physiobiological environment. Discoveries from this study support the need to develop innovative approaches to characterize biofilms and their phenotypes longitudinally based on their physical properties, such as cell density, thickness, permeability, and mechanical properties, an approach that could potentially enhance our overall understanding of biofilms.",
"introduction": "1 INTRODUCTION Biofilms are microbial communities encased within a self‐produced extracellular polymeric substance (EPS). The extracellular matrix regulates essential features of the biofilm, such as their adherence to surfaces, hydration, permeability, and tolerance to mechanical forces (Stoodley et al., 2002 ). The formation of biofilm commences when planktonic bacteria adhere to a surface and forms irreversible attachments. The anchored bacteria multiply and produce EPS, gradually forming a complex 3D biofilm structure that matures over time, which ensues infectious bacteria's dispersal to colonize other areas (Ranganathan, 2014 ). This biofilm feature enhances their ability to propagate and develop tolerance against adverse conditions, including limited nutrients and when exposed to high concentrations of antimicrobial agents such as biocides, antibiotic and antifungal compounds (Lebeaux et al., 2013 ). It has been estimated that 65%−80% of human infections are associated with biofilms, with Pseudomonas aeruginosa biofilms as the major cause of many chronic lung infections such as cystic fibrosis, chronic obstructive pulmonary disease, and bronchiectasis (Maurice et al., 2018 ). Established biofilm infection in the lung is extremely difficult to eradicate. In addition to biofilm's deleterious impact on human health, they also present formidable challenges to many other industries, such as the fouling and corrosion of pipes, which drastically reduce the pipes' life span and present imminent health risks contaminating the water. In agriculture, biofilms may colonize the surfaces and interiors of plants which negatively impact the health and growth of the plant, and this alone is associated with more than 10% of annual loss of global food production (Strange & Scott, 2005 ). The growth of biofilms can be affected by the properties of the surfaces they adhere to, such as surface roughness (Cowle et al., 2020 ), topography (Scheuerman et al., 1998 ), and hydrophobicity (Callow & Fletcher, 1994 ). Biofilm proliferation is also likely affected by environmental conditions (e.g., temperature, humidity, and dynamic flow conditions) (Toyofuku et al., 2016 ), although the detailed effects of these factors on biofilm development remain unclear. In general, there are four different types of interfaces where biofilm develops: air–liquid interface (ALI), liquid–liquid interface (LLI), solid–liquid interface (SLI), and solid–air interfaces (SAI). It is important to model biofilms by considering these interfaces because critical aspects of biofilm developmental processes, such as their attachment, nutrient uptake, mass exchange, are likely incontrovertibly linked to the interfaces. Hence, biofilms proliferating in the respiratory system should be investigated in an ALI environment, whereas biofilms in the urinary tract or interior of the plant should be cultured in LLI. Biofilms in pipes, ship hulls, and medical devices embedded in the sea should be studied using the SLI interface. Moreover, in most of the scenarios, biofilms grow in dynamic conditions where airflow and/or liquid flow are involved. The shear stresses induced by these flows will modulate the development of biofilms. The majority of experimental biofilm models investigated to date were developed on SLI, and they were cultured using microtiter plate (Peeters et al., 2008 ), CDC biofilm reactor (Williams & Bloebaum, 2010 ), rotating disk biofilm reactor (Schwartz et al., 2010 ), the Calgary Biofilm Device (Ceri et al., 1999 ) or microfluidic devices (Kim et al., 2010 ). Only a handful of studies have studied biofilms cultured on LLI, and they are also limited to studying biofilms under static flow conditions (Rühsa et al., 2014 ). ALI biofilms have previously been produced and studied using the Transwell plate (Woodworth et al., 2008 ) and the drip flow reactor (Goeres et al., 2009 ). However, in both systems, understanding the effects of dynamic flow conditions on biofilm growth are challenging. While the Transwell plate can only facilitate biofilm culture under a static flow environment, manipulating airflow rate in the drip flow reactor is limited to a low and small range of flow rates. In general, there remains a significant lack of knowledge of the effects of mechanical shear stresses (induced by airflow and/or hydrodynamic flow) and the multidimensional flow transport mechanisms (e.g., diffusion and convection) of flow medium across different types of interfaces and their comparisons, on biofilms growth and development. This study aimed to understand the effects of environmental factors specifically, nutrient availability, mechanical shear stress and interface type on the development of biofilm. A dynamic platform was set up in the ALI and LLI configuration using a dual‐channel microfluidic device (Ye et al., 2022 ) for this purpose, with the dynamic system providing a controllable air or liquid flow rate for biofilm culture. We hypothesized that biofilm properties, as determined by the viable cell number, spatial distribution, resistance to physical disruption, permeability, and antibiotic susceptibility to a model antibiotic (ciprofloxacin—CIP) was significantly different between biofilms cultured in ALI and LLI. We further hypothesized that mechanical shear stresses would result in significant changes in biofilm physical structures but could enhance bacteria growth, at least at the surface of the biofilms contacting the dynamic flow.",
"discussion": "4 DISCUSSION Understanding how biofilm properties may be affected by different environmental conditions such as determining the effects of airflow and hydrodynamic flow‐induced shear forces on biofilm growth is important to inform the development of effective biofilm eradication strategies. To the best of the authors' knowledge, this study presents the first comparison of ALI and LLI biofilms cultured using flow conditions associated with mechanical shear stresses of the same magnitude. While the superficial appearances of biofilms cultured under different conditions appear similar under visual inspection, results from the analysis provide clear evidence that biofilm's developmental process and structure is dramatically different when cultured on ALI and LLI, with dynamic flow conditions having significant effects on various aspects of biofilms' properties and especially on the permeability, morphology, and proportion of live cells distribution profile across the biofilm thickness. Furthermore, compared to LLI biofilms subjected to hydrodynamic forces, this study shows that ALI biofilms exposed to dynamic airflow result in thicker, denser, but less permeable biofilm with more homogenous surfaces. 4.1 Nutrient‐related biofilm developed on ALI and LLI (static culture vs. dynamic culture τ w = 0 Pa) Using a dual‐channel microfluidic device, biofilms were cultured under static flow or dynamic flow conditions and replicated a nutrient limiting environment (Ye et al., 2022 ) or an environment where nutrients are supplied continuously. As demonstrated in this study, the distinct difference in biofilm's physical properties obtained from these two different culture methods (static vs. dynamic flow) shows that nutrient supply plays an important role in biofilm growth. While several existing studies have reported the effect of nutrients and showed how high nutrient concentration promotes biofilm growth (Cunningham et al., 1991 ; Liu et al., 2019 ; Stoodley et al., 1998 ; Vandevivere & Baveye, 1992 ), our study shows further that with continuous nutrient supply, bacteria proliferated within 48 h, and this time frame is sufficient to form significantly thicker and denser biofilms with lower permeability that has higher antibiotic tolerance compared to the biofilms cultivated under nutrient‐limited conditions. Such information is not only useful to estimate the time of biofilm establishments in an ALI and LLI environment but it can also be used to predict biofilm properties, including their susceptibility to eradication for a given antibiotic concentration. Liu et al. ( 2019 ) demonstrated that biofilm cultured on SLI with high nutrient concentration has weak adhesive strength. However, our study shows that ALI biofilms cultured under continuous nutrient supply has reduced adhesive strength, but this is not observed in LLI biofilms. The above may be related to the difference in nutrient transportation mechanisms associated with different interfaces. Our results further show that biofilms developed with continuous nutrient supply are 10–100 times more resistant to antibiotics than those cultured under limited nutrient conditions. In addition, the MBEC for continuously nurtured biofilm is 8 times higher than mal‐nourished biofilm. This is because biofilm cultured with continuous nutrient supply produced more EPS content that effectively protects encased cells against the antimicrobials. 4.2 Interface‐related effects—ALI biofilms versus LLI biofilms Under dynamic conditions, the biofilms formed on ALI and LLI developed remarkably different structures with LLI biofilms producing significantly lower viable bacteria numbers. The differences are due to the different environmental impacts on the biofilm‐forming process when growing on different interfaces. In ALI, the matured biofilm exposed to the air is less likely to be detached from the developed biofilm structure. However, detachment is facilitated in LLI since bacteria can move more freely in the liquid submerged environment. Thus, the LLI biofilm structure is constantly changing in the fluid environment resulting in more heterogeneous structures and fewer bacterial cell numbers compared to ALI biofilm. Differences in biofilm's permeability and live‐cell distribution between the ALI and LLI biofilms are likely related to differences in mass transfer mechanisms across the interfaces. In this study, the dual‐channel microfluidic device is separated by a hydrophilic porous and permeable membrane. The pressure in the channel induced by airflow is lower than the membrane's bubble point (Hydrophilic Polyester Track Etch PETE Membranes Specification, 2016 ), which means that air is unlikely to pass through the membrane to the channel containing the media. It has been well recognized that biofilm structures are interspersed with water channels for nutrient transport and waste expel (Costerton, 1995 ; Stoodley et al., 1994 ). Yang et al.'s study (Yang & Lewandowski, 1995 ) suggest that in these heterogeneous distributions of internal water channels, transportation of nutrients in the channels are likely driven by a convective term in addition to molecular diffusion. In this study, the top and bottom channel in the microfluidic device is connected by the membrane pores. Hence, it is not difficult to envisage that the transportation of nutrients via molecular diffusion and convection is facilitated in the LLI experimental configuration and less in the gas‐liquid configuration. This may explain why LLI biofilm develops as a more permeable and porous structure, with presumably larger water channels. Another plausible evidence supporting this hypothesis is that the live bacteria proportion is constant across the LLI biofilm thickness in continuous media supply, further supporting the concept that the LLI biofilm is saturated with media, facilitating larger mass transport across the membrane and the biofilm compared to the ALI biofilm. The biofilm growth on ALI and LLI showed no significant difference in antibiotic susceptibility. Given the obvious change in live‐cell distribution and differences in physical properties of the biofilms, the reason for this is unclear and could be related to the mechanisms of actions in the antimicrobials, such as their limited ability to eradicate certain constituents of the biofilm. This is supported by the observation that the trend of continuous improvement in biofilm eradication with higher antimicrobial concentration is only observed for biofilms cultured in nutrient‐limited conditions. These results suggest that MBEC of an antibiotic should be tested by first considering the number of nutrients supplied to the biofilms of interest during the drug development phase given that the improper management of biofilms through the use of inadequate antimicrobial may lead to super‐resistant bacteria (Stewart, 2002 ). 4.3 Mechanical shear stress‐related biofilm This study shows that mechanical shear stress has a significant impact on biofilm growth and its properties. It does not significantly affect bacteria count in ALI but results in significantly lower CFU counts in an LLI biofilm culture. Mechanical shear stress also appears to have a significant impact on biofilm thickness, density, permeability, and the proportion of live‐cell distribution across the biofilm. Previous studies demonstrated that biofilms exposed to hydrodynamic shear forces are thin, have heterogeneous structures (Stoodley et al., 1998 ; Vieira et al., 1993 ), and have high adhesive strength (Chen et al., 1998 ). Our findings show that they not only correspond well with the existing work, but we further show that such properties are also associated with significantly lower permeability across the LLI biofilm. Furthermore, Hassanpourfard et al. ( 2016 ) demonstrated that hydrodynamic stresses could detach portions of biofilm, leading to extended water channels in developed structures. Hence, it is plausible that the lower permeability observed in our LLI biofilm at the higher shear stress is related to the distortion of water channels, which has increased mass transport resistance. Mechanical shear stress is ubiquitous in biological processes, with airflow in human lungs being a common example. During respiration, shear stresses in a representative Horsfield generation 10 region (the generation number commonly used in a symmetrical dichotomously branching system) (Horsfield & Cumming, 1968 ) ranges from −4 × 10 −3 to 4 × 10 −3 cmH 2 O (0–0.4 Pa) (Nucci et al., 2003 ). The impact of airflow‐induced shear stress of this magnitude is investigated in our study. We show that biofilms developed under such conditions are different from biofilms exposed to hydrodynamic shear stress of the same magnitude. The increase in airflow induced shear stress produced biofilms that are thinner, denser, and less permeable compared to their controls cultured under shear stress negligible conditions. The increase in shear stress is also associated with changes in live‐cell distribution across the biofilm thickness, especially at the outer surface, and this may be related to the biofilm being exposed to more oxygen from the higher airflow rate. It is also interesting to note that biofilm shows more homogenous planar morphology when developed under the influence of airflow. In addition, there is no significant difference in the resistance of ALI biofilms against antimicrobials, at least based on the range of shear stress investigated in this study. The effect of higher mechanical shear stress on LLI biofilms is distinctively different from the ALI biofilms. This can be readily observed in the viable CFU counts, live bacteria distribution across the biofilm and biofilm's antibiotic susceptibility. Higher hydrodynamic shear stress appears to reduce viable cell number, homogenize the distribution of live cells across the biofilm, and increase antibiotic resistance. While there appear to be insignificant changes in biofilm volume with increased hydrodynamic shear stress, extended studies to delineate the potential effects for a wider range of shear stress magnitude (e.g., in orders of magnitude) comprehensively and as a function of culturing time is warranted to improve knowledge of the mechanisms of biofilm growth in a dynamic environment. While biofilms have been widely studied, how these microorganisms thrive in dynamic mechanical environments remains unclear. Being able to understand the impact of mechanical forces and, specifically, how they may change the properties of biofilm over time would be extremely useful for a wide range of engineering applications from the perspective of developing novel products to control, monitor their growth. The above knowledge will also pave the way to enable the development of effective eradication strategies to treat biofilms. 4.4 Limitations Despite the new findings obtained from this study, there are several inherent limitations in the experimental study design. First, the impact of mechanical shear stress has been investigated within the same order of magnitude. As discussed, future work is warranted to explore the effects and implications for a wider range of mechanical shear stress. Second, high shear flow LLI biofilm cultured using a high flow rate was achieved using a peristaltic pump because the syringe pump was unable to handle the large volume of liquid associated with the high flow rate cases. However, it is important to note that the concern where the peristaltic pump may have supplied a higher level of nutrient due to recirculation of the CaMHB media is negligible given that the volume of nutrients that biofilm is exposed to during its course of growth is constant (determined by the volume of the bottom flow chamber), and the change of the media concentration is negligible considering the small volume of biofilm sample. Third, the flow rate provided by the peristaltic pump is pulsatile, which is not constant compared to the airflow provided by the syringe pump. However, implications of this are likely minor as the frequency of pulsation is high. However, in‐depth studies exploring the effects of pulsatile flow conditions are warranted as they are useful to shed light on how such dynamic flow condition alters biofilm properties common in infectious lung diseases. In addition, EPS plays an important role in biofilm surface adhesion. Unfortunately, the properties of the EPS and quantifying whether they are loose or dense have not been thoroughly investigated as it was not defined as a study's scope in this current work. Future work to include the investigation of these changes may shed more insights on biofilm growth behavior. Finally, different bacterial strains are likely to exhibit different growth trends in a given environmental condition. So far, we have studied PAO1, and further plans are currently underway to investigate the growth properties of other strains such as Escherichia coli , Staphylococcus aureus in biofilm formation. It would also be interesting to further study their co‐growth behavior, which is ubiquitous in nature and represents how biofilm usually grows and thrive realistically."
} | 5,235 |
38589497 | PMC11001880 | pmc | 8,162 | {
"abstract": "Flexible pressure sensors can convert mechanical stimuli to electrical signals to interact with the surroundings, mimicking the functionality of the human skins. Piezocapacitive pressure sensors, a class of most widely used devices for artificial skins, however, often suffer from slow response-relaxation speed (tens of milliseconds) and thus fail to detect dynamic stimuli or high-frequency vibrations. Here, we show that the contact-separation behavior of the electrode-dielectric interface is an energy dissipation process that substantially determines the response-relaxation time of the sensors. We thus reduce the response and relaxation time to ~0.04 ms using a bonded microstructured interface that effectively diminishes interfacial friction and energy dissipation. The high response-relaxation speed allows the sensor to detect vibrations over 10 kHz, which enables not only dynamic force detection, but also acoustic applications. This sensor also shows negligible hysteresis to precisely track dynamic stimuli. Our work opens a path that can substantially promote the response-relaxation speed of piezocapacitive pressure sensors into submillisecond range and extend their applications in acoustic range.",
"introduction": "Introduction The perception of touch of the human skin is enabled by mechanoreceptors that respond to not only static forces (by slow adaptors) but also vibrational stimuli (by fast adaptors) 1 . Electronic skins or flexible pressure sensors are emerging devices that mimic the functionalities of the mechanoreceptors 2 – 5 , which have been widely studied because of their potential applications in the fields of robot haptics 6 – 8 , human-machine interfaces 9 , 10 , intelligent wearables 11 – 13 , and metaverse 14 – 16 . Many applications, such as texture recognition, sound recognition, and pressure/vibration detection, require sensors to respond to both static pressure and high-frequency vibrations up to thousands of hertz. Piezocapacitive flexible pressure sensors are a class of the most widely studied sensing devices that can detect static pressure, but these devices perform insufficiently in responding to dynamic stimuli. While elastic elastomers can respond to mechanical stimuli in nanoseconds 17 – 19 , conventional piezocapacitive flexible pressure sensors often exhibit a response-relaxation time on the level of tens of milliseconds, corresponding to a narrow frequency range up to tens of hertz. This low response-relaxation speed is primarily attributed to energy dissipation associated with viscoelastic materials and interfacial frictions. Soft dielectrics are typically viscoelastic materials that dissipate energy during loading-unloading cycles. Such an energy loss gets more pronounced when softer materials are used for detecting subtle pressures 20 , 21 . Also, during the contact-separation process, the interfacial friction and adhesion between the electrode and dielectric further contribute to energy loss 22 , 23 . To improve the response-relaxation speed, a common strategy is to engineer the dielectric layer with microstructured surfaces 20 . This strategy works through two principles. First, the microstructures reduce the bulk viscoelasticity of the dielectric by storing more elastic energy in smaller deformations. Second, they reduce the contact area between the dielectric and electrode, thereby lowering energy dissipation due to interfacial friction and adhesion. However, despite the reduced energy dissipation by introducing microstructures, the response-relaxation time remains largely above 1 ms to date. This limitation seems to be unreconcilable as long as viscoelastic materials are used and interfacial gaps persist. Although there are a few very recent advances reporting sensors with a shorter response-relaxation time down to a few milliseconds 6 , 24 – 26 , such sensors can still not be used to detect high-frequency vibrations of hundreds or thousands of hertz, and thus the application of the sensors to high-frequency or acoustic purpose is still unavailable. In this work, we present a strategy for downscaling the response-relaxation time of flexible piezocapacitive pressure sensors to ~0.04 ms by seamlessly bonding a low-viscosity microstructured dielectric with the electrode. The dielectric is made by dispersing 2 wt.% carbon nanotubes (CNTs) within a polydimethylsiloxane (PDMS) matrix, which reduces the material viscosity and surface adhesion. Without interfacial gaps, the bonded microstructured interfaces substantially diminish the friction-induced energy dissipation. We show that our sensor can quickly respond to stimuli from steady pressures to high-frequency vibrations over 10 kHz. In addition, the sensor also exhibits a high frequency-resolution of 0.2 Hz at 1000 Hz, and negligible capacitance-pressure hysteresis. Such behaviors enable its applications for dynamic pressure detection including acoustic scenarios. We further designed an artificial ear system based on the sensor and used the system for sound detection. We expect that our sensor to be used in more applications that require the detection of both static pressure and high-vibrational stimuli, and the method of using a bonded interface to improve response-relaxation speed might be extended to other devices.",
"discussion": "Discussion In conventional capacitive soft pressure sensors, a common issue is the presence of gaps between the electrode and the viscoelastic dielectric layer. This often leads to high energy dissipation, resulting in response-relaxation times exceeding 10 ms. Such durations limit their effectiveness in detecting high-frequency signals. In our study, we address this limitation by reducing the viscosity of the PDMS dielectric through the incorporation of 2 wt.% CNTs. Additionally, we have engineered a bonding between microstructured microcones and the electrode, effectively reducing the response-relaxation time to ~0.04 ms. We observed that the friction-induced energy dissipation is directly proportional to the contact area created during deformation. Our findings reveal that sharp microcones with larger bonded heads are more efficient in minimizing energy dissipation upon compression. However, a drawback is that these sharp microcones are prone to buckling instability, which can compromise mechanical stability. Furthermore, a larger bonded head can reduce the sensor’s sensitivity. Through our analysis, we conclude that microcones with a moderate height and bonded head size offer an optimal balance, achieving rapid response-relaxation times, high mechanical stability, and enhanced sensitivity. Our sensor design represents a significant advancement over existing PVDF-based flexible sensors, which may possess a broader frequency bandwidth 52 . Notably, our sensors can detect both static and dynamic pressures simultaneously. They are also softer and more flexible, broadening their practical applications in fields such as robotics, the metaverse, and biomedical engineering. It is also important to highlight that standard commercial LCR meters may not be able to detect the limits of response time. For instance, the 6 ms response time of our previous sensors also represented the detection threshold of the LCR meter used 27 . Interestingly, the remarkably short 0.04 ms response time of our current sensor is partly attributed to the 25,000 Hz bandwidth of our customized circuit board. This underscores the necessity of upgrading the testing systems in tandem with sensor improvements in the future. Note that there are sensors that have similar or wider bandwidth ranges, including traditional silicon-based capacitive sensors 53 , and flexible piezoelectric and triboelectric sensors 54 , 55 . However, the traditional silicon-based capacitive sensors are stiff, while the piezoelectric and triboelectric sensors are limited to the detection of dynamic signals. By contrast, our capacitive sensor is soft and has little signal drift 56 ; it can not only detect static pressure but also record high-frequency vibrations over 10 kHz, making it potential for a wider range of applications."
} | 2,024 |
30153857 | PMC6114274 | pmc | 8,165 | {
"abstract": "Background Biochemical and regulatory pathways have until recently been thought and modelled within one cell type, one organism and one species. This vision is being dramatically changed by the advent of whole microbiome sequencing studies, revealing the role of symbiotic microbial populations in fundamental biochemical functions. The new landscape we face requires the reconstruction of biochemical and regulatory pathways at the community level in a given environment. In order to understand how environmental factors affect the genetic material and the dynamics of the expression from one environment to another, we want to evaluate the quantity of gene protein sequences or transcripts associated to a given pathway by precisely estimating the abundance of protein domains, their weak presence or absence in environmental samples. Results MetaCLADE is a novel profile-based domain annotation pipeline based on a multi-source domain annotation strategy. It applies directly to reads and improves identification of the catalog of functions in microbiomes. MetaCLADE is applied to simulated data and to more than ten metagenomic and metatranscriptomic datasets from different environments where it outperforms InterProScan in the number of annotated domains. It is compared to the state-of-the-art non-profile-based and profile-based methods, UProC and HMM-GRASPx, showing complementary predictions to UProC. A combination of MetaCLADE and UProC improves even further the functional annotation of environmental samples. Conclusions Learning about the functional activity of environmental microbial communities is a crucial step to understand microbial interactions and large-scale environmental impact. MetaCLADE has been explicitly designed for metagenomic and metatranscriptomic data and allows for the discovery of patterns in divergent sequences, thanks to its multi-source strategy. MetaCLADE highly improves current domain annotation methods and reaches a fine degree of accuracy in annotation of very different environments such as soil and marine ecosystems, ancient metagenomes and human tissues. Electronic supplementary material The online version of this article (10.1186/s40168-018-0532-2) contains supplementary material, which is available to authorized users.",
"conclusion": "Conclusion MG and MT datasets have been explored mostly to learn about which and in what abundance species are present in the community. Learning about the functional activity of the community and its subcommunities is a crucial step to understand species interactions and large-scale environmental impact. Ecological questions, such as how limited availability of abiotic factors in an ocean shape most abundant genes in a community, or how temperature affects eukaryotic phytoplankton growth strategies, for example, can be approached with an accurate domain annotation and a precise functional mapping. In this respect, one might need to zoom into functional activities and metabolic pathways employed by the environmental communities that might involve non-highly expressed genes. This means searching for lowly abundant domains that, through cooperation, might imply important functional effects. In order to capture common and rare entities in a given environment, functional annotation methods need to be as precise as possible in identifying remote homology. Nowadays, the bottleneck resides in the annotation step, directly influencing an appropriate quantitative estimation of the domains. Here, we show how MetaCLADE, based on a multi-source annotation strategy especially designed for MG/MT data, allows for the discovery of patterns in very divergent sequences and provides a way to overcome this fundamental barrier. With the ongoing generation of new MG/MT data, unknown sequences will augment in number and probabilistic models are expected to play a major role in the annotation of sequences that span unrepresented sequence spaces. This point is clearly shown in our comparison with UProC, which is based on k-mer recognition, and therefore particularly adapted to the identification of already known domain sequences. By construction, UProC approach cannot be successful on unknown diverged domain sequences, a context where probabilistic domain modelling fully reveals its predictive power.",
"discussion": "Discussion MetaCLADE was especially designed to consider the partial information contained in domain fragments, localised in reads. For this, we defined a powerful two-dimensional domain-dependent gathering threshold and we use multiple models to represent each domain, possibly characterising small conserved motifs for the domain. In future development, we foresee to improve the tool in several ways (see also [ 43 ]): More domains and new models for an improved MetaCLADE annotation. New CCMs could be added to the library with the hope to reach novel and unrepresented evolutionary solutions for a domain. An obvious improvement could be obtained by extending the library with the set of new domains included in Gene3D and TIGRFAM. The motifs represented in PRINTS and ProSite could be also considered and the associated profiles handled in MetaCLADE. Note that MetaCLADE package provides the program to pre-compute gathering thresholds for all domain models. This allows the user to compute appropriate thresholds based on new CCMs. Constructing a library of conserved small motifs. The search for sequence motifs in an environmental sample might be realised with a computationally costly “all against all” read comparison. Alternatively, starting from the most conserved patterns comprised in CCMs, we can generate a repertoire of significant motifs specific of each domain in order to improve hit selection criteria. A systematic classification of these motifs might lead to datasets of motifs that could be used as environmental signatures of metabolic activities. These “environmental patterns” could be also used to find new domains in environmental samples with MetaCLADE. The advantage in this search approach, compared to an “all against all” strategy, is that patterns are constructed starting from domains, possibly functionally annotated and that this annotation could be used to associate a potential functional role to new domains discovered through the pattern. Annotation of longer sequences. Availability of long reads and read assembly in contigs allow reconstructing longer stretches, and possibly entire, ORF sequences. In this case, one could replace the third filter in MetaCLADE with DAMA [ 73 ], to reconstruct the best domain architecture as done in CLADE. Reduction of the number of redundant models in MetaCLADE. Some of the probabilistic models in MetaCLADE library are expected to be redundant, and a suitable handling of these models, after clustering, should help to increase the speed of the method and to preserve the same predictive power. Future development of MetaCLADE will reduce the number of redundant models representing domains. New criteria to filter overlapping hits in MetaCLADE. Different domain hits could be selected by exploiting further the characteristics of the two-dimensional space of sequences pre-computed for the domains. For instance, one could privilege the domain hits with larger bit-score/mean-bit-score distance from the closest negative in the space. These filtering conditions could improve the annotation and need to be tested at large scale. MetaCLADE differences with CLADE. MetaCLADE has been designed with the purpose of annotating MG and MT reads. It exploits the multi-source annotation strategy introduced in CLADE and the CLADE model library, but it handles the models and their output in a different manner. Indeed, the CLADE pipeline combines the output of its rich database of probabilistic models with a machine-learning strategy in order to determine a set of best predictions for each domain sequence. Then, DAMA [ 73 ] is used to find the best domain architecture, by using information on domain co-occurrence and by exploiting multi-objective optimisation criteria. Neither CLADE machine-learning algorithm nor DAMA are used in MetaCLADE. In fact, the characteristics of MG and MT reads, compared to full protein sequences, are their short lengths and the presence of multiple sequencing errors in them compared to full-length ORFs. Hence, they demand the design of a special computational protocol taking into account the particular nature of the data; namely: \n CCMs cannot be used with tailored GA thresholds as in CLADE. Instead, we introduce an original bi-dimensional gathering threshold that is specifically designed for evaluating short hits. For each domain, we compute a probability space on which to evaluate hits. This is done with a naïve Bayes classifier. Note that the computation of such a probability space depends on an appropriate generation of positive and negative sequences on which evaluate models for a domain. CLADE machine-learning algorithm cannot be used for protein fragment identification. Indeed, CLADE works well with the full domain annotation of known genomes. In its design, it explicitly considers E value, hit length, consensus on multiple domain hits and clade-specific hits. On the other hand, read annotation should be less sensitive to sequence errors and hit length and should disregard the species the sequence comes from. In MetaCLADE, we do not use a SVM combining the above characteristics but instead we create a simpler pipeline of hit selection. DAMA, the tool used in CLADE to reconstruct protein architectures, cannot be used on short reads. Indeed, reads might be long enough to contain at most one adjacent pair of domains and certainly cannot provide information to evaluate the contextual annotation of a domain within a potential domain architecture. In MetaCLADE, knowledge of adjacent pairs of domains could be considered but we left it for future developments."
} | 2,474 |
35243218 | PMC8867051 | pmc | 8,166 | {
"abstract": "Summary Microbial inoculations contribute to reducing agricultural systems' environmental footprint by supporting sustainable production and regulating climate change. However, the indirect and cascading effects of microbial inoculants through the reshaping of soil microbiome are largely overlooked. By discussing the underlying mechanisms of plant- and soil-based microbial inoculants, we suggest that a key challenge in microbial inoculation is to understand their legacy on indigenous microbial communities and the corresponding impacts on agroecosystem functions and services relevant to climate change. We explain how these legacy effects on the soil microbiome can be understood by building on the mechanisms driving microbial invasions and placing inoculation into the context of ecological succession and community assembly. Overall, we advocate that generalizing field trials to systematically test inoculants' effectiveness and developing knowledge anchored in the scientific field of biological/microbial invasion are two essential requirements for applying microbial inoculants in agricultural ecosystems to tackle climate change challenges.",
"conclusion": "Concluding remarks Microbial inoculations can influence agroecosystem functions in multiple ways and show great opportunities to increase services such as climate change mitigation and adaptation to climate change in agroecosystems. This potential is related to the direct effects of inoculants when they harbor relevant specific functions, as well as indirect, cascading effects through plants and/or native microbial communities. However, whereas the mechanisms underlying plant-inoculant interactions are well studied, how microbial inoculants cause changes in native microbial communities and their functioning is rarely studied. Understanding these cascading effects may help design new approaches to tackle climate change, as envisaged, for example, part of the Moonshot project “Cool Earth via Microbes in Agriculture” promoted by the New Energy and Industrial Technology Development Organization (NEDO) in Japan. Conversely, neglecting these cascading effects may lead to unintended disservices regarding climate change regulation in agroecosystems ( Bounaffaa et al., 2018 ). It remains particularly challenging to evaluate how the inoculation-induced effects change over time ( Estoup et al., 2016 ) and how they might go beyond the intended purpose owing to, for example, the long-term persistence of microbial inoculants in soil or legacy effects of inoculants on the soil microbiome ( Narożna et al., 2015 ). Understanding inoculant effects should be based on a framework of microbial invasion explicating the ecological successions of microbial populations induced by inoculation. We thus proposed a conceptual model that links microbial successional mechanisms and assembly processes to guide the analyses of the ecological consequences of microbial inoculations. We anticipate that there will be increasing empirical evidence to apply and test this conceptual model to improve our ability to understand the role of microbial inoculations (and, more generally, invasions) in soil ecosystems over space and time. In doing so, an increasing number of strategies based on microbial inoculants will likely be developed to steer agroecosystem functions and particularly to tackle the climate change challenge, and their actual benefits will have to be more systematically tested under natural field conditions. Applied at a large scale, inoculations could then contribute to pursuing agricultural sustainability in the Anthropocene. Resource availability Lead contact Further information and requests should be directed to the Lead contact, Joana Falcão Salles ( J.falcao.salles@rug.nl ). Materials availability This study did not generate new unique reagents.",
"introduction": "Introduction Increasing or even maintaining crop yields is becoming more difficult with the need to decrease the environmental footprint of agricultural practices and their potential effects on climate change ( Brisson et al., 2010 ; Pe'er et al., 2014 ; Smit and Skinner, 2002 ). With a faster-growing global market compared to agrochemicals ( Batista and Singh, 2021 ), microbial inoculants, (i.e., beneficial microorganisms or mixtures of microorganisms applied to either the soil or the plant to improve soil quality and crop productivity; hereafter called soil inoculants and plant-based inoculants, respectively), are gaining importance in enhancing the sustainability of agroecosystems. For instance, plant-based inoculants contribute to higher plant growth, yield, resistance to abiotic (e.g., higher plant resistance to drought), and biotic (e.g., soil-borne pathogens) stresses (reviewed in ( Bashan, 1998 ; Vejan et al., 2016 )). Microbial inoculation thus potentially offers nature-based solutions ( Eggermont et al., 2015 ) to climate change through their influence on plant growth and different relevant agroecosystem functions and services. Despite the benefits of plant growth, inoculation practices can potentially lead to changes in soil microbial communities, which are often neglected ( Trabelsi and Mhamdi, 2013 ). Recent reports have indicated that microbial inoculation, even when not successful, triggers significant changes at the level of indigenous soil microbial communities, following a framework linked to microbial invasions (see Glossary ). Specifically, microbial invaders (here inoculants) favor or outcompete native microbial populations, leading to changes in species diversity and community composition ( Bannar-Martin et al., 2018 ). As a consequence, the invaders can also reshape the functionality of resident soil communities, such as carbon resources utilization and CO 2 emission ( Amor et al., 2020 ; Mallon et al., 2018 ; Xing et al., 2020a , 2020b ). Such an impact can propagate to the ecosystem level, depending on the strength and duration of the invasion. Microbial inoculations can thus have widespread and unexpected influences on soil microbial communities by triggering secondary succession ( Horn, 1974 ) with cascading effects on the functions and services provided by agroecosystems. Therefore, climate change regulation or the adaptation of agroecosystems to climate change could be influenced by reshaping the soil microbiome in response to inoculant applications. Glossary Deterministic processes: predictable process governing community assembly with a determinable outcome such as selection. Ecological secondary succession: one of two main ecological successions, referring to the process of community changes which started by disturbances. Functional redundancy: the ecological phenomena that different species performed a similar or the same function in a microbial community. Microbial community assembly: rules shaping the microbial community diversity and its distribution, functions, succession, and biogeography, including four basic processes (diversification, dispersal, selection, and drift). Microbial invasion : the process by which alien microorganisms enter and affect the resident community. Microbial keystone taxa: highly connected taxa significantly affect microbiome structure and functioning relative to their abundance across space and time. Plant growth-promoting rhizobacteria (PGPR): are microbes that colonize the rhizosphere and directly or indirectly benefit plant growth and development. Resilience: the capacity for a system to recover in response to disturbances. Resistance: the ability of a system to remain unchanged when subjected to disturbances. Stochastic processes: ecological processes that control community assembly in a random manner, such as ecological drift. Tipping point: the critical threshold where a system shifts abruptly into a different state. Broadly, microbial inoculants can cause legacy effects on agroecosystem functions relevant to climate change through three main pathways: direct effects of inoculants carrying specific functions relevant to climate change; indirect effects through modified plant growth and development; and indirect effects through the reshaping of soil microbial community ( Table 1 ). Although our current understanding of the potential of microbial inoculations to tackle climate change challenges largely derives from studies on plant-based inoculants, such as plant growth-promoting rhizobacteria (PGPR) , the three pathways mentioned above are often inseparable and work together with the introduction of inoculants to influence soil functions, although their effects may be different ( Trabelsi and Mhamdi, 2013 ). Table 1 Synthetic view on the potential of microbial inoculants to steer biogeochemical processes in agroecosystems to tackle climate change (CC) challenges Type of inoculant Object Effect Modified function(s) Service regarding climate change (CC) Demonstration of effectiveness Examples of ref N 2 O-reducing bacteria Soil Increased N 2 O reduction ( A) Decreased soil N 2 O emissions CC mitigation through reduced GHG emissions In soil microcosms and the field: N 2 O emissions diminished by 28%–189% ( Akiyama et al., 2016 ; Domeignoz-Horta et al., 2016 ) Methanotrophs Soil Increased biological CH 4 oxidation (A) Decreased soil CH 4 emissions and removal of CH 4 from the atmosphere CC mitigation through reduced GHG emissions In paddy field: CH 4 emissions diminished by 6.9%–12% ( Rani et al., 2021 ) (Engineered) CO 2 -fixing microorganisms Soil Promoted microbial CO 2 sequestration (A) Reduced soil CO 2 emissions CC mitigation through reduced GHG emissions In culture medium: the CO 2 fixation rates achieved were comparable to the capacity of the autotrophic microbes ( Gong et al., 2015 ) Microorganisms producing EPS-like compounds Soil The input of organic compounds like extracellular polymeric (EPS) substances into the soil (A) Better soil aggregates formation and water-holding capacity Better crop adaptation to drought/salinity and CC mitigation via better carbon (C) sequestration In planted soil pots: dry matter yield of roots and shoots increased by 149%–527 and 85%–281% under drought stress ( Ashraf et al., 2004 ; Sandhya and Ali, 2015 ) Plant Growth Promoting Rhizobacteria (PGPR) - general Plant Stimulated root growth and development (B) Better water uptake by roots from deep soil layers and enhanced physiological traits of seedlings Better crop adaptation to drought/salinity In planted soil pots and the field: plant biomass increased vary from 11% to 87% ( Chandra et al., 2019 ; Silambarasan et al., 2019 ; Zhang et al., 2020 ) PGPR - general Plant Increased whole plant biomass production (B) Better plant carbon sequestration CC mitigation via better carbon sequestration (if plant C is well managed) In planted soil pots: plant growth and plant-derived C inputs to soil increased by an average of 42 and 91% under elevated CO 2 ( Nie et al., 2015 ) PGPR producing VOCs Plant Production of volatile organic compounds (VOCs) (B) Better germination, higher plant activities of antioxidant defense enzymes Better crop adaptation to drought/salinity In planted soil pots: plant phytohormones increased by 49%–255%; the activities of antioxidant defense enzymes increased by 9%–70% ( Yasmin et al., 2020 ) PGPR producing IAA Plant Production of phytohormone indole acetic acid (IAA) (B) Adjustment of the timing of plant flowering Better crop adaptation to CC via modulation of plant phenology In planted soil pots: plant flowering time delayed by ∼3 days ( Lu et al., 2018 ) Plant-nodulating rhizobia influencing interactions within the rhizosphere microbiome Plant Reshaped community interaction networks (though the same composition) (C) Modified interactions between microbial populations change their ability to express the genes required to help plants tolerate stresses Better crop adaptation to drought/salinity In planted soil pots: the salt stress-induced loss of plant shoot weight diminished by 50% ( Benidire et al., 2020 ) PGPR Azospirillum lipoferum Plant Increased nitrite reducer abundance (up to 60–90%) but only moderately increased abundances of N 2 O-reducers in sites with high C limitation; decreased nirS-denitrifier abundance (0 to -20%) and N 2 O reducer abundance (down to -20%) in sites with low C limitation (C) Increased gross (up to +113%) and net (+37%) N 2 O production in sites with high C limitation; decreased gross and net N 2 O productions (-15 and -40%, respectively) in sites with low C limitation Modification of CC mitigation through GHG emissions (on soils with a high C content, GHG emissions at the regional level can be increased by 2–5%) In planted soil mesocosms and the field: variable outcomes in situ , from -6% to +25% ( Bounaffaa et al., 2018 ; Florio et al., 2017 , 2019 ) We distinguish the effect directly linked to the inoculant (A) and cascading effect through plants (B) or native soil community (C). In this opinion article, we first provide an overview of the potential benefits of microbial inoculants regarding climate change mitigation and adaptation and to what extent these have been demonstrated to be effective under field conditions. We then present the three main pathways mentioned above and discuss the potential effects of inoculants by placing them in a framework of ecological successions. We argue that grasping the legacy of microbial inoculations is essential to promote the more widespread and effective use of inoculants in the face of climate change."
} | 3,372 |
23387867 | null | s2 | 8,167 | {
"abstract": "Next-generation sequencing has dramatically changed the landscape of microbial ecology, large-scale and in-depth diversity studies being now widely accessible. However, determining the accuracy of taxonomic and quantitative inferences and comparing results obtained with different approaches are complicated by incongruence of experimental and computational data types and also by lack of knowledge of the true ecological diversity. Here we used highly diverse bacterial and archaeal synthetic communities assembled from pure genomic DNAs to compare inferences from metagenomic and SSU rRNA amplicon sequencing. Both Illumina and 454 metagenomic data outperformed amplicon sequencing in quantifying the community composition, but the outcome was dependent on analysis parameters and platform. New approaches in processing and classifying amplicons can reconstruct the taxonomic composition of the community with high reproducibility within primer sets, but all tested primers sets lead to significant taxon-specific biases. Controlled synthetic communities assembled to broadly mimic the phylogenetic richness in target environments can provide important validation for fine-tuning experimental and computational parameters used to characterize natural communities."
} | 316 |
37862415 | PMC10588950 | pmc | 8,169 | {
"abstract": "Mechanosensing, the transduction of extracellular mechanical stimuli into intracellular biochemical signals, is a fundamental property of living cells. However, endowing synthetic materials with mechanosensing capabilities comparable to biological levels is challenging. Here, we developed ultrasensitive and robust mechanoluminescent living composites using hydrogels embedded with dinoflagellates, unicellular microalgae with a near-instantaneous and ultrasensitive bioluminescent response to mechanical stress. Not only did embedded dinoflagellates retain their intrinsic mechanoluminescence, but with hydrophobic coatings, living composites had a lifetime of ~5 months under harsh conditions with minimal maintenance. We 3D-printed living composites into large-scale mechanoluminescent structures with high spatial resolution, and we also enhanced their mechanical properties with double-network hydrogels. We propose a counterpart mathematical model that captured experimental mechanoluminescent observations to predict mechanoluminescence based on deformation and applied stress. We also demonstrated the use of the mechanosensing composites for biomimetic soft actuators that emitted colored light upon magnetic actuation. These mechanosensing composites have substantial potential in biohybrid sensors and robotics.",
"introduction": "INTRODUCTION Mechanosensing, the transduction of extracellular mechanical stimuli into intracellular biochemical signals, is a fundamental property of living cells, which enables living cells to detect, interpret, respond, and further adapt to external mechanical cues ( 1 – 3 ). Mechanosensing is a ubiquitous complex process and vital for the survival of living organisms in complex environments. It can occur across scales, ranging from the nanoscale ion channels ( 4 ) to microscale cells ( 5 ) and further to macroscale tissues/organs ( 6 , 7 ). Living cells can sense diverse mechanical stimuli (e.g., normal stress, shear stress, and strain) and then generate biochemical signals to modulate cellular physiological functions (e.g., activation/deactivation of ion channels, migration, proliferation, differentiation, and growth) ( 1 – 3 ). Besides biological systems, mechanosensing can also play a key role to increase functionality in various engineering applications including wearable devices, human-machine interfaces, and soft robotics ( 8 – 10 ). Although diverse materials and structures have been developed to respond to mechanical stimuli, the performance of the human-made engineering systems is still far below biological systems, in terms of sensitivity, compactness, energy efficiency, and autonomy level. Mechanoluminescence, the conversion of mechanical stimuli into light emission, can be visually detected in a dark environment. Mechanoluminescence has been widely explored in engineering materials including inorganic solids ( 11 ), ceramic particle infilled stretchable composites ( 12 ), and mechanophores modified polymers ( 13 ), demonstrating diverse applications including wearable devices, stress sensors, structure diagnosis, and biomechanical imaging ( 11 – 13 ). Compared to the widely explored flexible electroluminescent devices ( 14 – 16 ), mechanoluminescent composites are much simpler in terms of structure designs and do not require any electrical components. Mechanoluminescent composites directly convert mechanical stimuli into light emission, readily suitable for mechanosensing. However, to achieve mechanosensing in flexible electroluminescent devices, a mechanical stimulus must be converted to an electrical signal such as capacitance change first that further triggers electroluminescence by an electrical circuit ( 14 – 16 ), which is more complicated than that in mechanoluminescent composites. However, engineering mechanoluminescent materials are hardly comparable to biological systems, especially regarding sensitivity. To this end, biohybrid systems provide a promising perspective. Recently, researchers have adopted biohybrid approaches to directly integrate living organisms with synthetic materials to create devices inheriting the functionalities of the organisms ( 17 – 21 ). Examples include biohybrid actuators/robots ( 17 , 22 ), living biochemical sensors ( 23 – 25 ), and mechanical property-tunable composites ( 26 , 27 ). However, most biohybrid systems share some common drawbacks including complicated fabrications, requirements of careful maintenance, low viability under harsh conditions, and slow response to sense stimuli ( 17 – 21 ). Moreover, mechanosensing organisms are abundant in nature, but living materials are rarely developed in the laboratory. Previously, we reported biohybrid mechanoluminescent devices with fast response, long lifetime of ~1 month and minimal need for maintenance by encapsulating bioluminescent dinoflagellates, marine unicellular algae, into liquid-filled elastomeric chambers ( 28 ). In this work, we further address the limitations of our previous work such as possible leakage, poor scalability, complex fabrication for complicated shapes, and the lack of a quantitative understanding of mechanoluminescence by developing mechanoluminescent living composites. Here, we demonstrate ultrasensitive and robust mechanoluminescent living composites by embedding dinoflagellates into soft and biocompatible hydrogel matrices, within which the dinoflagellates retained their intrinsic near-instantaneous luminescent response to mechanical stress with ultrahigh sensitivity (as low as several pascals). With hydrophobic coatings, our living composites demonstrated a lifetime of ~5 months under harsh conditions (e.g., acidic and basic solutions) with minimal maintenance. We also three-dimensional (3D)–printed the living composites into large-scale (~5 cm) mechanoluminescent structures with high spatial resolution (~0.39 mm). We further markedly enhanced the mechanical properties of living composites with double-network hydrogels. Moreover, we applied well-defined loadings to the living composites and developed a mathematical model to fit the mechanoluminescence and achieved good agreements compared to experiments. Last, we demonstrated the application in soft robotics with a biomimetic swarm of soft actuators that emitted colored light upon magnetic actuation. Our mechanoluminescent living composites not only provide a high-throughput platform to study the bioluminescence of dinoflagellates but also have potential applications in biohybrid sensors and robotics.",
"discussion": "DISCUSSION Mechanosensing, the transduction of mechanical stimuli into biochemical signals, is a fundamental property of living cells ( 1 – 3 ). Here, using a biohybrid approach, we developed ultrasensitive and robust mechanoluminescent living composites by embedding marine dinoflagellates into hydrogel matrices. Biocompatible hydrogels allowed dinoflagellates to retain their intrinsic near-instantaneous luminescent response to mechanical stimuli with ultrahigh sensitivity (~stress as low as several pascals). In the dark phase, upon external deformation/force, stress/strain was transferred through the matrix toward embedded cells, which responded by emitting visible light. We demonstrated that with permeable and hydrophobic coatings, the living composites had a lifetime of ~5 months under various conditions, e.g., seawater and acidic and basic solutions. We also printed the composites into large-scale (~5 cm) mechanoluminescent structures with high spatial resolution (~0.39 mm). We further markedly enhanced the mechanical properties of the living composites with double-network hydrogels. Moreover, we proposed a phenomenological model to quantitatively predict the mechanoluminescence and achieved good agreements compared to experiments. Last, we demonstrated a biomimetic swarm of soft actuators that emitted colored light upon magnetic actuation (Supplementary Text, figs. S15 and S16, and movie S13), which are suitable for proprioceptive sensing and optical signaling in the dark ( 14 , 28 ). The excellent viability of coated living composites immersed in seawater and under other harsh conditions confirmed that the composite created a self-sustaining environment for the dinoflagellates: (i) The culture medium contained all necessary nutrients for maintenance and growth over a period of ~5 months as indicated by our viability tests; (ii) oxygen created during photosynthesis by P. lunula in the light phase was respired during the dark phase, producing carbon dioxide that was used during the subsequent light phase; and (iii) alginate is a widely-known biocompatible hydrogel. The initial cell concentration and availability of nutrients in the living composites was important for cell viability. In this study, the initial cell concentration of P. lunula culture was ~10 cells/mm 2 . Revealed by the cell viability results ( Fig. 2 and figs. S4 to S8), this initial cell concentration was suitable for the maintenance of bioluminescence over a long time and provided sufficient light intensity for mechanoluminescence. A much higher initial cell concentration would lead to a much shorter lifetime of the living composites due to the rapid consumption of available nutrients. Compared to our previous work ( 28 ), the current living composites have several advantages: (i) They do not have a leakage issue that is a common failure mechanism for fluid chamber-based devices; (ii) we can print the living composite into arbitrary and complex structures with an approximately millimeter scale resolution, which is challenging to do with fluid chamber-based devices; (iii) P. lunula was encapsulated inside the living composite, and the stress applied to them are much better controlled. With the phenomenological model proposed for the single-cell bioluminescence ( 31 ), we can quantitatively predict the mechanoluminescence of the living composite and achieved relatively good agreements with experimental results. However, it was very challenging to model the mechanoluminescence of the fluid chamber-based device because of the complex flow field of P. lunula culture; (iv) the coated living composites demonstrated a lifetime of ~5 months, which is much longer than the lifetime of ~1 month of fluid chamber-based devices; (v) a smaller force was needed for the soft living composite to activate the bioluminescence compared to deforming the much stiffer fluid elastomer chamber-based device. Therefore, the living composite is expected to have a much higher force sensitivity. Compared to flexible electroluminescent devices that usually require an external power source ( 14 – 16 ), our living composite is energetically self-sustained, performing photosynthesis to convert solar energy to chemical energy for cell maintenance and mechanoluminescence. The high sensitivity of our living composites can be compared to other widely explored mechanoluminescent materials, such as inorganic mechanoluminescent solids (e.g., SrAl 2 O 4 :Eu 2+ and ZnS:Cu 2+ /Mn 2+ ) that requires a large activation stress on the order of magnitude of megapascal to gigpascal ( 11 , 50 ), inorganic particle-filled reusable mechanoluminescent elastomers that operate at an activation stress on the order of magnitude of approximately megapascal ( 12 , 51 ), and mechanophore molecule-modified mechanoluminescent elastomers that need an activation stress on the order of magnitude of kilopascal to megapascal but are not reusable ( 13 , 52 ). Thus, our living composites outperformed those classical mechanoluminescent materials in terms of reusability and sensitivity. The response time (15 to 20 ms) of our living composites is comparable to that of the ceramic particle-filled soft mechanoluminescent materials ( 12 , 53 – 55 ). Our current work may shed some light on the future study of living materials. First, it is often challenging to construct functional materials with comparable performance as biological systems. Direct integration of biological organisms such as cells into synthetic material can be an effective way to create functional materials with unusual and desired properties ( 17 – 21 ). The current study of combining bioluminescent dinoflagellates and biocompatible hydrogels to create ultrasensitive mechanoluminescent living composites provides such an example. Second, high robustness and minimal maintenance of living materials are often essential for many potential applications ( 17 – 21 ). However, many currently developed living materials are fragile and require careful maintenance. Our studies have demonstrated the possibility of creating highly durable living materials through a careful combination of biology and synthesized materials. Third, it is worthwhile considering the printability of living materials in future studies, which can be an important factor determining their potential applications ( 18 , 19 , 27 , 56 ). Last, living material may also provide a unique platform for studying the behaviors of biological organisms. For example, the living composite developed in the current work can be used to investigate the intrinsic luminescent behaviors of dinoflagellates under mechanical stimuli, many aspects of which remain elusive from the biology perspective. Experimentally, deforming single dinoflagellate cell and monitoring its luminescence are complicated ( 30 , 31 ). In contrast, as demonstrated in our study, it is very straightforward to control/vary the loading conditions of the living composites and simultaneously monitor the light intensity of the illumination generated by the cells. Our living composite may find its application as a mechanical sensor. Because of the solid-state form, fresh samples from the same batch showed quite good repeatability of mechanoluminescence under each condition ( Fig. 5, C and D ). Moreover, the predictions from the proposed model agreed relatively well with experimental results with several fitting parameters, providing a solid basis for mechanosensing applications. However, one limitation of dinoflagellate bioluminescence as a sensor is that the light intensity decays after a few cycles of loadings due to the consumption of bioluminescence substrates (figs. S3 to S8) ( 30 , 31 , 57 ). Typically, 30 min is needed to fully recover the bioluminescence ( 57 ). However, in each loading cycle, the curve of light intensity versus time showed the same dynamics, similar to the bioluminescence of a single cell ( 31 ). This single-cell kernel function will be stretched as a result of the loading protocol, and therefore, the temporal dynamics of the emitted light will still contain information about the mechanical loading process and suggest the suitability as a mechanical sensor. A second application is to integrate mechanoluminescent living composite with optogenetically modified muscle cells, whose motion can be triggered by external blue light, to construct biohybrid robots ( 22 , 58 – 60 ). For example, an external mechanical stimulus would trigger light emission from the living composite, which provides the light source to activate the movement of the muscle cells, resulting in a mechanical feedback loop. A third application is for biomedical purposes. A recent review has discussed the potential applications of mechanoluminescent materials in biomedical fields, e.g., in vivo local light sources for precisely controllable drug release, photothermal therapy, and photodynamic therapy ( 12 ). Our living composites are highly biocompatible and can be made biodegradable, which might be suitable for these applications. Despite the promising results demonstrated above, there are still some limitations that need to be further addressed before practical applications of our mechanoluminescent living composites. First, both Ca 2+ -alginate hydrogels and Ca 2+ -alginate/PEGDA hydrogels used in this study are viscoelastic with the degradation of mechanical properties under cyclic loadings ( 44 , 45 ). Thus, the light emission of the living composite may vary from cycle to cycle due to the change of stress state, undesirable for sensing applications. Developing biocompatible and tough hydrogels with low hysteresis may tackle this problem ( 61 , 62 ). Second, for the dip-coating strategy, there was no chemical bonding between the alginate hydrogel and elastomer layer. This physical encapsulation worked well in the current study, but the debonding between the coating and hydrogel may occur under large deformations. With the adhesion strategies explored in recent years ( 63 , 64 ), using biocompatible chemistry to achieve strong bonding is feasible. Third, our current model is still at the preliminary level with fitting parameters, although it agreed reasonably well with the uniaxial tensile test of our living composites and the indentation test of a single cell ( 31 ). Our efforts display the first step toward developing a model that helps to predict and analyze the nonuniform stress field. Last, P. lunula used in the current study can only be maintained at temperatures between 18° and 27°C but cannot tolerate extreme environmental conditions ( 65 – 67 ). Similar limitations are probably shared by most living materials and devices ( 17 – 21 ). Overall, we have developed ultrasensitive, fast-response, highly robust, printable, and tough mechanoluminescent living composites, with the potential for applications biohybrid sensors and robotics."
} | 4,358 |
34522731 | PMC8426196 | pmc | 8,170 | {
"abstract": "Here, we report data of the principal component analysis (PCA) assessment and clustering analysis related to low-temperature thermal hydrolysis process (THP) for enhancing the anaerobic digestion (AD) of sludge in wastewater treatment plants (WWTPs) with primary sludge fermentation (Azizi et al., 2021). The PCA was examined to pinpoint the influence of different THP schemes on the variations of macromolecular compounds solubilization after low-temperature THP and the relative performances in enhancing methane potential in AD. We established 2 experimental setups with a total of 18 treatment conditions (3 exposure times, 30, 60, and 90 min at three temperature levels 50, 70 and 90 °C) in comparison to the untreated control samples. Scheme-1 comprises the THP of a mixture of (1:1 vol ratio) fermented primary sludge (FPS) and thickened waste activated sludge (TWAS); while scheme-2 comprised the THP of TWAS only. The factors employed in the assessment of the PCA encompassed the variations in the macromolecular compounds and other solubilization metrics. This included the variations in the levels of carbohydrates, lipids, proteins, and solubilization of chemical oxygen demand (COD) and volatile suspended solids (VSS). Furthermore, the evaluation considered the changes of volatile fatty acids (VFAs) and total ammonia nitrogen (TAN) with respect to time and temperature. The assessment of PCA classified the THP based on their differences and alterations that occurred after the treatment. The indices of the PCA assessments differed based on the factors of concern and the focus of each individual PCA assessment. In every individual PCA assessment, the respective contribution to the total variance in PCA analysis was calculated and manifested by the highest distribution of the principal components (PCs) axis PC1 and PC2. The differences in distributions of PCs after various PCA examinations can describe the relative influence of THP schemes and the most significant variables that can trigger major differences among THP conditions. The comparative differences demonstrated by PCA support the potential investigations of the efficiency of THPs conditions and their performance categories."
} | 552 |
40346188 | PMC12064722 | pmc | 8,171 | {
"abstract": "Microbial desalination cells (MDCs), as an emerging desalination technology, have attracted increasing attention in recent years due to their ability to simultaneously achieve salt removal and wastewater treatment without the need for external energy input. In this study, the performance of two MDC systems with different cathode types—a biocathode (MDC1 # ) and a permanganate cathode (MDC2 # )—was comparatively evaluated for the treatment of saline wastewater, with a particular focus on voltage output, desalination efficiency, and chemical oxygen demand (COD) removal. Experimental results showed that the average output voltage of MDC2 # reached 742.02 mV, which was significantly higher than that of MDC1 # (695.6 mV). Its maximum power density was as high as 6.22 W/m3, approximately six times that of MDC1 # . Moreover, MDC2 # exhibited a higher average chloride removal rate in the desalination chamber (32.34 mg/h), compared to 17.13 mg/h for MDC1 # , indicating superior desalination performance. However, in terms of electron recovery, MDC1 # achieved a much higher average Coulombic efficiency (28.8 ± 18.7%), nearly three times that of MDC2 # , suggesting more efficient electron utilization with the biocathode. Regarding ammonium removal, MDC1 # demonstrated a higher initial removal efficiency within the first 96 h (74.3%, with an average rate of 4.17 mg/h), but this declined sharply over time, with the later-stage rate dropping to only 0.32 mg/h (less than 10% of the initial rate). In contrast, MDC2 # maintained a relatively stable ammonium removal rate throughout the operation (ranging from 0.58 to 3.27 mg/h, with an average of 1.92 mg/h). In addition, both systems achieved stable COD removal at the anode, with efficiencies consistently above 85%. Overall, the permanganate cathode is more suitable for applications that require high voltage output and efficient desalination, whereas the biocathode shows significant advantages in organic pollutant removal and energy recovery. This study provides a theoretical foundation for the rational selection of cathode types based on the characteristics of saline wastewater, offering valuable guidance for optimizing MDC system performance.",
"conclusion": "Conclusion This study demonstrates that the cathode type has a certain influence on the electrochemical behavior and pollutant removal performance of microbial desalination cells (MDCs) treating saline wastewater. The permanganate cathode, due to its high redox potential (E⁰ = 1.7 V in acidic and 0.59 V in alkaline conditions), exhibited higher voltage output and faster chloride ion removal. However, its Coulombic efficiency was relatively low, possibly due to side reactions and electron loss during oxidation. Although permanganate is not the most eco-friendly oxidant, the concentration used in this study was low (100 mg/L), and the final precipitate was identified as manganese dioxide (MnO₂), which helps mitigate environmental risks. In contrast, the aerobic biocathode system showed more efficient and stable performance in organic pollutant removal, along with a significantly higher Coulombic efficiency, indicating greater potential for energy recovery and long-term operational sustainability. Differences in pH variation between the cathode chambers of the two systems further reflected the impact of cathodic reactions on ion migration, providing insight into internal mass transfer and performance regulation. Therefore, cathode type should be selected based on specific treatment objectives. Permanganate cathodes are suitable when high desalination efficiency and voltage output are desired, whereas biocathodes are preferable for ammonium removal, COD degradation, and improved electron utilization. However, several important aspects need to be addressed in future studies: Long-Term Performance Assessment: The study mainly focuses on short-term performance. Long-term stability, operational sustainability, and cost-effectiveness need to be evaluated, as these factors are crucial for practical application and system optimization. Real Industrial Wastewater Testing: The experiments were conducted using synthetic saline wastewater. Testing with real industrial wastewater is essential, as its composition may vary and introduce additional challenges, such as complex organic pollutants or varying ionic strengths, that could impact system performance. Economic and Environmental Considerations: While the permanganate cathode demonstrated promising performance, the economic feasibility and potential environmental impact of its use need further investigation. In particular, reagent consumption, secondary pollution, and disposal of waste products (e.g., manganese dioxide) may pose operational challenges. A more comprehensive cost–benefit analysis and environmental impact assessment are necessary to evaluate the practicality of permanganate-based systems in large-scale applications. Future research should focus on multifunctional cathode materials, environmentally friendly alternatives to permanganate, and validation under real saline wastewater conditions. Additionally, the potential transfer of organic compounds—such as acetate—from the saline chamber to the anodic chamber was not assessed in this study and should be carefully monitored in future experiments, as it may affect COD removal interpretation and system efficiency over time.",
"introduction": "Introduction Saline wastewater is primarily generated from industries such as chemical manufacturing, pharmaceuticals, and dye production. It is characterized by a complex chemical composition and high toxicity, typically containing high concentrations of salts and recalcitrant organic pollutants. If discharged without effective treatment, such wastewater can lead to soil salinization, eutrophication of water bodies, and ecological imbalance 1 .Moreover, the high chloride concentration (> 8 g/L) in the wastewater can significantly inhibit the metabolic activities of aerobic microorganisms, thereby affecting the stability and efficiency of conventional wastewater treatment systems 2 . With the rapid economic development and ongoing industrialization in China, the total volume of saline wastewater discharged has been continuously increasing, posing severe challenges to both ecological environments and water resource security. Therefore, developing efficient and sustainable treatment technologies is critical for mitigating the environmental risks associated with saline wastewater. Currently, the desalination of saline wastewater mainly relies on physicochemical methods such as reverse osmosis (RO) and membrane distillation (MD), whose principles are based on separation and concentration techniques for salt removal. However, these technologies usually suffer from high energy consumption, limited applicability, and operational complexity 3 , 4 . For example, treating 1m 3 of saline wastewater by RO consumes approximately 3–7 kWh, while multi-stage flash (MSF) desalination may consume as much as 6–8 kWh/m 3 and generate roughly 6.7 kg CO₂eq of greenhouse gases 5 . Furthermore, when dealing with saline wastewater with complex components, the treatment efficiency of these conventional methods is often constrained, making it challenging to achieve economical and effective pollutant removal 6 . In contrast, microbial desalination cells (MDC) have gradually become a research focus for saline wastewater treatment due to their advantages of low energy consumption, process flexibility, and multifunctional integration. Unlike traditional microbial fuel cells (MFCs), MDCs introduce an additional desalination chamber between the anode and cathode, forming a three-chamber structure. The working principle of MDCs is based on the metabolic activity of electroactive bacteria attached to the anode electrode. These bacteria oxidize the organic matter in the wastewater to produce electrons and protons; the electrons are then transferred through an external circuit to the cathode, where they react with an electron acceptor, resulting in a stable current output. Simultaneously, the electric field established between the anode and cathode drives the directional migration of ions in the desalination chamber: anions move through an anion exchange membrane into the anode chamber, while cations pass through a cation exchange membrane into the cathode chamber, thus achieving the desalination process. By effectively combining the degradation of organic pollutants with salt removal, MDCs not only enable simultaneous wastewater desalination and pollutant removal but also allow for partial energy recovery, demonstrating broad developmental prospects and excellent application potential 7 , 8 . Since the MDC concept was first proposed, researchers have continually enhanced its performance through reactor design optimization, proper selection of cathode electron acceptors, and the development of efficient, low-cost catalysts. For instance, Chen et al. 9 improved desalination efficiency by increasing the number of desalination chambers in a stacked MDC (SMDC); Jacobson et al. 10 optimized ion migration pathways in a tubular MDC (UMDC) by employing ion exchange membranes; and Jafary et al. 11 enhanced the stability and long-term reliability of quadruple MDCs (QMDCs) using polygonal configurations, such as quadrilateral structures. Furthermore, as MDC operation involves both anodic oxidation and cathodic reduction reactions, the type of cathode used has a certain influence on MDC performance 12 .Common abiotic cathode electron acceptors include metal ions with high redox potentials and oxygen. For example, Cao et al. 13 developed the first MDC using potassium ferricyanide as the catholyte [Fe(CN)₆ 3 ⁻ + e⁻ → Fe(CN)₆ 4 ⁻, E° = 0.37 V]. In reactors with a desalination chamber volume of 3 mL and a cross-sectional area of 9 cm 2 , desalination rates exceeding 90% were achieved when treating saltwater with initial concentrations of 5, 20, and 35 g/L. Under a load of 200 Ω, the maximum output power reached 2 W/m 2 , demonstrating that the potential difference can effectively drive the desalination process. When using permanganate as a cathode electron acceptor, its significant advantage is its high redox potential. For instance, You et al. 14 employed permanganate in a dual-chamber MFC [MnO₄⁻ + 4H⁺ + 3e⁻ → MnO₂ + 2H₂O, E° = 1.70 V] and achieved a maximum power density of 115.60 mW/m 2 —4.5 and 11.3 times higher than that achieved with potassium ferricyanide (25.62 mW/m 2 ) and oxygen (10.2 mW/m 2 ), respectively. The tubular microbial fuel cell (MFC) uses potassium permanganate as the cathode electron acceptor, achieving a maximum power density of 3986.72 mW/m 2 . In contrast, in the bio-cathode configuration, an aerobic bio-cathode is employed. Through microbial catalysis, the aerobic bio-cathode enhances electron transfer, improves the electrochemical performance of the system, and may reduce pollutant discharge. To further improve the efficiency of the oxygen reduction reaction (ORR), researchers commonly use precious metals or transition metals as catalysts 15 . Mehanna et al. 16 employed an air cathode with platinum (0.5 mg/cm 2 Pt) as the catalyst, coating four layers of PTFE on 30% moisture-resistant carbon cloth. In a desalination chamber with a volume of 14 mL and a cross-sectional area of 7 cm 2 , seawater with initial salinities of 5 and 20 g/L achieved desalination rates of 43–67%. To reduce costs and minimize the use of precious metals, researchers have developed more affordable transition metal catalysts. Li et al. 17 used cobalt-copper catalysts (Co-OMS-2, Cu-OMS-2) synthesized from metal-doped octahedral manganese dioxide molecular sieves (OMS-2). When the load resistance was 100Ω, the peak currents were 0.19 mA and 0.2 mA, respectively, demonstrating higher performance than platinum catalysts. Liu et al. 18 prepared the MNOX catalyst via electrodeposition as a replacement for precious metal catalysts, achieving a maximum power density of 772.8 mW/m 3 . Cheng et al. 19 demonstrated that CoTMPP could also serve as a cathode catalyst, replacing platinum. Given the high costs of catalysts during long-term operation, researchers have further developed cost-effective biological cathodes 20 . Electrochemically active microorganisms, as biocatalysts, not only effectively promote redox reactions and enhance desalination rates but also possess sustainability and self-regeneration abilities 21 . According to the type of terminal electron acceptor, biological cathodes can be classified as aerobic or anaerobic. Meng et al. 22 constructed a BMDC, where the anode pH remained between 6.6 and 7.6 during stable operation. When the initial salinities were 5 g/L and 10 g/L, the desalination rates after 24 h were 46.37 ± 1.14% and 40.74 ± 0.89%, respectively. After 130 days of stable operation, the BMDC achieved a maximum output power of 3.178 W/m 3 and an open-circuit voltage (OCV) of 1.118 V. Wen et al. 23 constructed an aerobic biological cathode using carbon felt and bacterial catalysts, achieving a peak voltage of 609 mV, which was 136 mV higher than that of an air cathode under the same conditions. When using 441 mL of anode solution to treat seawater with an initial salinity of 35 g/L, the desalination rate reached 92%, with a coulombic efficiency of 96.2 ± 3.8%. In summary, the development of MDCs shows that optimizing reactor structures, selecting appropriate cathode electron acceptors, and developing efficient and low-cost catalysts can effectively improve desalination efficiency and energy recovery. Based on the above considerations, this study selected representative types of bio-cathodes and abiotic cathodes to compare their performance in treating saline wastewater. In the abiotic cathode configuration, potassium permanganate was chosen as the electron acceptor due to its high redox potential [MnO₄⁻ + 4H⁺ + 3e⁻ → MnO₂ + 2H₂O, E° = 1.70 V; MnO₄⁻ + 2H₂O + 3e⁻ → MnO₂ + 4OH⁻, E° = 0.59 V]. Potassium permanganate can accept electrons and be reduced to manganese dioxide under both acidic and alkaline conditions; as the reaction proceeds, its oxidizing ability gradually diminishes, eventually forming stable MnO₂ precipitates. Experimental observations revealed that, after cyclic operation, the color of the permanganate solution gradually changed from deep purple to brown, indirectly indicating a gradual reduction in its oxidizing capacity. In the bio-cathode configuration, an aerobic bio-cathode was used. The aerobic bio-cathode enhances electron transfer through microbial catalysis, thereby improving the overall electrochemical performance of the system and potentially reducing pollutant discharge. This study investigates the impact of cathode type on the comprehensive performance of MDCs by evaluating output voltage, organic pollutant removal, and desalination efficiency, with the aim of providing a theoretical basis and experimental support for further optimization of MDC systems.",
"discussion": "Results and discussion Output voltage analysis of the MDC To effectively compare the performance differences between the bio-cathode and the potassium permanganate cathode in treating saline wastewater, the output voltage between the electrodes of MDC1 # and MDC2 # was systematically analyzed after completing a full desalination cycle. As shown in Fig. 3 a, when the chloride ion (Cl⁻) concentration in the desalination chamber dropped below 1000 mg/L, MDC1 # underwent six stable voltage generation cycles, with a total operation time of 529.3 h and an average liquid replacement cycle of 88.2 h. In contrast, MDC2 # completed seven voltage generation cycles (Fig. 3 b) within a total of 275.0 h, with an average liquid replacement cycle of only 39.3 h, approximately 44.56% of that of MDC1 # . Fig. 3 Voltage output profiles for different cathode types: ( a ) MDC1 # and ( b ) MDC2 # . The maximum and minimum electrode potential differences of MDC1 # were 744.15 mV and 661 mV, respectively, with an average potential between electrodes of 695.6 mV. For MDC2 # , the maximum and minimum electrode potential differences were 828.05 mV and 650.5 mV, respectively, with an average of 742.02 mV. This indicates that the electrode potential difference of MDC2 # was consistently higher than that of MDC1 # throughout the operation period. Additionally, during each output voltage cycle, the voltage in both systems exhibited a gradual declining trend. This could be attributed to the progressive maturation of electroactive microbial communities at the anode, leading to an accelerated consumption rate of anodic substrates during the reaction, thereby resulting in a voltage drop 33 . Current density is an important parameter that reflects the ion migration rate and the intensity of electrochemical reactions. It directly affects the ion transfer process from the desalination chamber to the adjacent chambers, thereby determining the overall desalination efficiency of the MDC system 34 . As shown in Fig. 4 , the current density curves over time reveal significant differences between the two types of cathodes. For the biocathode (MDC1 # ), the current density is relatively low at the beginning of the desalination process. As the reaction progresses, it gradually increases and eventually stabilizes after a certain period. The fluctuations in current density during this stage may be attributed to the formation of the biofilm and the gradual enhancement of microbial activity. As the biofilm adapts and matures, the current density correspondingly increases. In contrast, the current density of the potassium permanganate cathode (MDC2 # ) remains consistently high and stable throughout the entire desalination cycle, indicating that this type of cathode possesses strong electrocatalytic properties and high electron transfer efficiency. The ability of the potassium permanganate cathode to maintain a stable current output over long-term operation contributes to the improved desalination efficiency of the system. Fluctuations in current density are closely related to the intensity of electrochemical reactions 34 , 35 . The larger variations observed in the biocathode may affect the long-term stability of the system. On the other hand, the potassium permanganate cathode demonstrates superior current stability, suggesting its clear advantage in sustaining continuous current output and enhancing desalination efficiency. Fig. 4 Variation of current density over time ( a ) MDC1 # with biocathode; ( b ) MDC2 # with potassium permanganate cathode). During the operation of MDC1 # , a significant drop in cathode potential was observed upon the replenishment of electron acceptors (e.g., oxygen) to compensate for oxygen loss due to aeration (Fig. 5 a, points A, B, and C). The potential gradually recovered to its initial level over time. This phenomenon is consistent with findings by Xie et al. 36 , who reported that, under oxygen-limited conditions, cathodic potential recovery typically exhibits a time lag. The observed decrease in cathode potential may be attributed to fouling and competition from non-electroactive aerobic microorganisms colonizing the cathode surface. Although these microorganisms do not directly participate in electron transfer processes, they consume dissolved oxygen through their own respiratory metabolism, thereby reducing the oxygen availability at the cathode surface for reduction reactions. This competition for oxygen may result in electron accumulation at the cathode surface, leading to a decline in cathodic potential. In addition, as indicated at point D in Fig. 5 a, the anodic potential in the MDC1 # system dropped rapidly to below –400 mV following substrate replenishment. In contrast, MDC2 # , which utilized a high redox potential oxidant (potassium permanganate) as the cathodic electron acceptor, did not exhibit comparable potential fluctuations upon substrate replacement. As shown in Fig. 5 b, the potential difference in MDC2 # continuously declined during the first 70 h of operation. This behavior may be ascribed to several factors. First, during the initial operation phase, the colonization of exoelectrogenic bacteria on the anode was still limited, resulting in a reduced rate of electron transfer to the permanganate cathode. This in turn constrained the cathodic reaction kinetics and aggravated activation polarization, thereby shifting the cathode potential in a more negative direction. Second, permanganate ions were rapidly reduced on the graphite felt cathode, while their diffusion rate from the bulk solution to the electrode surface remained relatively low. This created local concentration depletion near the electrode surface, inducing concentration polarization that further inhibited cathodic potential rise. Fig. 5 Electrode potential profiles during a single cycle: ( a ) MDC1 # ; ( b ) MDC2 # . With continued operation and sufficient permanganate replenishment, both activation and concentration polarization effects were gradually mitigated, and the cathodic reaction became more stable. Once the electron acceptance rate at the cathode reached a dynamic equilibrium with the rate of electron release by anodic microorganisms, the system’s potential fluctuations stabilized. Meanwhile, as the substrate was gradually consumed, the metabolic activity of the electroactive microbial community on the anode declined, reducing the electron generation rate and resulting in a gradual increase in anodic potential. At approximately 70 h into the operation, replacement of both the anodic and cathodic electrolytes in MDC2 # (Fig. 5 b, points F and E) led to a marked recovery in the overall potential difference of the system. As indicated by points G and H (representing catholyte replenishment), the cathodic potential began to rise steadily. This could be due to the high concentration of permanganate in the fresh catholyte, which alleviated the previous concentration polarization and improved electron acceptance at the cathode. Simultaneously, the replenished substrate stimulated the metabolic activity of the anodic electroactive microorganisms, enhancing electron generation and transfer to the anode electrode. As a result, the anodic potential decreased, further increasing the overall potential difference of the system. After the output voltage of the MDC systems stabilized, power density and polarization curve tests were conducted, and the results are presented in Fig. 6 . During the tests, the electrolyte in the desalination chamber had a pH of 6.85 and a conductivity of 31.5 mS/cm. For MDC2 # , the catholyte (permanganate solution) had a concentration of 100 mg/L, conductivity of 329 μS/cm, and a pH of 7.28. The pH values of both the anolyte and catholyte in MDC1 # and the anolyte in MDC2 # were all 7.3, with a conductivity of 6.45 mS/cm. Under external resistances of 1000 Ω and 200 Ω, the maximum power density of MDC1 # was consistently 1.02 W/m3, while that of MDC2 # reached 6.22 W/m3, approximately six times higher than that of MDC1 # . This result indicates that, compared to a bio-cathode utilizing aerobic microorganisms as the catalyst, employing potassium permanganate as the cathodic electron acceptor can significantly enhance the power output performance of the system. In addition, potassium permanganate possesses a relatively high redox potential, with a standard electrode potential of 1.70 V (vs. SHE) under acidic conditions and 0.59 V (vs. SHE) under alkaline conditions. This higher redox potential facilitates electron migration toward the cathode, thereby improving the overall charge transfer efficiency of the system and enhancing power output performance 14 . The polarization curve was obtained using a variable external resistance method, in which the external load was gradually adjusted to measure the relationship between current and voltage. The results showed that the open-circuit voltages of MDC1 # and MDC2 # were 871 mV and 976 mV, respectively. As the external resistance gradually decreased, the current densities of both systems increased accordingly, while the output voltage declined. As shown in Fig. 6 , the slope of the polarization curve for MDC1 # was significantly steeper than that of MDC2 # , corresponding to internal resistances of 855 Ω and 151 Ω, respectively. This indicates that MDC2 # exhibits lower resistive losses and superior electron transfer performance. Moreover, the internal resistance of MDC2 # was lower than that of the manganese-based MFC developed by You et al. 14 , further confirming the significant electrochemical advantages of the system. Fig. 6 Polarization and power density curves of the microbial desalination cell. Analysis of solute removal in the desalination chamber In the initial stage of seawater desalination, the salt concentration in the desalination chamber is significantly higher than that in the anode and cathode chambers. Under the combined influence of the electric field and concentration gradient, the desalination efficiency is significantly enhanced. Specifically, the electric field drives the cations (Na⁺) in the desalination chamber toward the cathode chamber and the anions (Cl⁻) toward the anode chamber. Meanwhile, the concentration gradient between the desalination chamber and the anode/cathode chambers further promotes the diffusion of ions from the high-concentration region (desalination chamber) to the low-concentration regions (anode and cathode chambers). The synergistic effect of these two driving forces accelerates the migration and removal of salts from the desalination chamber, thereby achieving high initial desalination efficiency. As shown in Fig. 7 , the removal rate of NH₄⁺ exhibits a steadily increasing trend throughout the entire desalination cycle, reaching a maximum of over 97%, with the overall removal efficiency of MDC1 # surpassing that of MDC2 # . Detailed analysis reveals that the NH₄⁺ removal rate in the MDC1 # system follows a \"fast-then-slow\" pattern, while that of the MDC2 # system remains relatively stable. During the first 96 h of operation, the NH₄⁺ concentration in MDC1 # rapidly decreased from 538.6 mg/L to 138.56 mg/L, achieving a removal rate of 4.17 mg/h, indicating high removal efficiency. However, after 96 h, the NH₄⁺ removal rate in MDC1 # declined significantly to only 0.32 mg/h, less than 10% of the rate observed in the initial period, suggesting a gradual reduction in removal capacity over long-term operation. In contrast, the NH₄⁺ removal rate in the MDC2 # system remained relatively stable throughout the entire experiment, ranging from 0.58 to 3.27 mg/h, with an average rate of 1.92 mg/h, indicating its capability to maintain good performance under prolonged operation. Fig. 7 Degradation curve and removal efficiency of NH₄⁺ in the desalination chamber. Overall, if a short-term high-efficiency removal of NH₄⁺ is required, the MDC1 # system demonstrates superior performance and is more suitable for rapid treatment applications. Conversely, for applications that require long-term and stable operation, the MDC2 # system offers more consistent removal efficiency. Therefore, in practical engineering applications, the selection of cathode type should be based on specific treatment needs: aerobic biocathodes are preferable for short-term, high-efficiency NH₄⁺ removal, whereas permanganate cathodes are more appropriate for systems that demand long-term operational stability. As shown in Fig. 8 , the Cl⁻ removal rate gradually increased over time, exhibiting a trend similar to that of NH₄⁺, and eventually reached a peak value of approximately 90% in the later stage of the experiment. However, significant differences were observed in the Cl⁻ removal performance between the two cathode types. Specifically, the average Cl⁻ removal rate of MDC2 # was 32.34 mg/h, which was substantially higher than that of MDC1 # (17.13 mg/h), indicating that the permanganate cathode system had a stronger driving force for ion migration and removal throughout the operational period. Further analysis revealed that the Cl⁻ removal rate in MDC2 # exhibited a \"fast-then-slow\" trend, with a notable inflection point at 144 h. At this stage, the removal rate dropped sharply from the initial peak of 53.17 mg/h to 9.63 mg/h, accounting for only 12.47% of its peak value. This suggests that the removal capacity of MDC2 # declined to a certain extent during prolonged operation. In contrast, the Cl⁻ removal rate in MDC1 # remained relatively stable, fluctuating around 15 mg/h throughout the experiment, demonstrating better performance stability during long-term operation. In summary, MDC2 # exhibited higher Cl⁻ removal efficiency during the initial stage, making it more suitable for short-term intensive desalination applications, while MDC1 # showed better long-term stability, making it preferable for continuous desalination scenarios. When considering the removal of NH₄⁺, MDC1 # demonstrated superior short-term efficiency, whereas MDC2 # showed more stable NH₄⁺ removal performance. Therefore, the cathode type should be selected based on the characteristics of the target pollutants and the expected operational duration to optimize the overall system performance. It is noteworthy that when the salt concentration in the desalination chamber becomes lower than that in the anode and cathode chambers, reverse diffusion may occur—where ions migrate from the low-concentration zones (anode/cathode chambers) back into the high-concentration desalination chamber—thereby reducing the overall salt removal rate and desalination efficiency. This phenomenon is particularly pronounced in MDC systems operating under low current density conditions, which typically arise when using high external resistances (e.g., 1000 Ω) or suboptimal cathode materials with limited electrocatalytic performance that hinder the generation of high current densities 37 . Fig. 8 Variation of Cl⁻ concentration and removal efficiency in the desalination chaber. COD removal performance analysis As shown in Fig. 9 , there are significant differences in Coulombic efficiency (CE) among the MDC systems with different cathode types. Specifically, the CE of MDC1 # was 28.8 ± 18.7%, which was significantly higher than that of MDC2 # (9.8 ± 4.4%), approximately three times higher. This indicates that the bio-cathode system has a superior advantage in electron utilization. Further analysis revealed that certain bacteria can connect to each other or to electrodes via specialized structures such as cytochromes and nanowires, enabling efficient electron transfer to specific ions or compounds and thus promoting the desalination process. This stable structure optimizes the electron transfer mechanism, reducing potential electron losses 38 , thereby enhancing the overall electron utilization efficiency of the system. In contrast, potassium permanganate, as a strong oxidant, undergoes a relatively straightforward reduction reaction at the cathode, without involving complex biological metabolism or electron transport mechanisms. However, during this process, potassium permanganate may still undergo some undesirable side reactions with various substances, resulting in partial loss 39 . Therefore, although the permanganate cathode operates through a relatively straightforward reaction pathway, its electron utilization efficiency is lower compared to that of the biocathode system. Throughout the entire desalination cycle, the COD removal performance of the MDC1 # cathode, MDC1 # anode, and MDC2 # anode was thoroughly monitored. The results show that COD removal efficiency increased progressively over time, but significant differences were observed among different electrode systems, with the order of performance being: MDC1 # cathode > MDC2 # anode > MDC1 # anode (Fig. 9 ). Specifically, the COD removal efficiency of the MDC1 # anode fluctuated between 35.93 ± 0.77% and 88.12 ± 0.73%, gradually stabilizing after the fourth power generation cycle, with a final average removal efficiency of 85.30 ± 0.81%. In comparison, the MDC2 # anode showed less fluctuation, ranging from 47.13 ± 2.61% to 90.40 ± 1.36%, and remained stable at approximately 87.82 ± 1.33% after the fourth cycle. These results indicate that both anode systems exhibited high COD removal capacities and could achieve stable organic matter removal after a certain period of operation. Fig. 9 COD degradation performance and removal efficiency curves of the anode and cathode during multiple power generation cycles ( a ): MDC1 # anode, ( b ): MDC2 # anode). Compared with the anode systems, the MDC1 # cathode exhibited a more stable and efficient performance in COD removal (Fig. 10 ). Specifically, the COD removal efficiency of the MDC1 # cathode consistently remained between 78.13 ± 1.25% and 92.87 ± 0.37% , stabilizing after the third power generation cycle and maintaining a high level throughout long-term operation. This performance is likely attributed to the application of aerobic biotechnologies at the cathode. In the MDC1 # cathode system, aeration was employed to increase the dissolved oxygen concentration in the solution, which significantly promoted microbial proliferation and metabolic activity. As a result, the microbial degradation capacity was greatly enhanced, providing strong support for efficient COD removal. Based on this mechanism, the MDC1 # cathode system was able to maintain high COD removal rates throughout the entire desalination cycle and demonstrated excellent performance even in the initial stages of operation. Fig. 10 COD degradation and removal efficiency of MDC1 # cathode over multiple power generation. Changes in anode and cathode pH under varying cathode types During the entire desalination cycle, the pH values of influent and effluent in both the anode and cathode chambers of MDC1 # and MDC2 # were monitored (Fig. 11 ). A phosphate buffer system composed of monopotassium phosphate (KH₂PO₄) and dipotassium phosphate (K₂HPO₄) was added to the anolyte and catholyte of MDC1 # as well as to the anolyte of MDC2 # to maintain pH stability and mitigate drastic acidification or alkalization. Despite the buffering effect, variations in pH were still observed, particularly in the cathode chambers 40 – 42 . In MDC1 # , the anode influent and effluent pH decreased from 7.04 ± 0.21 to 6.62 ± 0.36, while the cathode effluent pH slightly increased to 7.36 ± 0.08. A similar trend was observed in the anode chamber of MDC2 # , where the influent pH was 7.04 ± 0.21 and the effluent pH decreased to 6.75 ± 0.15. However, a more pronounced pH increase was detected in the cathode chamber of MDC2 # , where the influent pH of 6.65 ± 0.12 rose sharply to 9.88 ± 0.70. This significant pH elevation is attributed to the electrochemical reduction of permanganate (MnO₄⁻), which served as the cathodic electron acceptor. According to the electrode reaction mechanism, MnO₄⁻ undergoes the following reaction at the cathode surface: MnO₄⁻ + 2H₂O + 3e⁻ → MnO₂ + 4OH⁻ (E⁰ = 0.59 V). The generation of hydroxide ions (OH⁻) during this process led to an alkaline shift in catholyte pH. This change provides direct evidence of MnO₄⁻ reduction, and the resulting MnO₂ exhibits excellent electrocatalytic properties, enhancing electron transfer and improving both desalination and power generation performance of the system 43 . Fig. 11 pH variations of influent and effluent in anode and cathode chambers under different electrode configurations."
} | 8,915 |
21375742 | PMC3060884 | pmc | 8,172 | {
"abstract": "Background Eucalyptus species are among the most planted hardwoods in the world because of their rapid growth, adaptability and valuable wood properties. The development and integration of genomic resources into breeding practice will be increasingly important in the decades to come. Bacterial artificial chromosome (BAC) libraries are key genomic tools that enable positional cloning of important traits, synteny evaluation, and the development of genome framework physical maps for genetic linkage and genome sequencing. Results We describe the construction and characterization of two deep-coverage BAC libraries EG_Ba and EG_Bb obtained from nuclear DNA fragments of E. grandis (clone BRASUZ1) digested with Hind III and BstY I, respectively. Genome coverages of 17 and 15 haploid genome equivalents were estimated for EG_Ba and EG_Bb, respectively. Both libraries contained large inserts, with average sizes ranging from 135 Kb (Eg_Bb) to 157 Kb (Eg_Ba), very low extra-nuclear genome contamination providing a probability of finding a single copy gene ≥ 99.99%. Libraries were screened for the presence of several genes of interest via hybridizations to high-density BAC filters followed by PCR validation. Five selected BAC clones were sequenced and assembled using the Roche GS FLX technology providing the whole sequence of the E. grandis chloroplast genome, and complete genomic sequences of important lignin biosynthesis genes. Conclusions The two E. grandis BAC libraries described in this study represent an important milestone for the advancement of Eucalyptus genomics and forest tree research. These BAC resources have a highly redundant genome coverage (> 15×), contain large average inserts and have a very low percentage of clones with organellar DNA or empty vectors. These publicly available BAC libraries are thus suitable for a broad range of applications in genetic and genomic research in Eucalyptus and possibly in related species of Myrtaceae , including genome sequencing, gene isolation, functional and comparative genomics. Because they have been constructed using the same tree ( E. grandis BRASUZ1) whose full genome is being sequenced, they should prove instrumental for assembly and gap filling of the upcoming Eucalyptus reference genome sequence.",
"conclusion": "Conclusions The two Eucalyptus BAC libraries described in this study represent an important milestone for the advancement of Eucalyptus genomics and forest tree research. These BAC resources have a highly redundant genome coverage (> 15×), contain large average inserts (157 and 135 kb) and have a very low percentage of clones with organellar DNA or empty vectors. This indicates that these publicly available BAC libraries are suitable for a broad range of applications in genetic and genomic research in Eucalyptus and possibly in related species of Myrtaceae , including genome sequencing, gene isolation, functional and comparative genomics. The analysis of ~0.6 Mb of BAC clone sequences generated by Roche GS FLX sequencing technology provided an overview of the composition of the Eucalyptus nuclear genome and the feasibility of using this high-throughput technology for low-cost and efficient sequencing and assembly of the targeted Eucalyptus sequences. SSRs identified within the BAC clone sequences could be used to develop new genetic markers for multiple genotyping purposes. In addition, we report the full chloroplast genome sequence of E. grandis (160,137 bp) allowing comparison of this genome with E. globulus and other plant species. Comparative analysis of the CAD2 and CCR genes between E. grandis and E. gunnii showed a high conservation of the structure of genes as well as a high identity both in the coding and non coding sequences.",
"discussion": "Results and discussion BAC library characterization Two genomic BAC libraries (EG_Ba and EG_Bb) were constructed using partially digested ( Hind III or BstY I) and size-selected nuclear DNA isolated from E. grandis (genotype BRASUZ1) and the pAGIBAC1 cloning vector, as described in Materials and Methods. Two libraries were constructed, using different restriction enzymes to avoid a biased distribution of clones along the Eucalyptus genome [ 26 - 29 ]. Each library, EG_Ba ( Hind III) and EG_Bb ( BstY I), contains 73,728 robotically picked clones arrayed into 192 384-well microtiter plates. To evaluate the average insert size of each library, BAC DNA was isolated about 384 randomly selected clones from each library, restriction enzyme digested with the rare cutter Not I, and analyzed by Pulsed-Field Gel Electrophoresis (PFGE). All fragments generated by Not I digestion contained the 7.5 kb vector band and various insert fragments (see Figure 1a, b ). Analysis of the insert sizes from the EG_Ba library showed that more than 87% of the library contained inserts >120 kb while the average insert size was 157 kb (Figure 2a ). Analysis of the insert sizes from the EG_Bb library showed that more than 89% of the library contained inserts >111 kb while the average insert size was 135 kb (Figure 2b ). Since the haploid genome of E. grandis is about 660 Mb, the library coverage is predicted to be 17 and 15 haploid genome equivalents for the EG_Ba and EG_Bb libraries, respectively: large large enough coverage to ensure these libraries will be useful for positional cloning, physical mapping and genome sequencing [ 30 ]. The estimated probability of finding any specific sequence is greater than 99.99% considering both libraries together [ 31 ]. Figure 1 NotI digest of random E. grandis BAC clones . PFGE random selected BAC clones from the a) EG__Ba and b) EG__Bb libraries. Size standards and cloning vector are indicated. Figure 2 Size distribution of the inserts in the two BAC libraries . Histogram shows the distribution of insert sizes from random selected BAC clones from the a) EG__Ba BAC library and the b) EG__Bb BAC library. BAC library screening To characterize these BAC libraries and facilitate clone identification, we prepared high-density macroarrays on nylon filters from a subset of the libraries representing 8.5× and 7.5× of genome coverage for EG_Ba and EG_Bb libraries, respectively. Membranes were hybridized with a series of pooled probes representing the chloroplast and mitochondria genome as well as with probes derived from lignin biosynthesis related genes. Contamination with extranuclear genomes was estimated at 0.55% of the total number of BAC clones in both libraries. BAC clones containing E. grandis chloroplast sequences represented about 0.48% of all BAC clones in both libraries, lower than the estimates for several other plant species libraries [ 32 - 37 ]. The mitochondrial genome was represented by 0.07% of our BAC clones; slightly higher as compared to the 0.03%, 0.012%, and 0.04% found in coffee [ 33 ], tomato [ 38 ], and banana [ 39 ] BAC libraries, respectively. To evaluate the potential of these two BAC libraries to supply genomic sequences to contain candidate genes for cell wall biosynthesis, we screened the libraries with cDNA probes for lignin biosynthesis genes, [ 14 , 40 - 42 ] and regulatory genes ( EgMyb1 , EgMyb2 , EgRAC1 ) [ 11 , 43 , 44 ]. An average of 15.6 positive clones (Table 1 ) was obtained when the EG_Ba library was screened with probes derived from the following genes: EguCAD2 (22 clones), EguCCoAOMT (19 clones), EguCCR1 (13 clones), EguF5 H (17 clones), EguHCT (13 clones), EguMyb1 (10 clones) and EguPAL (15 clones). An average of 14.2 positive clones was obtained by probing the EG_Bb macroarray against Egu4CL (15 clones), EguC3 H (5 clones), EguC4 H (6 clones), EguMyb2 (7 clones), EguCOMT (19 clones) and EguRAC1 (33 clones) genes. The results of the macroarray hybridization gene screening suggest an over-representation of positive clones for CAD2, CCoAOMT, COMT and RAC genes. However, the probes used in the macroarray hybridizations were relatively long allowing the possibility of cross-hybridization with other members of multigene families. Table 1 BAC libraries screening. cDNA probes used were either involved in the lignin biosynthetic pathway ( PAL, C4 H, HCT, C3 H, CCoAOMT, CCR, F5H/CAld5H , CAD ) or regulating this pathway ( MYB1 and 2 , RAC1 ). Gene family Gene EMBL accession library Positive clones Array hybridization (a) PCR (b) Ratio (b/a) % Phenylalanine ammonia lyase (PAL) EguPAL_ CT987001 EG__Ba 15 11 63 Cinnamate 4-hydroxylase (C4H) EguC4H CT988030 EG__Bb 6 2 33 4 coumarate CoA ligase (4CL) Egu4CL AJ244010 EG__Bb 15 12 80 Hydroxycinnamoyl-CoA:shikimate/quinate hydroxycinnamoyl transferase (HTC) EguHTC CT980202 EG_Ba 13 13 100 p -coumarate 3-hydroxylase (C3H) EguC3H CT986440 EG__Bb 5 3 60 Caffeoyl-CoA O- methyltransferase (CCoaOMT) EguCCoaOMT AF168778 EG_Ba 19 7 37 Cinnamoyl CoA reductase (CCR) EguCCR X79566 EG_Ba 13 10 77 Ferulate 5 hydroxylase/coniferaldehyde 5 hydroxylase (F5H/CAld5H) EguF5H CT987560 Eg_Ba 17 5 29 Caffeic acid/5- hydroxyoniferaldehyde O -methyltransferase (COMT) EguCOMT X74814 EG__Bb 19 12 63 Cinnamyl Alcohol dehydrogenase EguCAD2 X65631 EG__Ba 22 13 59 Rho-related small GTP-binding protein EguRAC1 DR410036 EG__Bb 33 3 9 R2R3 MYB transcription factor EguMyb1 AJ576024 EB_Ba 10 9 90 R2R3 MYB transcription factor EguMyb2 AJ576023 EG__Bb 7 6 86 Screening was made on subsets of EG_Ba and EG_Bb BAC libraries with redundant genome coverage of 8.5× and 7.5×, respectively. Number of positive clones obtained (a) and validated by PCR (b). To remove false positives and target a single genetic locus, we performed an additional confirmation by PCR, using specific primer pairs designed from available E. gunnii available cDNA sequences. On average, the estimated proportions of 62% and 57% of the hybridization positive clones were confirmed by PCR screening for the EG_Ba and EG_Bb libraries, respectively (Table 1 ). The results of the hybridization experiments compared to those obtained by PCR validation suggest that these genes may be present in duplicate or belong to multigene families present in the Eucalyptus genome, in agreement with the EST analysis of Rengel et al. [ 14 ] that found different unigene members for some of these genes. Sequencing of selected BAC clones Five BAC clones were selected, sequenced and assembled with Roche GS FLX sequencing, and Newbler assembly methodology, respectively (Table 2 ). They included one BAC clone containing the chloroplast genome (EG_Ba_35H24), one BAC clone randomly selected from the EG_Ba library (EG_Ba_18G23), and three BAC clones (EG_Ba_2B15, EG_Ba_11K15, EG_Bb_94G18) that were hybridization positive for three genes of interest - EguCCR, EguCAD2 and EguRAC1 - respectively. Table 2 Sequencing (454FLX) and assembly of selected BAC clones. BAC Gene/Probe Reads % in the assembly % repeat Scaffolds Large Contigs (> 500Kb) Observed lenght Assembly length EG__Ba_2B15 EguCCR 27,155 97.3 4.1 1 19 ~145 kb 147,199 EG__Ba_11K15 EguCAD2 30,804 97.8 1.6 2 21 ~130 kb 136,759 EG__Bb_94G18 EguRAC1 28,580 98 2.9 6 25 ~120 kb - EG__Ba_18G23 Randomly chosen 29,837 97.7 0.9 4 24 ~160 kb 174,430 EG__Ba_35H24 Chloroplast 31,081 98.3 1.72 1 4 ~170 kb 160,137 The shotgun sequencing of these BAC clones produced on average 29,491 high quality reads per clone sequenced with a mean read length of 261nt. These clones were sequenced to different levels of sequence coverage ranging from 44.6× to 62.1×. On average 98% of these reads were used to assemble the full sequences of each clone into a minimum of five and a maximum of 36 contigs, for EG_Ba_35H24 and EG_Ba_18G23, respectively. The number of large contigs (> 500 bp) varied among clones from 4 (for clone EG_Ba_35H24) to 25 (for clone EG_Bb_94G18). These contigs were then reassembled into one (clone EG_Ba_2B15) to six (clone EG_Bb_94G18) scaffolds, allowing the reconstruction of the full sequence of all BAC clones except clone EG_Bb_94G18 (6 scaffolds). In this latter case, the presence of repetitive sequences prevented our ability to order and orient four of the six. Such a problem was already reported in barley by Wicker et al. [ 45 ]. Despite the relatively restricted number of clones sequenced, our results suggest the feasibility of using 454 sequencing for rapid and cost-effective sequencing and assembly of Eucalyptus BAC clones. The increased length of the 454 reads currently achievable with the Titanium chemistry (expected size ~400-550 bp) should result in regions of high-quality finished genomic sequences. BAC clone sequences were deposited into GenBank (accession ID HM366540 to HM366544 ). Characterization of genomic nuclear BAC clone sequences RepeatMasker http://www.repeatmasker.org/ was used to estimate GC content and search for interspersed repeats and low complexity DNA sequences in the four BAC sequences (588,509 bp, 13 scaffolds). The four BAC clone sequences revealed a low number of transposable elements, and low percentage of low complexity sequences (Additional file 1 ) when compared to 10 to 35% in other plant genomes analyzed (The Arabidopsis Genome Initiative 2000; International Rice Genome Sequencing Project 2005). However, these estimates cannot be generalized to the entire E. grandis genome due to the small number of BAC clones analyzed as well as the ascertainment bias resulting from the selection of gene-containing BACs. Furthermore, it is known that the distribution of these repetitive elements vary along the genomes. Indeed, clone EG_Ba_2B15 presented the lowest repeat content (1.2% of 152,083 bp) while the highest levels were found for clones EG_Bb_94G18 (5.5% of 129,018 bp) and clone EG_Ba_11K15 (5.4% of 137,697 bp). Six putative retroelements were found within the scaffold sequences of clone EG_Ba_18G23 (1.2% total length, 3 LINES and 3 LTRs) whereas none were identified in clone EG_Ba_2B15. Four LTR class retroelements were found within the scaffold sequences of clone EG_Bb_94G18 (2.5% total length, 3 Ty1/Copia and one Gypsy/DIRS1) and EG_Ba_11K15 (3,6% total length, 4 Gypsy/DIRS1). Low complexity sequences covered 0.8%, to 1.9% of the BAC clones sequence scaffolds analysed Using Sputnik (Abajian, 1994, http://www.cbib.u-bordeaux2.fr/pise/sputnik.html) a total of 88 microsatellites also called simple sequence repeats (SSR) were found within the BAC scaffold sequences and these can be developed as new genetic markers. SSR markers have been extensively used in linkage analysis and comparative mapping of Eucalyptus species[ 10 , 46 - 48 ], genetic fingerprinting [ 49 ], population genetics [ 50 ], and for clonal fidelity assessment [ 51 ]. One possible application of these new BAC-derived genetic markers could be the anchoring of a physical map to the available Eucalyptus genetic maps. Within the BAC clone sequence scaffolds we found 57 perfect Class I SSRs, that are more likely to be polymorphic [ 52 ] as SSR mutations tend to be positively correlated with SSR length [ 53 ]. This class of SSRs was found to occur on average every 10.3 kb within the sequences of the four nuclear BAC clones sequenced, from a minimum of 7.2 kb [EG_Bb_94G18] to 15.3 kb [EG_Ba_11K15]. These differences could reflect the non-random distribution of SSR in plant genomes [ 52 , 54 , 55 ]. The frequency of ClassI SSRs observed in this study was similar to that observed by Mun et al. [ 54 ] for selected Medicago truncatula BAC clone sequences, but very low when compared to that observed in another tree species Populus , where SSRs occurred on average every 2.5 kb [ 55 ]. However, one cannot discard the possibility that the existence of gaps generated by the presence of repetitive sequences, where 454 sequencing has trouble going through, and/or due to a low sequence coverage in the region, might potentially underestimate the SSR frequency. Furthermore, the frequency of SSRs within the BAC clones is also very low when compared to that observed by Ceresini et al. [ 56 ]) and Rabello et al. [ 57 ] that reported a frequency of one Class I SSR every 2.5-2.7 kb in Eucalyptus cDNA libraries. Comparison of the structure and sequence of the CCR and CAD2 genes from E. grandis and E. gunnii Cinnamoyl CoA reductase (CCR; EC 1.2.1.44), that catalyzes the conversion of cinnamoyl CoA esters to their corresponding cinnamaldehydes and Cinnamyl alcohol dehydrogenase (CAD; EC 1.1.195) that catalyzes the conversion of these aldehydes to the corresponding alcohols are considered key enzymes in lignin biosynthesis. For instance, it was previously found that CCR down-regulated plants had lower lignin levels than controls [ 58 , 59 ] and the extractability of the lignin polymer was improved in CAD down-regulated plants [ 60 , 61 ]. Sequences of positive BAC clones for CAD and CCR probes were analyzed for gene identification via homology searches. Searches with BLASTN performed against a non-redundant databases (e-value cut off of 1e -5 ) allowed us to easily identify the E. grandis homologous genes to the E. gunnii CCR gene (within scaffold #1 of EG_Ba_2B15) and the CAD2 gene (within scaffold#2 of clone EG_Ba_11K15). Global sequence alignments performed using the Needle algorithm included within the Emboss package [ 62 ] allowed for the calculation of the percent identity between E. grandis and E. gunnii CCR and CAD sequences and to compare their intron/exon structure. Results of these structural comparisons are schematically presented in Figure 3(a, b) , and revealed that the number of exons, the intron/exon structure and junction boundaries were strictly conserved for both genes in both species. Moreover, the sequences are highly conserved particularly in the exons, with identity percentages varying between 98 and 100%. It seems that for both genes, the three first exons are slightly more conserved between the two species than the fourth and fifth exons. Figure 3 Genomic structure comparison of the CCR (a) and the CAD2 (b) genomic clones between E. grandis and E. gunnii . Global alignement was performed by the Needle software (EMBOSS package). (a) The predicted E. grandis CCR genomic sequence found in scaffold #1 (108,677 to 111,886bp in Eg__Ba_2B15) was compared to the E. gunnii CCR promoter (EMBL AJ132750 ) linked to the genomic sequence (EMBL X97433 ). (b) The predicted E. grandis CAD genomic sequence found in scaffold #2 (3,492 to 8553bp in Eg__Ba_2B15) was compared to the E. gunnii CAD genomic sequence (EMBL X75480 ). Non-coding regions were also very well conserved between E. gunnii and E. grandis CAD and CCR sequences, respectively. The CAD promoter regions exhibited sequence identity of 96% for the first 2.5 kb. Whereas the sequence conservation was lower between the CCR promoter sequences, showing 88% sequence identity within the first 500 pb upstream the transcription start. The alignment was interrupted by an insertion of 444 bp in the E. grandis sequence and the identity level in the remaining 5' sequence dropped to 85%. Concerning introns, identities between E. grandis and E. gunnii increased from 90% to 97%, the less conserved being intron 4, showing 90% and 93% identity for CAD2 and CCR respectively. Intron 1 in the E. grandis CAD2 gene exhibited a deletion of 71 nt after position 122 as compared to the corresponding intron in E. gunnii . The indels, reported here, between the CCR promoters and between introns 1 of CAD2 could be used to develop markers to discriminate these two species, as they seem highly conserved within each species (Additional file 2 ). The E. grandis chloroplast genome characterization The large size of E. grandis BAC clone inserts coupled with the library screening strategy (arrays hybridized with a pool of probes selected along the E. globulus chloroplast genome, NCBI accession number NC_008115) allowed for the identification of BAC clones with inserts that potentially contained the entire sequence of the E. grandis choloroplast genome. Clone EG_Ba_35H24 was selected for sequencing since this clone was shown to be positive for several chloroplast probes and also presented an insert size close to that of the previously sequence of E. globulus chloroplast genome (~160 kbp). The The Roche GS FLX reads were assembled into four long contigs sharing more than 99% sequence identity with the E. globulus chloroplast genome sequence. Due to the presence of inverted repeats (IRs) in the E. globulus chloroplast, a manual rearrangement of the sequences was performed based on this reference chloroplast genome [ 17 ] which allowed us to obtain a unique continuous fragment with 160,137 bp (EMBL HM347959 ), a length that is close to that observed for E. globulus (160,268 bp) [ 17 ] and Vitis (160,928 bp) [ 63 ] but larger than the one reported earlier for E. nitens (151 Kbp) based on restriction enzyme mapping [ 64 ]. The chloroplast genome of E. grandis includes a pair of inverted repeats 26,390 bp long, separated by a small, and large single copy regions of 18,478 bp and 88,879 bp, respectively. The GC-content of the E. grandis chloroplast genome is 36.9%, which is comparable to that of the E. globulus chloroplast genome and to other tree plant plastids (e.g. 38.5% in Pinus thunbergii [ 65 ], 36.7% in Populus trichocarpa [ 66 ], 37.4% in Vitis vinifera [ 63 ]). Figure 4 illustrates a very high conservation of the chloroplast genomes between the E. grandis and the E. globulus . Moreover, the annotation of the E. grandis chloroplast genome sequences reveals that gene order is conserved in the two species. The large size of the inserts in the BAC libraries allowed us to obtain the chloroplast genome sequence in one single BAC clone. The sequencing of this BAC clone with Roche GS FLX technology was as efficient as the bridging shotgun library strategy used for E. globulus [ 17 ] or Vitis [ 63 ] chloroplast genome sequencing. Such an approach could be readily applied to study other non-nuclear genomes, such as mitochondrial genomes. The complete and annotated E. grandis chloroplast genome sequence was deposited into GenBank (accession ID NC_014570 ). Figure 4 CSA algorithm output showing the conserved sequences identified between the E. grandis (EMBL HM347959 ) and E. globulus (NC_008115) chloroplast genomes ."
} | 5,588 |
38065083 | PMC10760936 | pmc | 8,173 | {
"abstract": "SUMMARY Photosynthesis is central to food production and the Earth’s biogeochemistry, yet the molecular basis for its regulation remains poorly understood. Here, using high-throughput genetics in the model eukaryotic alga Chlamydomonas reinhardtii , we identify with high confidence (false discovery rate [FDR] < 0.11) 70 poorly characterized genes required for photosynthesis. We then enable the functional characterization of these genes by providing a resource of proteomes of mutant strains, each lacking one of these genes. The data allow assignment of 34 genes to the biogenesis or regulation of one or more specific photosynthetic complexes. Further analysis uncovers biogenesis/regulatory roles for at least seven proteins, including five photosystem I mRNA maturation factors, the chloroplast translation factor MTF1, and the master regulator PMR1, which regulates chloroplast genes via nuclear-expressed factors. Our work provides a rich resource identifying regulatory and functional genes and placing them into pathways, thereby opening the door to a system-level understanding of photosynthesis.",
"introduction": "INTRODUCTION In photosynthetic eukaryotes, the photosynthetic apparatus consists of a series of protein complexes in the chloroplast thylakoid membrane that use light energy to produce NADPH, ATP, and other cellular energy carriers. 1 NADPH and ATP, in turn, power many pathways, notably CO 2 assimilation into sugar by the Calvin-Benson-Bassham metabolic cycle 2 ( Figure 1A ). 2 As a sophisticated system central to cellular fitness, hundreds of genes encoded in both the nucleus and chloroplast are required to assemble these complexes 3 and regulate their activity 3 under nuclear control. 4 In plants and green algae, this coordination is known to involve a range of different mechanisms, including post-transcriptional regulation of chloroplast-expressed genes by nuclear-encoded proteins, 5 translational regulation of chloroplast-expressed subunits by assembly intermediates of photosynthetic complexes, 6 and protease-mediated degradation of unassembled subunits. 7 Although photosynthesis and its regulation have been extensively studied for 70 years, 8 , 9 phylogenetics suggests that hundreds of genes participating in photosynthesis remain to be identified and characterized. Indeed, approximately half of the GreenCut2 genes—a set of 597 genes conserved only in the green photosynthetic eukaryotic lineage and therefore likely to be involved in photosynthesis 10 —have not been functionally characterized. Genetic screens have been done in land plants and algae to identify missing photosynthesis genes. Land plant screens have identified photosynthesis-deficient mutants based on leaf coloration, 11 , 12 seedling lethality, 13 and chlorophyll fluorescence. 14 , 15 As a complementary system to plants, the leading unicellular model eukaryotic alga Chlamydomonas reinhardtii ( Chlamydomonas ) has provided advantages of higher throughput and physiology that facilitate the identification and characterization of genes essential to photosynthesis. 16 , 17 These characteristics have been leveraged to identify and characterize many core components of the photosynthetic electron transport chain. 18 – 20 In the past decade, several hundred candidates for genes involved in photosynthesis have been uncovered by screens of two large Chlamydomonas mutant collections, Niyogi CAL 21 – 23 and CLiP. 24 , 25 However, these screens had many false positives and there are indications that fewer than half of these candidates are actually involved in photosynthesis. 25 Current challenges facing the field include (1) determining which of these candidates are genuinely involved in photosynthesis and (2) determining the functions of validated photosynthesis genes. Here, we address these two challenges by combining genetics and proteomics to identify and functionally characterize genes required for photosynthesis with high confidence on a global scale. We first identified with high confidence (false discovery rate [FDR] < 0.11) a total of 115 genes required for photosynthesis–including 70 genes whose molecular function in photosynthesis had not been previously characterized in any organism–by confirming linkage of each mutation with the observed photosynthetic defect and validating insertion site mappings. We then determined the proteomic profiles of mutants representing these genes to initiate their functional characterization, including assigning 34 of them to specific photosynthetic pathways. As proof of principle for the utility of our resource, we performed additional analyses, which revealed that five of these factors work with known factors to regulate mRNA maturation of key photosystem I (PSI) subunit PsaA. We also discovered and characterized two post-transcriptional regulators of photosynthetic apparatus biogenesis, providing insights into how cells leverage the chloroplast translation machinery and the regulation of nuclear gene expression to control photosynthetic complex abundance. Together, our dataset opens the door to rapid characterization of photosynthesis genes and provides systems-level insights into photosynthesis regulation.",
"discussion": "DISCUSSION In this study, we identified with high confidence (FDR < 0.11) 115 genes required for photosynthesis, including 70 whose functions in photosynthesis had not been previously characterized in any organism. We then showed that mutant proteomes provide key insights into the functions of these genes in photosynthesis, in many cases allowing the assignment of genes to specific pathways. We identified five ROGEs that are essential for the biogenesis of PSI. Including these genes, 76% (16/21) of genes with known functions in our dataset that lead to the depletion of an entire photosystem complex are ROGEs ( Figure 4 ), demonstrating their significant impact on photosynthesis. Growing evidence indicates that ROGEs play a regulatory role rather than being merely required for complex biogenesis 46 : different ROGEs affect different chloroplast-encoded genes, 5 are differentially transcriptionally regulated, 59 and participate in feedback loops, 6 , 76 a classical transcription network motif. 77 Moreover, several ROGEs can coregulate the same protein 56 , 76 ( Table S4 ), and the expression of photosystem proteins with a stronger effect on growth, including the largest subunit of each complex, tends to be impacted by more ROGEs ( Table S4 ). Our results further support a regulatory role for ROGEs by showing that different ROGEs can be limiting factors in different conditions: RAT2 is a limiting factor for psaA expression in the light but not in the dark ( Figures 4K and 5C – 5E ), and by discovering that multiple ROGEs are controlled by a master regulator ( Figure 6O ). Together, ROGE-mediated regulation raises the intriguing possibility that during the endosymbiosis process, as transcriptional regulation in the chloroplast was lost, 5 ROGEs evolved to replace transcription factors in a regulatory network for chloroplast-expressed proteins. In order to respond effectively to changing conditions, the cell must simultaneously regulate multiple photosynthetic complexes. Such coordinated regulation cannot be achieved by the ROGEs alone, since each regulates only one or two chloroplast-encoded proteins. 5 Our results suggest the existence of two mechanisms that operate on a larger scale to coordinate the expression of multiple complexes. First, the cell appears to leverage the chloroplast translation machinery to coregulate multiple complexes. Specifically, while translation factors MTF1 and CIF2 may look like classical housekeeping genes, our data suggest that they are leveraged for regulatory functions. Whereas classical housekeeping translation initiation factors mediate all translation, 78 MTF1 and CIF2 each affect specific subsets of chloroplast-expressed proteins, a property associated with regulatory factors. 79 CIF2 is mostly required for expression of photosynthetic machinery, whereas MTF1 loss also affects ribosomal large subunits ( Figures 6A and S7A – S7C ). Consistent with a regulatory role, MTF1 overexpression leads to overexpression of proteins downregulated in the mtf1 mutant ( Figure 6A ). The differences in the proteomic impacts of mtf1 and cif2 , combined with the differential regulation of the MTF1 and CIF2 transcripts ( Figure S7G ), suggest that MTF1 and CIF2 coordinate chloroplast gene expression in response to light and nitrogen availability. Second, our data suggest that the master regulator PMR1 regulates the mRNA levels of multiple nuclear-encoded ROGEs, thus coordinating the expression of the overall photosynthetic apparatus. We hypothesize that the higher-level regulatory mechanisms mediated by PMR1, MTF1, and CIF2 are essential for the cell’s rapid and coordinated response to changes in growth conditions. More than 65% of the 115 genes we identified as required for photosynthesis have homologs in land plants ( Figure S1H ). In most cases, the functions of these conserved genes appear to be similar in Chlamydomonas and land plants, supporting the value of Chlamydomonas as a model system and expanding the significance of our findings. Genes with no clear homologs in land plants could reflect homolog search failure due to sequence divergence 80 , 81 and/or different evolutionary innovations in the algal lineage such as the algal-specific CO 2 -concentrating mechanism (CCM), the study of which has the potential to enhance crop yields. 82 We anticipate that future studies of the genes identified here and explored in our proteomics dataset will enable further discoveries in photosynthesis. Limitations of the study Considering our FDR cutoff of 0.11, up to 11% of our hits may be false positives. We have validated by genetic rescue 12 of the 70 genes not previously known to be required for photosynthesis; future work on other genes will require independent validation. In addition, although protein localization by Venus-tagging is generally reliable, 30 , 31 increased confidence in the conclusions on cellular localization will require validation by an independent method such as immunofluorescence. 31 While we have initiated here the characterization of several of the identified genes, additional work is needed to fully characterize the molecular mechanisms by which they and other factors impact photosynthesis."
} | 2,615 |
38065083 | PMC10760936 | pmc | 8,173 | {
"abstract": "SUMMARY Photosynthesis is central to food production and the Earth’s biogeochemistry, yet the molecular basis for its regulation remains poorly understood. Here, using high-throughput genetics in the model eukaryotic alga Chlamydomonas reinhardtii , we identify with high confidence (false discovery rate [FDR] < 0.11) 70 poorly characterized genes required for photosynthesis. We then enable the functional characterization of these genes by providing a resource of proteomes of mutant strains, each lacking one of these genes. The data allow assignment of 34 genes to the biogenesis or regulation of one or more specific photosynthetic complexes. Further analysis uncovers biogenesis/regulatory roles for at least seven proteins, including five photosystem I mRNA maturation factors, the chloroplast translation factor MTF1, and the master regulator PMR1, which regulates chloroplast genes via nuclear-expressed factors. Our work provides a rich resource identifying regulatory and functional genes and placing them into pathways, thereby opening the door to a system-level understanding of photosynthesis.",
"introduction": "INTRODUCTION In photosynthetic eukaryotes, the photosynthetic apparatus consists of a series of protein complexes in the chloroplast thylakoid membrane that use light energy to produce NADPH, ATP, and other cellular energy carriers. 1 NADPH and ATP, in turn, power many pathways, notably CO 2 assimilation into sugar by the Calvin-Benson-Bassham metabolic cycle 2 ( Figure 1A ). 2 As a sophisticated system central to cellular fitness, hundreds of genes encoded in both the nucleus and chloroplast are required to assemble these complexes 3 and regulate their activity 3 under nuclear control. 4 In plants and green algae, this coordination is known to involve a range of different mechanisms, including post-transcriptional regulation of chloroplast-expressed genes by nuclear-encoded proteins, 5 translational regulation of chloroplast-expressed subunits by assembly intermediates of photosynthetic complexes, 6 and protease-mediated degradation of unassembled subunits. 7 Although photosynthesis and its regulation have been extensively studied for 70 years, 8 , 9 phylogenetics suggests that hundreds of genes participating in photosynthesis remain to be identified and characterized. Indeed, approximately half of the GreenCut2 genes—a set of 597 genes conserved only in the green photosynthetic eukaryotic lineage and therefore likely to be involved in photosynthesis 10 —have not been functionally characterized. Genetic screens have been done in land plants and algae to identify missing photosynthesis genes. Land plant screens have identified photosynthesis-deficient mutants based on leaf coloration, 11 , 12 seedling lethality, 13 and chlorophyll fluorescence. 14 , 15 As a complementary system to plants, the leading unicellular model eukaryotic alga Chlamydomonas reinhardtii ( Chlamydomonas ) has provided advantages of higher throughput and physiology that facilitate the identification and characterization of genes essential to photosynthesis. 16 , 17 These characteristics have been leveraged to identify and characterize many core components of the photosynthetic electron transport chain. 18 – 20 In the past decade, several hundred candidates for genes involved in photosynthesis have been uncovered by screens of two large Chlamydomonas mutant collections, Niyogi CAL 21 – 23 and CLiP. 24 , 25 However, these screens had many false positives and there are indications that fewer than half of these candidates are actually involved in photosynthesis. 25 Current challenges facing the field include (1) determining which of these candidates are genuinely involved in photosynthesis and (2) determining the functions of validated photosynthesis genes. Here, we address these two challenges by combining genetics and proteomics to identify and functionally characterize genes required for photosynthesis with high confidence on a global scale. We first identified with high confidence (false discovery rate [FDR] < 0.11) a total of 115 genes required for photosynthesis–including 70 genes whose molecular function in photosynthesis had not been previously characterized in any organism–by confirming linkage of each mutation with the observed photosynthetic defect and validating insertion site mappings. We then determined the proteomic profiles of mutants representing these genes to initiate their functional characterization, including assigning 34 of them to specific photosynthetic pathways. As proof of principle for the utility of our resource, we performed additional analyses, which revealed that five of these factors work with known factors to regulate mRNA maturation of key photosystem I (PSI) subunit PsaA. We also discovered and characterized two post-transcriptional regulators of photosynthetic apparatus biogenesis, providing insights into how cells leverage the chloroplast translation machinery and the regulation of nuclear gene expression to control photosynthetic complex abundance. Together, our dataset opens the door to rapid characterization of photosynthesis genes and provides systems-level insights into photosynthesis regulation.",
"discussion": "DISCUSSION In this study, we identified with high confidence (FDR < 0.11) 115 genes required for photosynthesis, including 70 whose functions in photosynthesis had not been previously characterized in any organism. We then showed that mutant proteomes provide key insights into the functions of these genes in photosynthesis, in many cases allowing the assignment of genes to specific pathways. We identified five ROGEs that are essential for the biogenesis of PSI. Including these genes, 76% (16/21) of genes with known functions in our dataset that lead to the depletion of an entire photosystem complex are ROGEs ( Figure 4 ), demonstrating their significant impact on photosynthesis. Growing evidence indicates that ROGEs play a regulatory role rather than being merely required for complex biogenesis 46 : different ROGEs affect different chloroplast-encoded genes, 5 are differentially transcriptionally regulated, 59 and participate in feedback loops, 6 , 76 a classical transcription network motif. 77 Moreover, several ROGEs can coregulate the same protein 56 , 76 ( Table S4 ), and the expression of photosystem proteins with a stronger effect on growth, including the largest subunit of each complex, tends to be impacted by more ROGEs ( Table S4 ). Our results further support a regulatory role for ROGEs by showing that different ROGEs can be limiting factors in different conditions: RAT2 is a limiting factor for psaA expression in the light but not in the dark ( Figures 4K and 5C – 5E ), and by discovering that multiple ROGEs are controlled by a master regulator ( Figure 6O ). Together, ROGE-mediated regulation raises the intriguing possibility that during the endosymbiosis process, as transcriptional regulation in the chloroplast was lost, 5 ROGEs evolved to replace transcription factors in a regulatory network for chloroplast-expressed proteins. In order to respond effectively to changing conditions, the cell must simultaneously regulate multiple photosynthetic complexes. Such coordinated regulation cannot be achieved by the ROGEs alone, since each regulates only one or two chloroplast-encoded proteins. 5 Our results suggest the existence of two mechanisms that operate on a larger scale to coordinate the expression of multiple complexes. First, the cell appears to leverage the chloroplast translation machinery to coregulate multiple complexes. Specifically, while translation factors MTF1 and CIF2 may look like classical housekeeping genes, our data suggest that they are leveraged for regulatory functions. Whereas classical housekeeping translation initiation factors mediate all translation, 78 MTF1 and CIF2 each affect specific subsets of chloroplast-expressed proteins, a property associated with regulatory factors. 79 CIF2 is mostly required for expression of photosynthetic machinery, whereas MTF1 loss also affects ribosomal large subunits ( Figures 6A and S7A – S7C ). Consistent with a regulatory role, MTF1 overexpression leads to overexpression of proteins downregulated in the mtf1 mutant ( Figure 6A ). The differences in the proteomic impacts of mtf1 and cif2 , combined with the differential regulation of the MTF1 and CIF2 transcripts ( Figure S7G ), suggest that MTF1 and CIF2 coordinate chloroplast gene expression in response to light and nitrogen availability. Second, our data suggest that the master regulator PMR1 regulates the mRNA levels of multiple nuclear-encoded ROGEs, thus coordinating the expression of the overall photosynthetic apparatus. We hypothesize that the higher-level regulatory mechanisms mediated by PMR1, MTF1, and CIF2 are essential for the cell’s rapid and coordinated response to changes in growth conditions. More than 65% of the 115 genes we identified as required for photosynthesis have homologs in land plants ( Figure S1H ). In most cases, the functions of these conserved genes appear to be similar in Chlamydomonas and land plants, supporting the value of Chlamydomonas as a model system and expanding the significance of our findings. Genes with no clear homologs in land plants could reflect homolog search failure due to sequence divergence 80 , 81 and/or different evolutionary innovations in the algal lineage such as the algal-specific CO 2 -concentrating mechanism (CCM), the study of which has the potential to enhance crop yields. 82 We anticipate that future studies of the genes identified here and explored in our proteomics dataset will enable further discoveries in photosynthesis. Limitations of the study Considering our FDR cutoff of 0.11, up to 11% of our hits may be false positives. We have validated by genetic rescue 12 of the 70 genes not previously known to be required for photosynthesis; future work on other genes will require independent validation. In addition, although protein localization by Venus-tagging is generally reliable, 30 , 31 increased confidence in the conclusions on cellular localization will require validation by an independent method such as immunofluorescence. 31 While we have initiated here the characterization of several of the identified genes, additional work is needed to fully characterize the molecular mechanisms by which they and other factors impact photosynthesis."
} | 2,615 |
38065083 | PMC10760936 | pmc | 8,174 | {
"abstract": "SUMMARY Photosynthesis is central to food production and the Earth’s biogeochemistry, yet the molecular basis for its regulation remains poorly understood. Here, using high-throughput genetics in the model eukaryotic alga Chlamydomonas reinhardtii , we identify with high confidence (false discovery rate [FDR] < 0.11) 70 poorly characterized genes required for photosynthesis. We then enable the functional characterization of these genes by providing a resource of proteomes of mutant strains, each lacking one of these genes. The data allow assignment of 34 genes to the biogenesis or regulation of one or more specific photosynthetic complexes. Further analysis uncovers biogenesis/regulatory roles for at least seven proteins, including five photosystem I mRNA maturation factors, the chloroplast translation factor MTF1, and the master regulator PMR1, which regulates chloroplast genes via nuclear-expressed factors. Our work provides a rich resource identifying regulatory and functional genes and placing them into pathways, thereby opening the door to a system-level understanding of photosynthesis.",
"introduction": "INTRODUCTION In photosynthetic eukaryotes, the photosynthetic apparatus consists of a series of protein complexes in the chloroplast thylakoid membrane that use light energy to produce NADPH, ATP, and other cellular energy carriers. 1 NADPH and ATP, in turn, power many pathways, notably CO 2 assimilation into sugar by the Calvin-Benson-Bassham metabolic cycle 2 ( Figure 1A ). 2 As a sophisticated system central to cellular fitness, hundreds of genes encoded in both the nucleus and chloroplast are required to assemble these complexes 3 and regulate their activity 3 under nuclear control. 4 In plants and green algae, this coordination is known to involve a range of different mechanisms, including post-transcriptional regulation of chloroplast-expressed genes by nuclear-encoded proteins, 5 translational regulation of chloroplast-expressed subunits by assembly intermediates of photosynthetic complexes, 6 and protease-mediated degradation of unassembled subunits. 7 Although photosynthesis and its regulation have been extensively studied for 70 years, 8 , 9 phylogenetics suggests that hundreds of genes participating in photosynthesis remain to be identified and characterized. Indeed, approximately half of the GreenCut2 genes—a set of 597 genes conserved only in the green photosynthetic eukaryotic lineage and therefore likely to be involved in photosynthesis 10 —have not been functionally characterized. Genetic screens have been done in land plants and algae to identify missing photosynthesis genes. Land plant screens have identified photosynthesis-deficient mutants based on leaf coloration, 11 , 12 seedling lethality, 13 and chlorophyll fluorescence. 14 , 15 As a complementary system to plants, the leading unicellular model eukaryotic alga Chlamydomonas reinhardtii ( Chlamydomonas ) has provided advantages of higher throughput and physiology that facilitate the identification and characterization of genes essential to photosynthesis. 16 , 17 These characteristics have been leveraged to identify and characterize many core components of the photosynthetic electron transport chain. 18 – 20 In the past decade, several hundred candidates for genes involved in photosynthesis have been uncovered by screens of two large Chlamydomonas mutant collections, Niyogi CAL 21 – 23 and CLiP. 24 , 25 However, these screens had many false positives and there are indications that fewer than half of these candidates are actually involved in photosynthesis. 25 Current challenges facing the field include (1) determining which of these candidates are genuinely involved in photosynthesis and (2) determining the functions of validated photosynthesis genes. Here, we address these two challenges by combining genetics and proteomics to identify and functionally characterize genes required for photosynthesis with high confidence on a global scale. We first identified with high confidence (false discovery rate [FDR] < 0.11) a total of 115 genes required for photosynthesis–including 70 genes whose molecular function in photosynthesis had not been previously characterized in any organism–by confirming linkage of each mutation with the observed photosynthetic defect and validating insertion site mappings. We then determined the proteomic profiles of mutants representing these genes to initiate their functional characterization, including assigning 34 of them to specific photosynthetic pathways. As proof of principle for the utility of our resource, we performed additional analyses, which revealed that five of these factors work with known factors to regulate mRNA maturation of key photosystem I (PSI) subunit PsaA. We also discovered and characterized two post-transcriptional regulators of photosynthetic apparatus biogenesis, providing insights into how cells leverage the chloroplast translation machinery and the regulation of nuclear gene expression to control photosynthetic complex abundance. Together, our dataset opens the door to rapid characterization of photosynthesis genes and provides systems-level insights into photosynthesis regulation.",
"discussion": "DISCUSSION In this study, we identified with high confidence (FDR < 0.11) 115 genes required for photosynthesis, including 70 whose functions in photosynthesis had not been previously characterized in any organism. We then showed that mutant proteomes provide key insights into the functions of these genes in photosynthesis, in many cases allowing the assignment of genes to specific pathways. We identified five ROGEs that are essential for the biogenesis of PSI. Including these genes, 76% (16/21) of genes with known functions in our dataset that lead to the depletion of an entire photosystem complex are ROGEs ( Figure 4 ), demonstrating their significant impact on photosynthesis. Growing evidence indicates that ROGEs play a regulatory role rather than being merely required for complex biogenesis 46 : different ROGEs affect different chloroplast-encoded genes, 5 are differentially transcriptionally regulated, 59 and participate in feedback loops, 6 , 76 a classical transcription network motif. 77 Moreover, several ROGEs can coregulate the same protein 56 , 76 ( Table S4 ), and the expression of photosystem proteins with a stronger effect on growth, including the largest subunit of each complex, tends to be impacted by more ROGEs ( Table S4 ). Our results further support a regulatory role for ROGEs by showing that different ROGEs can be limiting factors in different conditions: RAT2 is a limiting factor for psaA expression in the light but not in the dark ( Figures 4K and 5C – 5E ), and by discovering that multiple ROGEs are controlled by a master regulator ( Figure 6O ). Together, ROGE-mediated regulation raises the intriguing possibility that during the endosymbiosis process, as transcriptional regulation in the chloroplast was lost, 5 ROGEs evolved to replace transcription factors in a regulatory network for chloroplast-expressed proteins. In order to respond effectively to changing conditions, the cell must simultaneously regulate multiple photosynthetic complexes. Such coordinated regulation cannot be achieved by the ROGEs alone, since each regulates only one or two chloroplast-encoded proteins. 5 Our results suggest the existence of two mechanisms that operate on a larger scale to coordinate the expression of multiple complexes. First, the cell appears to leverage the chloroplast translation machinery to coregulate multiple complexes. Specifically, while translation factors MTF1 and CIF2 may look like classical housekeeping genes, our data suggest that they are leveraged for regulatory functions. Whereas classical housekeeping translation initiation factors mediate all translation, 78 MTF1 and CIF2 each affect specific subsets of chloroplast-expressed proteins, a property associated with regulatory factors. 79 CIF2 is mostly required for expression of photosynthetic machinery, whereas MTF1 loss also affects ribosomal large subunits ( Figures 6A and S7A – S7C ). Consistent with a regulatory role, MTF1 overexpression leads to overexpression of proteins downregulated in the mtf1 mutant ( Figure 6A ). The differences in the proteomic impacts of mtf1 and cif2 , combined with the differential regulation of the MTF1 and CIF2 transcripts ( Figure S7G ), suggest that MTF1 and CIF2 coordinate chloroplast gene expression in response to light and nitrogen availability. Second, our data suggest that the master regulator PMR1 regulates the mRNA levels of multiple nuclear-encoded ROGEs, thus coordinating the expression of the overall photosynthetic apparatus. We hypothesize that the higher-level regulatory mechanisms mediated by PMR1, MTF1, and CIF2 are essential for the cell’s rapid and coordinated response to changes in growth conditions. More than 65% of the 115 genes we identified as required for photosynthesis have homologs in land plants ( Figure S1H ). In most cases, the functions of these conserved genes appear to be similar in Chlamydomonas and land plants, supporting the value of Chlamydomonas as a model system and expanding the significance of our findings. Genes with no clear homologs in land plants could reflect homolog search failure due to sequence divergence 80 , 81 and/or different evolutionary innovations in the algal lineage such as the algal-specific CO 2 -concentrating mechanism (CCM), the study of which has the potential to enhance crop yields. 82 We anticipate that future studies of the genes identified here and explored in our proteomics dataset will enable further discoveries in photosynthesis. Limitations of the study Considering our FDR cutoff of 0.11, up to 11% of our hits may be false positives. We have validated by genetic rescue 12 of the 70 genes not previously known to be required for photosynthesis; future work on other genes will require independent validation. In addition, although protein localization by Venus-tagging is generally reliable, 30 , 31 increased confidence in the conclusions on cellular localization will require validation by an independent method such as immunofluorescence. 31 While we have initiated here the characterization of several of the identified genes, additional work is needed to fully characterize the molecular mechanisms by which they and other factors impact photosynthesis."
} | 2,615 |
38065083 | PMC10760936 | pmc | 8,174 | {
"abstract": "SUMMARY Photosynthesis is central to food production and the Earth’s biogeochemistry, yet the molecular basis for its regulation remains poorly understood. Here, using high-throughput genetics in the model eukaryotic alga Chlamydomonas reinhardtii , we identify with high confidence (false discovery rate [FDR] < 0.11) 70 poorly characterized genes required for photosynthesis. We then enable the functional characterization of these genes by providing a resource of proteomes of mutant strains, each lacking one of these genes. The data allow assignment of 34 genes to the biogenesis or regulation of one or more specific photosynthetic complexes. Further analysis uncovers biogenesis/regulatory roles for at least seven proteins, including five photosystem I mRNA maturation factors, the chloroplast translation factor MTF1, and the master regulator PMR1, which regulates chloroplast genes via nuclear-expressed factors. Our work provides a rich resource identifying regulatory and functional genes and placing them into pathways, thereby opening the door to a system-level understanding of photosynthesis.",
"introduction": "INTRODUCTION In photosynthetic eukaryotes, the photosynthetic apparatus consists of a series of protein complexes in the chloroplast thylakoid membrane that use light energy to produce NADPH, ATP, and other cellular energy carriers. 1 NADPH and ATP, in turn, power many pathways, notably CO 2 assimilation into sugar by the Calvin-Benson-Bassham metabolic cycle 2 ( Figure 1A ). 2 As a sophisticated system central to cellular fitness, hundreds of genes encoded in both the nucleus and chloroplast are required to assemble these complexes 3 and regulate their activity 3 under nuclear control. 4 In plants and green algae, this coordination is known to involve a range of different mechanisms, including post-transcriptional regulation of chloroplast-expressed genes by nuclear-encoded proteins, 5 translational regulation of chloroplast-expressed subunits by assembly intermediates of photosynthetic complexes, 6 and protease-mediated degradation of unassembled subunits. 7 Although photosynthesis and its regulation have been extensively studied for 70 years, 8 , 9 phylogenetics suggests that hundreds of genes participating in photosynthesis remain to be identified and characterized. Indeed, approximately half of the GreenCut2 genes—a set of 597 genes conserved only in the green photosynthetic eukaryotic lineage and therefore likely to be involved in photosynthesis 10 —have not been functionally characterized. Genetic screens have been done in land plants and algae to identify missing photosynthesis genes. Land plant screens have identified photosynthesis-deficient mutants based on leaf coloration, 11 , 12 seedling lethality, 13 and chlorophyll fluorescence. 14 , 15 As a complementary system to plants, the leading unicellular model eukaryotic alga Chlamydomonas reinhardtii ( Chlamydomonas ) has provided advantages of higher throughput and physiology that facilitate the identification and characterization of genes essential to photosynthesis. 16 , 17 These characteristics have been leveraged to identify and characterize many core components of the photosynthetic electron transport chain. 18 – 20 In the past decade, several hundred candidates for genes involved in photosynthesis have been uncovered by screens of two large Chlamydomonas mutant collections, Niyogi CAL 21 – 23 and CLiP. 24 , 25 However, these screens had many false positives and there are indications that fewer than half of these candidates are actually involved in photosynthesis. 25 Current challenges facing the field include (1) determining which of these candidates are genuinely involved in photosynthesis and (2) determining the functions of validated photosynthesis genes. Here, we address these two challenges by combining genetics and proteomics to identify and functionally characterize genes required for photosynthesis with high confidence on a global scale. We first identified with high confidence (false discovery rate [FDR] < 0.11) a total of 115 genes required for photosynthesis–including 70 genes whose molecular function in photosynthesis had not been previously characterized in any organism–by confirming linkage of each mutation with the observed photosynthetic defect and validating insertion site mappings. We then determined the proteomic profiles of mutants representing these genes to initiate their functional characterization, including assigning 34 of them to specific photosynthetic pathways. As proof of principle for the utility of our resource, we performed additional analyses, which revealed that five of these factors work with known factors to regulate mRNA maturation of key photosystem I (PSI) subunit PsaA. We also discovered and characterized two post-transcriptional regulators of photosynthetic apparatus biogenesis, providing insights into how cells leverage the chloroplast translation machinery and the regulation of nuclear gene expression to control photosynthetic complex abundance. Together, our dataset opens the door to rapid characterization of photosynthesis genes and provides systems-level insights into photosynthesis regulation.",
"discussion": "DISCUSSION In this study, we identified with high confidence (FDR < 0.11) 115 genes required for photosynthesis, including 70 whose functions in photosynthesis had not been previously characterized in any organism. We then showed that mutant proteomes provide key insights into the functions of these genes in photosynthesis, in many cases allowing the assignment of genes to specific pathways. We identified five ROGEs that are essential for the biogenesis of PSI. Including these genes, 76% (16/21) of genes with known functions in our dataset that lead to the depletion of an entire photosystem complex are ROGEs ( Figure 4 ), demonstrating their significant impact on photosynthesis. Growing evidence indicates that ROGEs play a regulatory role rather than being merely required for complex biogenesis 46 : different ROGEs affect different chloroplast-encoded genes, 5 are differentially transcriptionally regulated, 59 and participate in feedback loops, 6 , 76 a classical transcription network motif. 77 Moreover, several ROGEs can coregulate the same protein 56 , 76 ( Table S4 ), and the expression of photosystem proteins with a stronger effect on growth, including the largest subunit of each complex, tends to be impacted by more ROGEs ( Table S4 ). Our results further support a regulatory role for ROGEs by showing that different ROGEs can be limiting factors in different conditions: RAT2 is a limiting factor for psaA expression in the light but not in the dark ( Figures 4K and 5C – 5E ), and by discovering that multiple ROGEs are controlled by a master regulator ( Figure 6O ). Together, ROGE-mediated regulation raises the intriguing possibility that during the endosymbiosis process, as transcriptional regulation in the chloroplast was lost, 5 ROGEs evolved to replace transcription factors in a regulatory network for chloroplast-expressed proteins. In order to respond effectively to changing conditions, the cell must simultaneously regulate multiple photosynthetic complexes. Such coordinated regulation cannot be achieved by the ROGEs alone, since each regulates only one or two chloroplast-encoded proteins. 5 Our results suggest the existence of two mechanisms that operate on a larger scale to coordinate the expression of multiple complexes. First, the cell appears to leverage the chloroplast translation machinery to coregulate multiple complexes. Specifically, while translation factors MTF1 and CIF2 may look like classical housekeeping genes, our data suggest that they are leveraged for regulatory functions. Whereas classical housekeeping translation initiation factors mediate all translation, 78 MTF1 and CIF2 each affect specific subsets of chloroplast-expressed proteins, a property associated with regulatory factors. 79 CIF2 is mostly required for expression of photosynthetic machinery, whereas MTF1 loss also affects ribosomal large subunits ( Figures 6A and S7A – S7C ). Consistent with a regulatory role, MTF1 overexpression leads to overexpression of proteins downregulated in the mtf1 mutant ( Figure 6A ). The differences in the proteomic impacts of mtf1 and cif2 , combined with the differential regulation of the MTF1 and CIF2 transcripts ( Figure S7G ), suggest that MTF1 and CIF2 coordinate chloroplast gene expression in response to light and nitrogen availability. Second, our data suggest that the master regulator PMR1 regulates the mRNA levels of multiple nuclear-encoded ROGEs, thus coordinating the expression of the overall photosynthetic apparatus. We hypothesize that the higher-level regulatory mechanisms mediated by PMR1, MTF1, and CIF2 are essential for the cell’s rapid and coordinated response to changes in growth conditions. More than 65% of the 115 genes we identified as required for photosynthesis have homologs in land plants ( Figure S1H ). In most cases, the functions of these conserved genes appear to be similar in Chlamydomonas and land plants, supporting the value of Chlamydomonas as a model system and expanding the significance of our findings. Genes with no clear homologs in land plants could reflect homolog search failure due to sequence divergence 80 , 81 and/or different evolutionary innovations in the algal lineage such as the algal-specific CO 2 -concentrating mechanism (CCM), the study of which has the potential to enhance crop yields. 82 We anticipate that future studies of the genes identified here and explored in our proteomics dataset will enable further discoveries in photosynthesis. Limitations of the study Considering our FDR cutoff of 0.11, up to 11% of our hits may be false positives. We have validated by genetic rescue 12 of the 70 genes not previously known to be required for photosynthesis; future work on other genes will require independent validation. In addition, although protein localization by Venus-tagging is generally reliable, 30 , 31 increased confidence in the conclusions on cellular localization will require validation by an independent method such as immunofluorescence. 31 While we have initiated here the characterization of several of the identified genes, additional work is needed to fully characterize the molecular mechanisms by which they and other factors impact photosynthesis."
} | 2,615 |
38317929 | PMC10839987 | pmc | 8,176 | {
"abstract": "Understanding the impact of various parameters on the kinetics of dissolved selenium (Se) removal in bioreactors can be a challenging task, primarily due to the mass transfer limitations inherent in bioreactors employing attached growth configurations. This study successfully established a proof-of-concept for the efficient removal of Se from aqueous solutions using a chemostat bioreactor that relies solely on suspended growth. The research investigated the effect of selenate-Se feed concentrations under two distinct Se concentration conditions. One experiment was conducted at a considerably elevated concentration of 25 mg/L to impose stress on the system and evaluate its response. Another experiment replicated an environmentally relevant concentration of 1 mg/L, mirroring the typical Se concentrations in mine water. The bioreactor, featuring a working volume of 0.35 L, was operated as an anaerobic, fully mixed chemostat with hydraulic retention times (HRTs) ranging from 5 to 0.25 days. The outcomes revealed the chemostat's capacity to remove up to 25 mg/L of dissolved Se from water for all HRTs exceeding 1 day, under otherwise optimal conditions encompassing temperature, pH, and salinity. The research's significance lies in the development of a versatile tool designed to examine Se removal kinetics within a system devoid of mass transfer limitations. Furthermore, this study verified the ability of the bacterial consortium, obtained from a mine-influenced environment and enriched in the laboratory, to grow and sustain Se removal activities within a chemostat operating with HRTs as short as 1 day.",
"conclusion": "4 Conclusions This study investigated the removal of total dissolved Se with 25 and 1 mg/L feed Se-selenate in a chemostat. The two experiments were conducted with different Se, carbon source stoichiometric ratios, and inoculums (same source but grown in different batch bottles). The following observations were made. 1. Proof-of-concept was provided for the use of a chemostat bioreactor as a laboratory tool with suspended biomass and well-mixed conditions in order to study intrinsic biological removal of Se-selenate in the absence of mass transfer limitations. 2. In two experiments with feed Se-selenate concentrations of 25 and 1 mg-Se/L, Se was removed to 30 μg/L and 1 μg/L, respectively. The removal percentage was 99.8 % at 25 mg/L and 99.9 % at 1 mg/L Se-selenate. This is a significant contribution because the total dissolved Se concentration in the feed of industrial bioreactors is highly variable. 3. The yield of carbon source consumed per amount of dissolved Se removed, Y C Se , was investigated for Experiment 2. The experimental yield was 4–7 times higher than the theoretical stoichiometric yield for all HRTs. This is a significant contribution for operation of industrial bioreactors as it provides information on the amount of carbon source that is required to achieve and maintain the removal of Se in bioreactors. Limiting amounts of carbon source can be inhibitory for microbial activity and too much carbon source can change the environment toward extremely negative oxidation-reduction potentials that result in permanently shifting microbial community toward sulfate reducing bacteria. 4. The bacteria that were collected from a mine-influenced environment and enriched in the laboratory were capable of growing and maintaining Se removal activities in a chemostat with HRTs as short as 1 day. 5. The study of microbial community compositions revealed that the composition of the community changes as a function of operating conditions (HRT and carbon source concentration) and over time. Some of the families such as Rhodocyclaceae and Comamonadaceae families were abundant in both experiments; otherwise, the families of microorganisms were different in the two experiments. A fraction of the bacteria was unclassified in the genus level for both experiments since the inoculum was started from an environmental source that possibly contained bacteria that had not been sequenced before. 6. The dominant microorganisms changed cyclically which is possibly due to the interactions between different families in the community; meaning that different populations can have facilitative or inhibitory effects on the growth of other populations. [ 39 ].",
"introduction": "1 Introduction Selenium (Se) exists naturally in the Earth's crust at concentrations between 0.05 and 0.5 mg Se kg −1 [ 1 ]. Anthropogenic activities that mobilize Se compounds into the environment include metal and coal mining. This may result in excessive concentrations of Se in receiving aquatic environments with total dissolved Se concentrations up to 12 mg-Se/L measured in some locations [ 2 ]. Regulated concentrations of total dissolved Se for discharge of contaminated industrial wastewater into receiving aquatic environments varies by region. The present selenium water quality guideline for the protection of freshwater and marine aquatic life set by the Canadian Council of Ministers of Environment is 1 μg/L, the British Columbia Ministry of Environment guideline is 2 μg/L, and the USA Environmental Protection Agency and Australia/New Zealand guidelines are 5 μg/L [ 3 ]. The reason that these guidelines are at the microgram level is because Se bioaccumulates up the food chain in the receiving aquatic environment, which results in toxic effects on wildlife such as deformity and reproductive failure in fish (Ohlendorf et al., 2011). Consequently, dissolved forms of Se, selenate ( Se O 4 2 − , Se [ VI ] ) and selenite ( Se O 3 2 − , Se [ IV ] ), must be removed from mine-influenced water (MIW) to meet the regulated concentrations. Biological treatment of selenium from industrial wastewater is currently identified as the preferred alternative to chemical and physical technologies due to the lower reagent requirements, more compact sludge, and possibility of treating high volumes of wastewater [ [4] , [5] , [6] ]. In biological treatment of Se, microorganisms gain energy from oxidation-reduction reactions between an electron donor such as acetate ( C H 3 COOH ) and an electron acceptor such as selenate or selenite in a controlled environment within the bioreactor [ 5 ]. Over the last few decades, active bioreactors in which nutrients and energy are added for better performance, have gained attention for Se bioremediation. The bioreactor is the core of the active biological treatment systems which is designed to facilitate the microbial growth and biological treatment of the target contaminant [ 7 ]. The performance of active bioreactors in dissolved selenium (Se) removal depends on the bioreactor configuration. Active bioreactors are configured as suspended-growth, attached-growth or hybrid (contains both attached growth and suspended growth) systems. The attached-growth bioreactors have been widely used as the well-established technology for metal-laden water treatment processes in the industry. In attached-growth systems the microorganisms form a biofilm on the surface of a solid media known as biomatrix [ 8 ] by forming extracellular polymeric substances (EPS). The formation of biofilm reduces the risk of the biomass being washed out from the system. The removal of dissolved Se was examined in several studies in batch or continuous reactors including packed bed or fluidized bed reactors. Two commercially available attached-growth bioreactor configurations specifically developed for removal of selenate from Se-laden industrial wastewater include fluidized bed reactors (FBR), such as the one produced by the company Envirogen, and packed-bed reactors, such as the ABMet produced by Suez [ 6 ]. Envirogen bioreactor has been implemented in pilot and full scale for removal of up to 0.55 ppm Se down to 4.7 μg/L (99 % removal efficiency) [ 9 , 10 ]. The ABMet technology has also been evaluated for its capability for selenate and nitrate removal from MIW and have shown to be effective in removal of up to 1.95 ppm Se from industrial wastewater to below 2 μg/L in pilot and full scale [ 11 , 12 ]; MSE Technology Applications Inc., 2001). Depending on the reactor configuration and influent water chemistry (concentration of contaminants to be removed and water matrix), the operational hydraulic retention time (HRT) of industrial bioreactors can be in the range of hours to days. In all these attached-growth systems, mass transfer limitations exist. To the best of authors’ knowledge only one other bench-scale study investigated the removal of dissolved Se in a continuous reactor with only suspended biomass in the reactor [ 13 ]. There has been no full-scale proof of concept for Se treatment plants running solely as suspended-growth system [ 7 ]; due to the risk of washout and difficulty of maintaining anaerobic conditions at full-scale. Chemostats are typically used for kinetic studies, which must be performed in the absence of mass transfer limitations. In this work, the objective was to provide a proof-of-concept for the removal of total dissolved Se in a chemostat bioreactor containing only suspended biomass as the first step toward developing a laboratory tool for studying kinetics of dissolved Se removal in bioreactors. The chemostat bioreactor was used to assess the effect of HRT on extent and rate of Se removal. It could be challenging to achieve the stringent regulated limits for total dissolved Se concentration using bioreactors relying solely on attached growth. Since the total dissolved Se concentration in mine-influenced water (MIW) is widely variable due to seasonal changes in the range of 0.002–12 mg/L measured in some receiving environments [ 2 ], operators of industrial bioreactors need to adjust the concentration of carbon source, that is provided as electron donor for bacterial growth and energy production, proportional to the influent Se concentrations as electron acceptor. In this work, it was hypothesized that the dissolved Se removal in a bioreactor, operating anaerobically at constant optimal temperature and pH where carbon source is supplied in excess, and in the absence of inhibitory compounds, is influenced by the concentration of selenium as the electron acceptor. The removal of dissolved Se was determined as a function of total dissolved Se concentration using Se in the form of selenate and excess stoichiometric amounts of carbon source acetate. These variabilities in electron acceptor and electron donor type and concentration can change the Se transformation mechanisms and consequently the microbial community composition inside the bioreactor [ 14 ]. The microorganisms used in this study were enriched from sub-aqueous sediments collected from seepage ponds receiving MIW high in dissolved Se, as opposed to a pure culture. Environmentally sourced bacteria were chosen since mixed culture consortia are more robust and resistant to environmental changes compared to pure cultures [ 15 ]. This study represents a proof-of-concept for the efficient removal of selenium at concentrations up to 25 mg/L from synthetic wastewater, employing a chemostat bioreactor that exclusively relies on suspended growth system. The significance of this research lies in the development of a versatile tool designed for examination of Se removal kinetics in a system without mass transfer limitations. This chemostat reactor configuration can be utilized for a broad spectrum of kinetic investigations. This includes, but is not limited to, the effect of diverse parameters such as Se concentration, carbon source concentration, or operational conditions (e.g., temperature, pH) on the Se removal process. By conducting such comprehensive kinetic studies using this innovative approach, there is great potential to assist mining companies in the optimization of their Se removal bioreactors. Ultimately, this advancement contributes to more effective and environmentally sustainable practices within the mining industry.",
"discussion": "3 Results and discussion 3.1 Se removal Dissolved Se concentrations in the chemostat effluent measured over time are plotted in Fig. 2 , Fig. 3 for Experiments 1 and 2, respectively. A logarithmic scale is used for dissolved Se concentration to observe more clearly the lowest effluent concentrations achieved, which is of interest given the strict regulatory requirements for the discharge of Se into aquatic environments. In Experiment 1 during the initial batch period, the dissolved Se concentration decreased from 25 to 10 mg Se/L within 5 days indicating an active selenate reducing microbial community. Upon the earliest appearance of orange colour on Day 5, the reactor was switched to continuous mode. Following the initiation of continuous flow, the chemostat was operated for 29 days at an HRT of 5 days. During this time, the effluent dissolved Se concentration achieved was 1.14 ± 0.21 mg Se/L. Subsequently, the HRT was reduced to 3 days for 15 days, during which the effluent dissolved Se concentration decreased over time reaching a steady-state concentration of 0.12 mg Se/L. The lowest effluent dissolved Se concentration achieved was 30 μg/L when the HRT was 2 days, which represents 99.9 % Se removal from the dissolved phase. When the HRT was shortened to 1 day and then 0.5 days, the effluent dissolved Se concentrations increased. Finally, when the HRT was further reduced to 0.25 days, the dissolved Se concentration increased to 27.5 mg/L, which is close to the concentration in the influent, meaning that dissolved Se was no longer being removed in the chemostat. This indicates that a minimum HRT was reached at which washout occurred. Washout happens when the concentration of microorganisms inside the bioreactor decreases to zero since the dilution rate (D), which equals the volumetric flow rate divided by the working volume of the bioreactor, and the inverse of the HRT, is greater than the cell specific growth rate (μ). Thus, for the conditions of this experiment, D becomes greater than μ for the selenate reducing bacteria in the chemostat somewhere between 2 and 4 day −1 . Fig. 2 Expected influent (dashed line) and measured effluent (circles) dissolved Se concentration over time in the chemostat with a feed concentration of 25 mg/L selenate-Se and 3 times stoichiometric amounts of carbon source acetate. Effluent concentrations are coloured according to HRT. Fig. 2 Fig. 3 Influent (triangles) and effluent (circles) dissolved Se concentration over time in chemostat with 1 mg/L feed selenate-Se and 10 times stoichiometric acetate. Effluent concentrations are coloured according to HRT. Fig. 3 In Experiment 2, the chemostat was switched from batch to continuous mode after 1 day due to the appearance of orange-coloured precipitates assumed to be elemental Se. The analytical data confirmed that Se was removed during the 1-day batch period from 0.72 to 0.61 mg Se/L. As opposed to Experiment 1, for this experiment a sampling port was added to the influent stream so that feed concentrations could be analyzed to confirm that they were as expected ( Fig. 3 ). Measurements indicated that the feed dissolved Se concentration was 0.98 ± 0.10 mg Se/L for the entire duration of the experiment. As was observed in Experiment 1, the effluent dissolved Se concentrations decreased over time as the HRT was decreased from 3 to 2.5 and then to 2 days. The lowest dissolved Se concentration observed in the chemostat effluent was 1 μg/L (99.9 % Se removal) when the chemostat was running at HRTs of 2 days and 1.5 days. After the HRT was reduced to 1 day, the effluent dissolved Se concentration increased to 0.1 mg/L. Finally, when the HRT was further shortened to 0.5 days, the effluent dissolved Se concentration reached the same concentration as that in the feed signaling that washout had occurred. Thus, under the conditions for Experiment 2, the dilution rate, D, exceeded the cell specific growth rate somewhere between 1 and 2 day −1 . Plotting steady-state effluent dissolved Se concentrations, obtained by averaging the final two or three data points in each HRT run, as a function of HRT ( Fig. 4 ) highlights similarities and differences between the two experiments. In both experiments, a similar pattern was observed where the steady-state effluent dissolved Se concentrations decreased as the HRT was reduced from 5 or 3, respectively, reaching a minimum at an HRT of 2 days. The second experiment, using 1 mg selenate-Se/L in the feed appears to achieve lower effluent concentrations that are close to regulatory requirements for some locations (1-2 μg-Se/L BCWQG) [ 3 ]. It was possible to operate the chemostat at an HRT of 0.5 days when the feed selenate-Se concentration was 25 mg/L, whereas washout occurred at this HRT when the lower feed concentration was used. Washout is expected to occur at lower HRTs for higher feed concentrations of substrate in a chemostat with microbial growth following Monod kinetics ( Fig. 3 .3, page 174 in Ref. [ 20 ]). However, in Experiment 1, the chemostat was run at an HRT of 0.5 days for only three days during which the effluent dissolved Se concentration was continuously increasing and steady-state may not have been achieved, thus there is some uncertainty as to if the chemostat could operate for longer than 3 days at this HRT. Fig. 4 The average steady-state dissolved Se concentrations at each HRT are plotted versus HRT with 25 mg/L Se-selenate and 3-times stoichiometric C-source (red circle) and 1 mg/L Se-selenate and 10-times stoichiometric C-source (blue circle) in the feed. The error bars show the standard deviation for the averaged dissolved Se concentrations at each HRT. (For interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article.) Fig. 4 By changing the HRTs from 5 to 2 days (for Experiment 1) and from 3 to 2 days (for Experiment 2), despite the fact that the HRTs were reduced, improving Se removal extent was observed. This trend was unexpected since longer retention times should lead to higher conversions. This could be potentially explained by the fact that longer HRTs (5 days for Experiment 1 and 3 days for Experiment 2) are leading to lower amounts of active biomass in the chemostat due to biomass decay, which might contribute to lower Se removal during long HRTs [ 20 ]. It is also possible that dead biomass releases elemental Se back into the solution while still in the chemostat. One of the study's limitations was the inability to perform a comprehensive statistical analysis, primarily due to the limited number of data points available at each concentration. Attempting statistical analysis with such a small sample size heightens the risk of overfitting and may lead to inaccurate estimates or interpretations. 3.2 Carbon source consumption The carbon source consumption was monitored for Experiment 1 by measuring SCOD versus time ( Fig. 5 ). The concentration of SCOD in the chemostat at the beginning of the batch phase was expected to be around 253 mg/L based on the mass of sodium acetate added to the growth medium. However, the measured SCOD concentration was over 600 mg/L. The additional SCOD was likely associated with the inoculum. The inoculum for this experiment was prepared using fresh sediments removed from the mine site location two weeks prior to beginning the enrichment process. Organic compounds in the sediments might have persisted in the enrichment culture even though the enrichment period was 30 days. With the commencement of continuous flow, the effluent SCOD concentration remained much greater than the expected feed concentration of 76 mg/L due to this additional SCOD from the inoculum. The SCOD effluent concentration declined continuously until Day 53, thereafter remaining steady at 77 ± 6 mg/L until Day 68. It is unknown what chemical compounds contributed to the excess SCOD nor to what degree they were consumed by the microorganisms in the bioreactor. It is possible that they were slowly washed out from the bioreactor without reacting. From Day 68 until the end of the experiment, the effluent SCOD concentrations dropped to levels below the expected influent concentration and decreased steadily to 6 mg/L. The effluent SCOD concentrations did not trend upwards towards the end of the experiment similar to what was observed for effluent dissolved Se concentrations. This might have been due to the activity of other heterotrophic organisms that continued growing in the system whose growth was not coupled to selenate or selenite reduction, such as fermentative organisms, for example. Fig. 5 Expected influent (dashed line) and measured effluent (circles) chemical oxygen demand (SCOD) versus time for an influent concentration of 25 mg Se-selenate/L and 3-times stoichiometric amounts of acetate in the feed. A selection of SCOD samples was analyzed in replicates and a standard deviation of less than 2 % was measured. Fig. 5 To test if the carbon source supply was sufficient, in addition to the SCOD of the effluent, the amount of SCOD in the influent was also monitored for Experiment 2. In Experiment 2, the influent SCOD concentration was expected to be 10.9 mg/L based on the amount of acetate provided and the flow rate of feed to the chemostat. Measured influent SCOD concentrations ( Fig. 6 ) verified this (10.9 ± 0.5 mg/L) and acetate was confirmed to be the sole carbon source entering the chemostat. During the first HRT run tested during continuous flow (3 days), the effluent concentration of SCOD decreased consistently from 9.0 to 4.6 mg/L over 7 days. On Day 7, the concentration of SCOD decreased due to a clog in the SCOD influent tubing downstream of the influent sample port which was resolved. The effluent SCOD concentration was less variable during the periods when the HRT was 2.5 and 2 days, with an average of 5.39 ± 0.7 mg/L. When the HRT was changed to 1.5 days, the SCOD concentration started to trend upwards and by the time the HRT was reduced to 1 day, SCOD concentration increased to 11.8 mg/L meaning that SCOD not being consumed in the chemostat. This trend was consistent with that of dissolved Se in the effluent. The observation that SCOD was not being consumed confirms that washout of microorganisms occurred at the end of the HRT of 1 day run and that this was likely the cause of no removal of dissolved Se at that time. Fig. 6 Influent (triangles) and effluent (circles) SCOD concentrations over time in the chemostat with feed concentrations of 1 mg/L Se-selenate and 10-times stoichiometric amounts of acetate. Effluent concentrations are coloured according to HRT. A selection of SCOD samples were analyzed in replicates and a standard deviation of less than 2 % was measured. Fig. 6 Data from Experiment 2 were used to estimate the yield of carbon source consumed per amount of dissolved Se removed, Y C Se , by dividing the moles of acetate used, calculated from Equation 3 (3) ( mg COD i n − mg COD o u t ) mmole COD 32 mg COD mmole acetate 2 mmole COD by the moles of dissolved Se removed, calculated from Equation 4 (4) ( mg Se i n − mg Se o u t ) mmole Se 78.9 mg Se for each data point. The yield coefficients were averaged for each HRT ( Fig. 7 ) and compared to the theoretical yield coefficient of 1.25 mol acetate/mole Se (dashed line in Fig. 7 ). The yield was most variable during an HRT of 1.5 days when the highest removal extents of dissolved Se were achieved. Calculated average yields Y C Se ranged from 4.8 to 8.5 mol COD/mole Se, which were 4–7 times higher than the theoretical yield. This suggests that more than ∼74–85 % of acetate is consumed by microorganisms whose growth is not coupled with reduction of Se compounds, or the assumed f S used to derive the stoichiometric equations is not valid for this microbial community. Fig. 7 Average yield coefficient for moles of acetate used per mole of Se removed at each HRT in presence of 10-times stoichiometric carbon source. The dashed line represents the theoretical yield coefficient based on balanced stoichiometric reactions. The error bars show the standard deviation for the averaged yields at each HRT. Fig. 7 3.3 Microbial population diversity In the first experiment, an average of 23,173 high quality 16S rRNA sequences per sample were obtained from samples removed during HRTs of 3, 2, 1, 0.5 and 0.25 days. In the second experiment, an average of 13,715 high quality reads per sample were obtained from samples removed during the batch period and HRTs of 3, 2.5, 2 and 1.5 days. These were used to assess the diversity of the chemostat microbial population over time ( Fig. 8 A and B). Overall, the average number of observed features in the samples from Experiment 1 was 115 ( ± 23) with the only observable trend being a much higher number of features on Day 67 when the HRT was 1 day. Overall, the number of observed features in samples from Experiment 2 were fewer, decreasing from 104 features during the batch phase to an average of 71 (±12) over the following 24 days indicating a trend towards a more constrained population. Fig. 8 Diversity indices versus time A) for Experiment 1 and B) for Experiment 2. Fig. 8 Faith phylogenetic diversity which assesses the diversity with respect to the genetic relatedness of the population, reached a maximum of 15.7 on Day 67 and reduced to an average of 10 ( ± 1.5) for Experiment 1. For Experiment 2, the Faith index was maximum on Day 1 at 10.6, then decreased to 4.9 on Day 9 and increased to a final value of 10 on Day 24 which was similar to Day 1. The Shannon index which is an indication of species richness and evenness in the community, was increasing and decreasing over the course of the experiments ranging between 2.8 and 5.5. 3.4 Microbial population composition In Experiment 1, the first sample for DNA analysis was collected on Day 49. Between Days 49–60 in Fig. 9 A when the HRT was 3 and 2 days, members of the Moraxellaceae , Rhodocyclaceae, Comamonadaceae, and Clostridiaceae families were dominant. Some Moraxellaceae were classified in the genus Acinetobacter . Members of the Acinetobacter genus are commonly present in soil, water, and sewage environments [ [28] , [29] , [30] ]. Some species such as Acinetobacter sp. VS3 are known for their ability to grow in selenate and selenite-rich environments [ 28 ]. After Day 64 and towards the end of the experiment, when the HRT was less than 1 day, Dechloromonas genus as well as some unclassified genera from the Rhodocyclaceae family were dominant. Multiple studies have reported the role of Dechloromonas species in the bioremediation of selenium oxyanions in the presence of acetate as electron donor [ 31 , 32 ]. The selenate reduction activity of these microorganisms is probably due to the similarity of its perchlorate reductase enzyme (pcrABCD) which is one nucleotide different from a selenate reductase (SerABC) [ 33 ]. In addition, metagenome-assembled genomes obtained from whole genome sequencing of samples collected from similar mine-impacted sediments in a previous study classified within Rhodocyclaceae family contained sequences for putative selenate reductases that were homologous with the NarG enzyme [ 34 ]. The NarG enzyme preferentially reduces nitrate, but it also has an affinity for selenate when nitrate is absent [ 14 ]. Fig. 9 Dominant ASVs (>2 % of the total bacterial population) families in the chemostat over time A) for Experiment 1 and B) for Experiment 2. Fig. 9 The Comamonadaceae and Clostridiaceae families were only dominant intermittently on (and potentially before) Day 49 and were not among the dominant families for the rest of the experiment. Family Comamonadaceae is in the beta-proteobacteria class and metagenome-assembled genomes for members of this family contained sequences for putative selenate reductases [ 34 ]. Some clades within the Clostridiaceae family are selenium dependent microorganisms that require selenium for the synthesis of selenoproteins that are needed for oxidoreductase pathways [ 35 ]. Clostridium sp. BXM from this family is capable of selenate and selenite reduction [ 36 ]. Some examples of selenium-dependent enzymes (selenoenzymes) from Clostridiaceae are glycine reductase protein A and B, proline reductase, purine hydroxylase, and glutaredoxin [ [40] , [41] ]. Dominance by organisms within this same clade ( Clostridium sensu stricto 13) was observed also by Ref. [ 16 ] in their study in which the same source of sediment inoculum was used and was demonstrated to remove dissolved Se-selenate in the presence of nitrate in batch cultures [ 16 ]. At the beginning of Experiment 2, the only dominant family was Carnobacteriaceae ( Fig. 9 B). This family is within the order Lactobacillales and the Firmicutes phylum that are commonly found in Se-contaminated mine sites [ 37 , 38 ]. After two days, the Clostridiaceae family became abundant in the community that was also dominant in Experiment 1. Between days 7 and 14, Rhodocyclaceae and Comamonadaceae families were abundant in the chemostat which were also abundant in Experiment 1. This study was one of the few studies that were conducted in a continuous bioreactor by solely relying on the suspended growth of microorganisms in the bioreactor. In this study, the effect of influent Se concentration was investigated on its removal extent using a consortium of organisms that was collected from a mine-influenced environment and enriched in the laboratory instead of a pure culture. The use of a bacterial consortium that contains a variety of bacteria that are capable of selenate and/or selenite reduction is advantageous to pure cultures due to resistance of the microbial community to changes in the influent water chemistry. To the best of authors' knowledge, one kinetic study was conducted in a chemostat using a pure culture of Bacillus Sp. SF- 1 to study the kinetics of selenate and selenite removal [ 13 ]. The use of chemostat bioreactor is especially recommended for kinetic studies because it makes it possible to study the intrinsic kinetics of total dissolved Se removal when it can be assumed that there is no mass transfer limitation. While there are advantages to using a bacterial consortium, it does introduce certain limitations, particularly in the context of microbial community analysis. Exploring microbial composition in diverse environmental samples presents significant challenges. These samples often contain an immense variety of microbial species, some of which may be present in low abundance or have specific growth requirements, making detection and characterization difficult. DNA extraction and sequencing of such samples generates massive datasets which require complicated bioinformatics analysis. Sample variability and the presence of unknown microorganisms, further contribute to the complexity. Despite ongoing technical advancements, the exploration of microbial diversity in environmental samples remains a scientific challenge. To address these challenges, DNA studies in this research were restricted to samples obtained from the chemostat, as they were anticipated to have a relatively limited microbial diversity. Taxonomic characterization of the enriched microbial consortia from the same mine-influenced site is reported in Ref. [ 16 ]."
} | 7,838 |
38183463 | PMC10998807 | pmc | 8,177 | {
"abstract": "Ectomycorrhizal (EM) associations can promote the dominance of tree species in otherwise diverse tropical forests. These EM associations between trees and their fungal mutualists have important consequences for soil organic matter cycling, yet the influence of these EM-associated effects on surrounding microbial communities is not well known, particularly in neotropical forests. We examined fungal and prokaryotic community composition in surface soil samples from mixed arbuscular mycorrhizal (AM) and ectomycorrhizal (EM) stands as well as stands dominated by EM-associated Oreomunnea mexicana (Juglandaceae) in four watersheds differing in soil fertility in the Fortuna Forest Reserve, Panama. We hypothesized that EM-dominated stands would support distinct microbial community assemblages relative to the mixed AM-EM stands due to differences in carbon and nitrogen cycling associated with the dominance of EM trees. We expected that this microbiome selection in EM-dominated stands would lead to lower overall microbial community diversity and turnover, with tighter correspondence between general fungal and prokaryotic communities. We measured fungal and prokaryotic community composition via high-throughput Illumina sequencing of the ITS2 (fungi) and 16S rRNA (prokaryotic) gene regions. We analyzed differences in alpha and beta diversity between forest stands associated with different mycorrhizal types, as well as the relative abundance of fungal functional groups and various microbial taxa. We found that fungal and prokaryotic community composition differed based on stand mycorrhizal type. There was lower prokaryotic diversity and lower relative abundance of fungal saprotrophs and pathogens in EM-dominated than AM-EM mixed stands. However, contrary to our prediction, there was lower homogeneity for fungal communities in EM-dominated stands compared to mixed AM-EM stands. Overall, we demonstrate that EM-dominated tropical forest stands have distinct soil microbiomes relative to surrounding diverse forests, suggesting that EM fungi may filter microbial functional groups in ways that could potentially influence plant performance or ecosystem function. Supplementary Information The online version contains supplementary material available at 10.1007/s00572-023-01134-4.",
"conclusion": "Conclusion In a tropical montane forest, we demonstrate that EM-dominated stands are characterized by decreased prokaryotic diversity and different relative abundances of several important microbial functional groups compared to surrounding diverse mixed AM-EM forest. Differences in microbial communities between stand mycorrhizal types could be driven by overall differences in plant communities or microbiome assembly but likely contribute to broad EM-associated ecosystem effects. The relationships among different constituents of the soil microbiome could be an important extension of mycorrhizal function in response to the surrounding environment. Distinct soil microbiomes in EM-dominated versus mixed AM-EM stands may contribute to the effects of EM relationships on plant health or ecosystem function via changes to the relative abundance of fungal pathogens and saprotrophs. Overall, we found that the effects of mycorrhizal associations extend beyond the plant-fungal partnership into the broader soil microbiome and are especially pronounced for soil bacterial/archaeal communities.",
"introduction": "Introduction Plants often influence surrounding soils via interactions with microbial organisms (Zak et al. 2003 ). Predominant among these plant–microbe interactions are mycorrhizal associations between fungi and plant roots (Hawkes et al. 2007 ). Two main types of mycorrhizal association, namely, arbuscular mycorrhizal (AM) and ectomycorrhizal (EM), can impact plant health (Revillini et al. 2016 ), litter decomposition (Jacobs et al. 2018 ), nutrient cycling (Phillips et al. 2013 ), and soil organic matter dynamics (Frey 2019 ). Importantly, mycorrhizal fungi facilitate plant acquisition of soil nutrients such as nitrogen and phosphorous to support plant growth and nutrition (Smith and Smith 2011 ). However, bacteria, archaea, and non-mycorrhizal fungi also contribute to mycorrhizal plant and ecosystem effects as they can influence nutrient transformation processes, promote plant root exudation, and benefit plant growth (Tarkka et al. 2018 ; Sangwan and Prasanna 2022 ; Berrios et al. 2023 ). These non-mycorrhizal microbial communities are likely responsible for the extended plant-soil effects of mycorrhizal associations that can influence overall forest population dynamics in temperate ecosystems (Bennett et al. 2017 ). However, we know very little about how these “mycorrhizosphere” (Rambelli 1973 ) microbiomes manifest in tropical ecosystems that have differing patterns of productivity, species diversity, and soil functionality (Barlow et al. 2018 ). Neotropical forests are generally a matrix of primarily AM-associating tree species interspersed with fewer EM-associated species (McGuire et al. 2012 ), but EM-associated species can also form stands where the majority of the total basal area comprises a single species (Hart et al. 1989 ; Peh et al. 2011 ). These EM-dominated stands often differ significantly from surrounding mixed-mycorrhizal forest stands in important ecosystem characteristics (Torti et al. 2001 ), with slower decomposition rates (McGuire et al. 2010 ) and lower soil inorganic nutrient availability (Corrales et al. 2016b ) than adjacent mixed AM-EM forest stands. Ectomycorrhizal-dominated stands in these forests are also associated with changes to the relative abundance of some fungal functional guilds (Seyfried et al. 2022 ), but the impact of EM dominance on the broader complex soil microbiome, particularly prokaryotic communities, is still unclear due to a relative lack of data from neotropical forests. Mycorrhizal fungi often interact with surrounding microbial communities in ways that create favorable ecological conditions for their plant partners (Uroz et al. 2019 ). These interactions may play an important role in the formation or proliferation of EM-dominated forest stands. For example, EM fungi are thought to outcompete saprotrophic organisms for nutrients in soil organic matter (SOM; Averill and Hawkes 2016 ) due to their ability to produce SOM-degrading extracellular enzymes (Pellitier and Zak 2018 ). This fungal interguild competition can increase soil carbon (C)-to-nutrient ratios, slowing C and inorganic nutrient cycling (Fernandez and Kennedy 2016 ). These changes could decrease overall microbial C and nutrient availability, potentially resulting in lower overall microbial diversity with increasing EM dominance (Eagar et al. 2021 ; Heděnec et al. 2023 ), as well as downstream impacts on copiotrophic or oligotrophic soil microbiota (Nemergut et al. 2010 ). Ectomycorrhizal-associated changes to non-mycorrhizal microbial communities associated with soil organic matter cycling could potentially further promote positive feedbacks to slow SOM cycling and create a competitive advantage for EM mutualisms. Arbuscular mycorrhizal fungi are often thought to scavenge inorganic nutrients from soil and must rely on surrounding microbial communities to degrade SOM (van Der Heijden et al. 2015 ). Mycorrhizal fungi also influence the activity and abundance of soil fungal pathogens (Borowicz 2001 ; Veresoglou and Rillig 2012 ). Both AM and EM fungi can confer pathogen resistance to their plant hosts through the release of volatile organic compounds (Dreischhoff et al. 2020 ) or extracellular secretion of secondary metabolites (Pellegrin et al. 2015 ). However, EM relationships can generate greater conspecific benefits for pathogen suppression than AM, potentially promoting EM dominance (Liang et al. 2020 ) and resulting in greater conmycorrhizal plant recruitment (Delavaux et al. 2023 ). Overall, the effects of mycorrhizal associations on surrounding microbial community function are highly context-dependent, with their outcome varying based on mycorrhizal fungal species (Emmett et al. 2021 ), plant species and litter quality (Fernandez et al. 2019 ), climate (Bennett and Classen 2020 ), and soil parent material (Seyfried et al. 2021b ). Quantifying the effects of EM dominance on the different constituents of the soil microbiome could provide valuable insight to help contextualize the effects of mycorrhizal associations on surrounding soil microbiomes. To characterize soil microbiome responses to EM tree dominance in a neotropical forest, we conducted a study comparing bulk soil fungal and prokaryotic communities between EM-dominated (by Oreomunnea mexicana at > 50% basal area per stand) and mixed AM-EM stands of highly diverse lower montane tropical forests in western Panama. While these sites also contain ~ 14 other EM tree species, those species occur in low abundance, and O . mexicana is the only one to form monodominant stands (Prada et al. 2017 ). Variation in parent material and geology among these stands leads to the formation of soils differing in nutrient availability, pH, and base saturation (Prada et al. 2017 ; Seyfried et al. 2021a ), allowing for the investigation of microbial relationships with stand mycorrhizal type across a range of soil fertilities. We hypothesized that EM-dominated stands would be associated with distinct fungal and prokaryotic community assemblages relative to the AM-EM mixed stands. Given that EM dominance can slow SOM cycling and reduce nutrient availability, we present three predictions related to microbiome differences between stand mycorrhizal types (1) that EM stands would be associated with decreased microbial community diversity and heterogeneity compared to mixed AM-EM stands; (2) that EM stands would be associated with decreased relative abundance of saprotrophic and pathogenic fungal functional guilds, as well as bacteria and archaea associated with SOM cycling, compared to AM-EM mixed stands; and (3) that microbial communities in EM-dominated stands would have tighter Procrustean correspondence between general fungal and prokaryotic communities, as EM fungi could be acting to filter the soil microbiome to select microbial organisms that may contribute to an EM competitive advantage.",
"discussion": "Discussion We found a strong relationship between EM dominance and bulk soil microbiome composition in diverse neotropical montane forests. The relative abundance of EM-associating tree species in tropical forests is associated with significant shifts in soil properties (Barceló et al. 2022 ), yet it remains unclear how these shifts influence soil microbial composition. In support of our hypothesis, we found clear differentiation in soil microbial communities between stand mycorrhizal types across a range of forests differing in parent material and soil chemical properties, with stand mycorrhizal type explaining slightly greater variation in prokaryotic communities than site location. In relation to our predictions, (1) we demonstrate that EM-dominated forest stands maintain a less diverse prokaryotic microbiome than mixed AM-EM stands, (2) EM-dominated stands have lower relative abundance of fungal saprotrophs and pathogens than surrounding AM-EM mixed stands with taxonomic shifts of prokaryotes aligning with expected functional shifts for SOM cycling, and (3) correspondence between fungal and prokaryotic communities was greater in EM-dominated stands than in AM-EM mixed stands, but this did not correlate with the relative abundance of EM fungi. Our findings suggest that mycorrhizal fungi may affect the composition of non-mycorrhizal communities either directly or indirectly, through mycorrhizal effects on plant communities (Liang et al. 2020 ) and soil nutrient economies (Corrales et al. 2016b ). Ectomycorrhizal tree and fungal microbiome recruitment may have influenced the lower prokaryotic diversity we observed in EM-dominated than in mixed AM-EM stands. Tree species can independently recruit distinct bacterial communities (Oh et al. 2012 ). The EM-dominated stands we studied have a much lower overall tree species diversity than surrounding AM-EM mixed stands (Prada et al. 2017 ), potentially explaining higher prokaryotic community diversity in these stands. At our study site, EM leaf litter is not necessarily lower quality than AM leaf litter (Seyfried et al. 2021b ). However, leaf litter in EM-dominated stands may be more chemically homogenous than leaf litter in mixed species AM stands where a high diversity of AM tree species contributes litter ranging widely in chemical quality to the forest floor (Seyfried et al. 2021b ). Chemical heterogeneity could increase overall niche breath for microorganisms to support a highly diverse bacterial/archaeal community in AM-EM mixed stands. Additionally, mycorrhizal fungi may affect prokaryote communities directly by recruiting specific hyphosphere bacterial communities (Liu et al. 2018 ; Heděnec et al. 2020 ; Zhang et al. 2022 ). Specifically, AM fungi can support bacterial growth to facilitate inorganic nutrient transformations (Wang et al. 2023 ), while EM fungi may compete with free-living decomposers for organic nutrients (Fernandez and Kennedy 2016 ). These mycorrhizal hyphosphere responses also may be driven by differing patterns of belowground C allocation and root/fungal exudation between AM- and EM-associating tree and fungal species (Xu et al. 2023 ). Mycorrhizae and prokaryotic communities show strong patterns of interactions that may be driven by above- and belowground functional differences between AM and EM guilds. Shifts in alpha diversity in response to stand mycorrhizal type also corresponded to functional changes among prokaryotic taxa. The influence of EM-dominated versus AM-EM mixed stand type on microbiome recruitment could have driven the lower relative abundance of bacterial/archaeal groups associated with inorganic nutrient transformation processes, such as Nitrospirota (Myrold 2021 ), in EM-dominated stands relative to AM-EM mixed stands. Microbiota beneficial to AM fungi could have been inhibited either directly, or indirectly through EM effects on inorganic N availability (Phillips et al. 2013 ; Corrales et al. 2016b ). Further, there may be a relationship between these changes to functionally important taxa and EM-associated C dynamics. Ectomycorrhizal-dominated stands favored bacterial groups associated with diminished C mineralization activities, with greater Acidobacteriota and lower Bacteroidota relative abundance (Fierer et al. 2007 ) than in mixed AM-EM stands. These differences in the taxonomic composition of prokaryotic communities suggest that there may also be lower overall rates of C cycling and SOM decomposition in AM-EM mixed stands than in EM-dominated stands. Alternatively, O. mexicana may produce allelochemicals, as has been observed for other members of Juglandaceae (Jose and Gillespie 1998 ), which could be responsible for directly inhibiting specific soil microbiota (Revillini et al. 2023 ). This mechanism could potentially explain negative plant-soil feedbacks associated with O. mexicana legacy, although further investigation is necessary to determine the presence and identity of any possible allelopathic chemicals this plant could produce. Overall, suppression of C and N cycling by EM fungi and low tree diversity resulting in chemically homogenous root and leaf litter inputs in EM-dominated stands may promote specific microbiome assembly, potentially providing EM trees a competitive advantage. Overall fungal community beta diversity may have been driven by mycorrhizal interactions with their environment. We did not find a decrease in fungal alpha diversity in EM-dominated stands relative to mixed AM-EM stands as has been reported in temperate forests (Eagar et al. 2021 ). Rather, fungal communities were more heterogeneous among EM-dominated stands than among mixed AM-EM stands. Greater fungal community heterogeneity across EM-dominated stands could have been driven by soil pH and fertility which varied across our four watersheds and can select for functionally distinct EM fungal communities (Corrales et al. 2016a ). Specifically, low pH and fertility in Honda may select EM fungi that have a great capacity to alter C and N cycling through organic N uptake and that contribute abundant, low-quality fungal biomass to SOM pools (Seyfried et al. 2022 ). In contrast, relatively high soil pH and fertility in Alto Frio may select EM fungi which exclusively take up inorganic N and contribute limited, high-quality biomass to SOM pools (Seyfried et al. 2021a ). These selective forces on EM fungi may represent a disproportionate influence on the overall fungal community based on the high relative abundance of EM fungi in these EM-dominated stands. In mixed AM-EM stands, environmental filtering for functionally robust, less specialized fungal communities may have driven greater homogeneity across sites despite differences in underlying soil pH and fertility (Kivlin et al. 2018 ). Shifts in fungal communities between stand mycorrhizal types were also correlated with higher relative abundance of EM fungi and lower relative abundance of fungal pathogens and saprotrophs in EM-dominated stands than in the AM-EM mixed stands. Interactions among fungal guilds may help promote positive plant-soil feedbacks associated with EM mutualisms (Bennett et al. 2017 ) by creating favorable microbiomes (i.e., lower potential pathogen loads and decomposer abundance) in EM-dominated forest stands. The functional composition of soil fungal communities in EM-dominated forest stands may partially drive the effects of EM trees on ecosystem function (McGuire et al. 2010 ) and be influenced by underlying soil pH and fertility. The relationships between fungal and prokaryotic communities may highlight tradeoffs in belowground dynamics for different stand mycorrhizal types. Fungal and prokaryotic communities were more closely associated in EM-dominated than in mixed AM-EM stands. However, the divergence in this relationship (Procrustean residuals) increased with the relative abundance of fungal pathogens. Tradeoffs between defense, growth, nutrient acquisition, and mutualist collaboration in the development and morphology of root structures are an important component of plant development (Ravanbakhsh et al. 2019 ; Bergmann et al. 2020 ; Monson et al. 2022 ). For example, in tropical forests, the species most proficient at acquiring soil phosphorus are also the most vulnerable to pathogens (Laliberté et al. 2015 ; Lambers et al. 2018 ), with this tradeoff hypothesized to potentially help to maintain high plant diversity in these systems. Divergences between fungal and prokaryotic communities could be representative of tradeoffs made by plants to alter resource allocation for microbiome assembly in favor of pathogen protection. While this argument is largely speculative, the relationships we present here provide further support for microbial communities (or the capacity to manipulate them) as an extended plant root trait, which may be influenced by the surrounding environmental context (Freschet et al. 2021 )."
} | 4,819 |
25622822 | PMC4306973 | pmc | 8,178 | {
"abstract": "Vanillin dehydrogenase (VDH) is a crucial enzyme involved in the degradation of lignin-derived aromatic compounds. Herein, the VDH from Corynebacterium glutamicum was characterized. The relative molecular mass (Mr) determined by SDS-PAGE was ~51kDa, whereas the apparent native Mr values revealed by gel filtration chromatography were 49.5, 92.3, 159.0 and 199.2kDa, indicating the presence of dimeric, trimeric and tetrameric forms. Moreover, the enzyme showed its highest level of activity toward vanillin at pH 7.0 and 30C, and interestingly, it could utilize NAD + and NADP + as coenzymes with similar efficiency and showed no obvious difference toward NAD + and NADP + . In addition to vanillin, this enzyme exhibited catalytic activity toward a broad range of substrates, including p -hydroxybenzaldehyde, 3,4-dihydroxybenzaldehyde, o -phthaldialdehyde, cinnamaldehyde, syringaldehyde and benzaldehyde. Conserved catalytic residues or putative cofactor interactive sites were identified based on sequence alignment and comparison with previous studies, and the function of selected residues were verified by site-directed mutagenesis analysis. Finally, the vdh deletion mutant partially lost its ability to grow on vanillin, indicating the presence of alternative VDH(s) in Corynebacterium glutamicum . Taken together, this study contributes to understanding the VDH diversity from bacteria and the aromatic metabolism pathways in C. glutamicum.",
"discussion": "Discussion The -ketoadipate pathway is the major catabolic route for lignin-derived aromatic compounds in soil bacteria 17 18 19 . In the present study, we cloned, expressed and functional characterized a vdh gene from C. glutamicum , which channeled a variety of lignin-derived aromatic compounds to the protocatechuate branch of -ketoadipate pathway for further degradation. Based on the genome sequence of C. glutamicum ATCC13032 5 20 , one putative aldehyde dehydrogenase gene, vdh ATCC13032 , was identified. But this gene does not show a remarkably high level of homology to those with verified vanillin dehydrogenase activity ( Fig. 1 ). Our study suggests that VDH ATCC13032 is probably a unique aldehyde dehydrogenase with special catalytic roles, as discussed in the following. The conclusion that VDH ATCC13032 is a vanillin dehydrogenase is based on at least three lines of independent evidence. First, analysis of a vdh deletion mutant revealed a delayed growth when 3, 4-dihydroxy benzaldehyde, 3-hydroxy benzaldehyde, vanillin, or ferulic acid was present as the sole carbon source, suggesting an important role of vdh ATCC13032 in assimilation of these compounds. It is known that caffeic acid is not a direct target of VDH. However, when grown in the presence of caffeic acid, the vdh deletion mutant showed a delayed growth as well, supporting the idea that vdh is at the central pathway for catabolism of aromatic compounds. The growth defects of the mutant strain were complemented by expressing wild type vdh ( Fig. 2 ). Thus the resulting phenotype was not due to the polar effects caused by deletion of the vdh gene. The delayed growth observed with the deletion mutant grown on the different substrates may indicate minor alternative pathways for catabolism of aromatic compounds. This is consistent with the Amycolatopsis sp . strain ATCC 39116, in which VDH also plays a significant role in the course of vanillin degradation, but the biotransformation from vanillin to vanillate were still observed when vdh gene was knocked out 21 . HPLC-MS was performed to examine the biotransformation products of VDH ATCC13032 in the presence of vanillin or 3,4-dihydroxybenzaldehyde. As expected, vanillin and 3,4-dihydroxybenzaldehyde were transformed to vanillate and 3,4-dihydroxybenzoic acid, respectively, confirming the activity of VDH ATCC13032 . Relatively high catalytic activity of purified VDH ATCC13032 was observed with p -hydroxylbenzaldehyde and vanillin, consistent with the fact that vdh gene is at the downstream of ferulate and p -coumarate metabolism pathway. Also, VDH ATCC13032 was heterologously expressed and its Mr was estimated by SDS-PAGE ( Fig. 3A ). Interestingly, the native molecular mass determined by gel filtration showed that VDH from C. glutamicum existed in the form of tetramers, trimers and dimers ( Fig. 3B and C ). This is consistent with previous reports that VDHs from Micrococcus sp. TA1 and Burkholderia cepacia TM1 exist as tetramers and dimers, respectively 22 . The enzyme displayed the highest activity at 30C, in accordance with the physiological temperature of C. glutamicum and the results of in vitro expression assay. VDH from most of the studied species showed specificity against vanillin and benzaldehyde compounds. While purified VDH from C. glutamicum ATCC13032 showed catalytic activity toward a relatively broad range of tested substrates ( Table S1 ), oxidation rate for phenylacetaldehyde, formaldehyde and aldehyde were not detected. The observed substrate specificity is consistent with the substrate specificity observed with vanillin dehydrogenases from S. paucimobilis SYK-6 14 and P. fluorescens 9 . While most VDHs identified so far tend to use NAD + as a sole cofactor, VDH ATCC13032 could utilize both NAD + and NADP + , which has been observed with the VDH from P. fluorescens in a previous study 9 . Taken together, these results have revealed the important roles of VDH in C. glutamicum ATCC13032, whereas substrate specificity may vary from one species to another. Extensive works have been done to examine the active amino acid residues in the ALDH enzymes, and the examined residues have been well documented 15 16 23 . Sequence alignment revealed conserved sequences in VDH ATCC13032 with relatively high amino acid similarity with the active residues identified in ALDH and Pa BADH. One strictly conserved residue is Cys-292 in VDH ATCC13032 , which has been demonstrated to be responsible for the dehydrogenase as well as esterase activities of aldehyde dehydrogenase 16 24 . In P. aeruginosa , the Pa BADH catalytic cysteine (C286, corresponding to C292 in VDH ATCC13032 ) is oxidized to sulfenic acid or forms a mixed disulfide with 2-mercaptoethanol 19 . The second active residue identified in VDH ATCC13032 was Glu-258, which is in the vicinity of the catalytic cysteine, corresponding to E268 in ALDH2, E252 in PaBADH and E258 in VDH. This conserved residue probably functions as a general base in ALDH catalysis 25 . Lys-180 is another conserved residue in VDH ATCC13032 and may have functions similar to that of ALDH (K78) and Pa BADH (K162) 16 23 . Glu-199 in the VDH ATCC13032 may also be an conserved residue that would produce a steric clash with the 2-phosphate of NAD(P) + , resulting in a low affinity of ALDH2 for NAD(P) + 16 . Therefore we speculated that these five conserved sites may be important for the catalytic activity of VDH ATCC13032 , and they were subjected to further characterization. Mutations of these residues decreased catalytic activities by more than 50% compared with wild type when NAD + was used as a cofactor. However, when NADP + was used as a cofactor, mutations of N157A, E258A and C292A caused 90% reduction of catalytic activities compared with wild type, while E199A mutation only reduced the activity by 12%. This indicates that Glu-199 may have less influence on NADP + binding compared with the other tested residues, whereas all these residues play roles in NAD + binding. Consistently, Glu-199 showed no effect on the affinity of VDH ATCC13032 to NADP + , but affected the affinity of VDH ATCC13032 for the substrates such as vanillin. Taken together, the examined residues play critical roles in VDH catalysis, but their mechanisms may be different. Lignocellulosic hydrolysates for biofuel production usually contain not only fermentable sugars but also non-fermentable growth inhibitors, including furan, weak acids, and various lignin-derived aromatic compounds (such as vanillin, ferulic acid, p -coumaric acid, 4-hydroxybenzoic acid, vanillic acid, etc.), which inhibit microbial fermentation to the desired products 26 27 . C. glutamicum could be applied to both detoxify and assimilate lignin-derived aromatic inhibitors as an alternative source to sugars for carbon and energy. The functional characterization of VDH ATCC13032 contributes not only to a systematical understanding of aromatic compound assimilation, but also to develop C. glutamicum as an efficient strain to convert lignocellulose to bioproducts, such as biofuels."
} | 2,156 |
18692469 | null | s2 | 8,183 | {
"abstract": "Quorum sensing, a process of bacterial cell-cell communication, relies on production, detection, and response to autoinducer signaling molecules. LuxN, a nine-transmembrane domain protein from Vibrio harveyi, is the founding example of membrane-bound receptors for acyl-homoserine lactone (AHL) autoinducers. We used mutagenesis and suppressor analyses to identify the AHL-binding domain of LuxN and discovered LuxN mutants that confer both decreased and increased AHL sensitivity. Our analysis of dose-response curves of multiple LuxN mutants pins these inverse phenotypes on quantifiable opposing shifts in the free-energy bias of LuxN for occupying its kinase and phosphatase states. To understand receptor activation and to characterize the pathway signaling parameters, we exploited a strong LuxN antagonist, one of fifteen small-molecule antagonists we identified. We find that quorum-sensing-mediated communication can be manipulated positively and negatively to control bacterial behavior and, more broadly, that signaling parameters can be deduced from in vivo data."
} | 268 |
23611564 | null | s2 | 8,184 | {
"abstract": "The mechanical properties of the extracellular matrix (ECM) in which cells reside have emerged as an important regulator of cell fate. While materials based on natural ECM have been used to implicate the role of substrate stiffness for cell fate decisions, it is difficult in these matrices to isolate mechanics from other structural parameters. In contrast, fully synthetic hydrogels offer independent control over physical and adhesive properties. New synthetic materials that also recreate the fibrous structural hierarchy of natural matrices are now being designed to study substrate mechanics in more complex ECMs. This perspective examines the ways in which new materials are being used to advance our understanding of how ECM stiffness impacts cell function."
} | 191 |
36844461 | PMC9961374 | pmc | 8,185 | {
"abstract": "Hydrogen is a clean, renewable energy source, that when combined with oxygen, produces heat and electricity with only water vapor as a biproduct. Furthermore, it has the highest energy content by weight of all known fuels. As a result, various strategies have engineered methods to produce hydrogen efficiently and in quantities that are of interest to the economy. To approach the notion of producing hydrogen from a biological perspective, we take our attention to hydrogenases which are naturally produced in microbes. These organisms have the machinery to produce hydrogen, which when cleverly engineered, could be useful in cell factories resulting in large production of hydrogen. Not all hydrogenases are efficient at hydrogen production, and those that are, tend to be oxygen sensitive. Therefore, we provide a new perspective on introducing selenocysteine, a highly reactive proteinogenic amino acid, as a strategy towards engineering hydrogenases with enhanced hydrogen production, or increased oxygen tolerance.",
"introduction": "1 Introduction C1-utilizing microbes, microorganisms which rely on one carbon molecule for survival, have been of interest to produce biofuels for industrial use ( Du et al., 2011 ). Advances in metabolic engineering have led to the design of biosynthetic pathways as a means to efficiently use cellular machinery ( Bar-Even et al., 2010 ). One application of these engineering strategies is to utilize the activity of [NiFe]- and [FeFe]-hydrogenases in C1 microbes. Hydrogenases are enzymes that catalyze the reversible oxidation of hydrogen and are used for hydrogen production, a renewable source of energy. To compete with existing chemical methods for hydrogen production, hydrogenases require a significant hydrogen production rate ( Khanna and Lindblad, 2015 ). Furthermore, the highest hydrogen producing hydrogenases are also the most oxygen sensitive, reducing their efficiency within these microbial factories. Detailed studies on the factors driving hydrogen production and oxygen sensitivity have facilitated engineering strategies to overcome this ( Wittkamp et al., 2018 ). More specifically, an investigation into the role of selenocysteine (Sec) in these key processes for the development of novel hydrogenases increases the applicability for industrial purposes ( Marques et al., 2017 ; Evans et al., 2021 ). Sec, a homolog of cysteine (Cys), is found in redox-associated enzymes across all domains of life ( Li et al., 2014 ). With a single sulfur to selenium replacement compared to Cys, Sec retains similar chemistry but with enhanced chemical properties (i.e., increased nucleophilicity, decreased side-chain p K a, and increased oxidation which is often reversible) ( Chung and Krahn, 2022 ). The distinct characteristics of this amino acid and its similarity to Cys suggests it is potential to affect the active site electronic properties, catalytic rate, or oxygen sensitivity of hydrogenases ( Hondal et al., 2013 ; Marques et al., 2017 ; Evans et al., 2021 ). Occurring naturally in bacteria, Sec is incorporated in proteins at a UGA codon that immediately precedes a hairpin loop (known as the Sec insertion sequence [SECIS] element) in the translated region of the mRNA. Biosynthesis of Sec occurs on tRNA Sec , where it is first aminoacylated with serine (Ser) by seryl-tRNA synthetase (SerRS) and then converted to Sec by selenocysteine synthase (SelA) ( Fu et al., 2018 ). SelA uses selenophosphate as a substrate for this conversion, provided by selenophosphate synthetase (SelD) ( Stock and Rother, 2009 ). The resulting Sec-tRNA Sec is recognized by a specialized elongation factor (SelB) for peptide elongation in the ribosome. SelB resembles the canonical elongation factor EF-Tu, but with a C-terminal extension for interaction with the SECIS element ( Stock and Rother, 2009 ). This complex and highly regulated path for insertion of Sec ( Figures 1A , B ) has been an obstacle for recombinant selenoprotein production ( Fu et al., 2018 ). In this perspective, we discuss the details of an emerging technology for site-specific Sec insertion in Escherichia coli and how it can be adapted to cell factories. We focus on applying these cell factories for hydrogen production, highlighting recent evidence on the novel properties imparted by Sec on hydrogenases."
} | 1,081 |
30107031 | PMC6175206 | pmc | 8,186 | {
"abstract": "In the model cyanobacterium Synechocystis sp. PCC 6803, the terminal enzyme of chlorophyll biosynthesis, chlorophyll synthase (ChlG), forms a complex with high light‐inducible proteins, the photosystem II assembly factor Ycf39 and the YidC/Alb3/OxaI membrane insertase, co‐ordinating chlorophyll delivery with cotranslational insertion of nascent photosystem polypeptides into the membrane. To gain insight into the ubiquity of this assembly complex in higher photosynthetic organisms, we produced functional foreign chlorophyll synthases in a cyanobacterial host. Synthesis of algal and plant chlorophyll synthases allowed deletion of the otherwise essential native cyanobacterial gene. Analysis of purified protein complexes shows that the interaction with YidC is maintained for both eukaryotic enzymes, indicating that a ChlG‐YidC/Alb3 complex may be evolutionarily conserved in algae and plants.",
"discussion": "Discussion The processes of chlorophyll biosynthesis and photosystem assembly in phototrophic organisms appear to be co‐ordinated, ensuring efficient channelling of newly produced chlorophyll pigments to de novo photosystem polypeptides to enable their cotranslational insertion into the thylakoid membrane and assembly into functioning photosystems 27 , 28 . Chidgey et al . 7 previously identified the ChlG‐HliD‐Ycf39‐YidC assembly complex in the model cyanobacterium Synechocystis . To gain insight into whether similar ChlG complexes may form in higher oxygenic phototrophs, FLAG‐tagged algal and plant chlorophyll synthases were heterologously produced in Synechocystis ; the enzymes replaced the function of the native cyanobacterial enzyme, allowing subsequent deletion of the normally essential endogenous chlG gene resulting in strains that had the same growth rate and pigment composition as the WT and a strain producing the FLAG‐tagged native enzyme. Immunoprecipitations of the FLAG‐tagged ChlG proteins showed that only the most closely related ChlG from Synechococcus sp. PCC 7002 eluted with HliD and Ycf39. Synechococcus sp. PCC 7002 has close homologues of the Synechocystis Ycf39 (SYNPCC7002_A0216) and HliD (SYNPCC7002_A0858) and it is very likely that the same ChlG‐Hlip‐Ycf39‐YidC complex is conserved in this and other related cyanobacteria. Conversely, HliD and Ycf39 were absent from the plant and algal complexes, indicating that the eukaryotic chlorophyll synthases do not interact with these cyanobacterial proteins. HliD binds β‐carotene and chlorophyll a allowing the dissipation of absorbed light energy by chlorophyll to β‐carotene energy transfer 5 , 10 , consistent with the proposed role of Hlips in photoprotection of chlorophyll‐binding proteins 29 , 30 , 31 . It is possible that similar interactions with Hlip‐like proteins occur in algae and plants 32 . Arabidopsis one‐helix protein 2 (OHP2), which shows sequence similarity to HliD, associates with YCF244, a homologue of cyanobacterial Ycf39 33 . There is evidence for a functional similarity between OHP2 and HliD; an OHP2‐YCF244 complex associates with the PSII intermediate RCII 33 essentially as described for the HliD‐Ycf39 complex in cyanobacteria 13 . Light harvesting‐like 3 protein (LIL3) is another Hlip‐like protein that binds pigments and interacts with chlorophyll biosynthesis enzymes (ChlP and protochlorophyllide oxidoreductase) 34 , 35 , 36 , although Hey et al . 36 did not find a LIL3–ChlG interaction in Arabidopsis . Ycf39 forms a subcomplex with HliD that appears to associate either with ChlG or the early PSII assembly intermediate RCII 13 . The discovery of the ChlG‐HliD‐Ycf39‐YidC complex led to the hypothesis that ChlG associated with HliD and Ycf39 binds the PSII core subunit D1 precursor protein (pD1) as it is being cotranslationally inserted into the membrane by YidC, during which time chlorophyll provided by ChlG can be bound to the polypeptide 7 , 13 . Consistent with this, the Arabidopsis Ycf39 homologue HCF244 is important for translational initiation of psbA mRNA 37 . The Synechocystis strains containing the algal/plant enzymes display no obvious growth or pigmentation phenotype, despite the apparent lack of a ChlG‐HliD/Ycf39 interaction, suggesting that the roles of HliD and Ycf39 within the ChlG complex are not essential to cyanobacteria in vivo , at least under the low‐stress conditions used here. This is consistent with the lack of a growth rate or pigment composition defect in Synechocystis Δ hliD \n 7 and Δ ycf39 \n 13 , 38 mutants. We additionally found that Ycf39 is lost from the FLAG‐6803 ChlG complex under high light stress. Photo‐damaging conditions result in the release of chlorophylls from photosystems; these pigments must be recycled back to the membrane. A Δ ycf39 mutant is more sensitive to photoinhibition, and there is evidence that the Ycf39‐Hlip complex has a role in chlorophyll recycling and D1 incorporation into PSII during sudden exposure to high irradiance 13 . Whether Ycf39 is lost from the complex under other stress conditions, and the mechanism of its release from the ChlG complex, will require further study. Unlike HliD and Ycf39, YidC was detected in the immunoprecipitation eluates for all enzymes, including the nonfunctional BchG. In the phototrophic bacterium Rba. sphaeroides, the photosystem assembly factor LhaA was found to comigrate in CN‐PAGE with the integral membrane protease FtsH, BchG and YidC 39 , so (bacterio)chlorophyll synthase‐YidC associations might be widespread in phototrophs. YidC/Alb3 is a member of the evolutionally conserved protein family of membrane insertases 14 , 40 and is essential for thylakoid membrane biogenesis in cyanobacteria, algae and plants 16 , 17 . The discovery of an association between YidC and ChlG led to the hypothesis that YidC fixes chlorophyll‐binding proteins into a configuration that allows for the insertion of newly synthesized chlorophyll molecules from the neighbouring ChlG 7 , 41 . Unlike HliD, which is visible on stained gels, YidC is not observable by Coomassie blue staining, indicating it is present in the complex in a less than 1 : 1 ratio with ChlG/HliD; this could be confirmed by determination of molar quantities of FLAG‐ChlG, HID and YidC by mass spectrometry in future studies. Nonetheless, the observed interaction between algal/plant ChlG proteins and cyanobacterial YidC provides evidence that these proteins may form similar interactions with Alb3 in their native organisms, implying that co‐ordinated delivery of chlorophyll to nascent light‐harvesting polypeptides via ChlG‐YidC/Alb3 interactions is conserved among photosynthetic organisms. Arabidopsis thaliana contains a paralog of Alb3 called Alb4, which is required for chloroplast biogenesis 42 and thylakoid protein targeting 43 ; it is possible the plant ChlG may also interact with this protein. Coimmunoprecipitations with solubilised Arabidopsis or spinach thylakoids will allow the in vivo partner proteins of the plant enzyme to be confirmed."
} | 1,763 |
24228812 | null | s2 | 8,188 | {
"abstract": "A capability that is key to increasing the performance of paper microfluidic devices is control of fluid transport in the devices. We present dissolvable bridges as a novel method of manipulating fluid volumes within paper-based devices. We demonstrate and characterize the operation of the bridges, including tunability of the volumes passed from 10 μL to 80 μL, using parameters such as geometry and composition. We further demonstrate the utility of dissolvable bridges in the important context of automated delivery of different volumes of a fluid from a common source to multiple locations in a device for simple device loading and activation."
} | 162 |
37303603 | PMC10254463 | pmc | 8,189 | {
"abstract": "The hybrid electromagnetic-triboelectric generator (HETG) is a prevalent device for mechanical energy harvesting. However, the energy utilization efficiency of the electromagnetic generator (EMG) is inferior to that of the triboelectric nanogenerator (TENG) at low driving frequencies, which limits the overall efficacy of the HETG. To tackle this issue, a layered hybrid generator consisting of a rotating disk TENG, a magnetic multiplier, and a coil panel is proposed. The magnetic multiplier not only forms the EMG part with its high-speed rotor and the coil panel but also facilitates the EMG to operate at a higher frequency than the TENG through frequency division operation. The systematic parameter optimization of the hybrid generator reveals that the energy utilization efficiency of EMG can be elevated to that of rotating disk TENG. Incorporating a power management circuit, the HETG assumes the responsibility for monitoring the water quality and fishing conditions by collecting low-frequency mechanical energy. The magnetic- multiplier-enabled hybrid generator demonstrated in this work offers a universal frequency division approach to improve the overall outputs of any hybrid generator that collects rotational energy, expanding its practical applications in diverse multifunctional self-powered systems.",
"introduction": "Introduction Rapid development of the next-generation wearable and autonomous devices intensified the need for a reliable energy supply. Currently, chemical batteries are the primary power source for these electronics, which are plagued by poor battery life, electrolyte leaks, and environmental contamination. Thus, how to collect energy from the ambient environment and convert it into efficient electricity is a hot issue that urgently needs to be addressed for the alternative means of conventional energy supply. Among the various energy sources, mechanical energy is the one that deserves to be investigated in terms of its abundance, accessibility, and ubiquity in surroundings. Harvesters for mechanical energy typically involve in electromagnetic [ 1 , 2 ], triboelectric [ 3 – 5 ], and piezoelectric effects [ 6 – 8 ], where the electromagnetic generator (EMG) and triboelectric nanogenerator (TENG) are 2 most efficient approaches [ 9 – 11 ]. The Faraday’s law of electromagnetic induction-based EMG requests the coordination of coils and magnets, which produces a high output current but low output voltage and excels in high-frequency energy scavenge [ 12 – 14 ]. Through the combination of triboelectrification and electrostatic induction, TENG exhibits superior performance in harvesting low-frequency energy with the advantages of diverse material selection, simple structure, low cost, and light weight [ 15 – 17 ]. Meanwhile, compared to single EMG or TENG device, the operating frequency range is broadened obviously by the introduction of hybrid electromagnetic-triboelectric generator (HETG) [ 18 , 19 ]. Since 2015, HETG has been designed into various structures to apply in biomechanical [ 20 – 23 ], wind [ 24 – 27 ], vibration [ 28 – 31 ], and wave energy harvesting [ 32 – 35 ]. However, the majority of HETGs consist of a simple combination of EMG and TENG operating at the same frequency, which necessitates a driving source with a wide input frequency range or allows only 1 of the 2 components to operate at its appropriate operating frequency. Hu et al. [ 36 ] conducted a detailed quantitative comparison between TENG and EMG, which reveals that TENG has a superior energy utilization efficiency compared to EMG at low frequencies. At a frequency of 2.5 Hz, TENG exhibits a remarkable energy utilization efficiency of 75%, whereas EMG yields only 3%. The highest energy utilization efficiency of TENG, reaching 98.84%, is observed at a frequency of 5 Hz. However, as the frequency increases, the energy utilization efficiency of TENG declines, while that of EMG gradually increases. That is, the optimal frequency ranges for energy utilization efficiency differs between TENG and EMG. Consequently, a division frequency strategy is crucial for the 2 components of HETG to operate in their respective frequency domains, thereby enhancing the energy utilization efficiency. In this work, a layered hybrid generator made by a rotating disk TENG (RTENG), a magnetic multiplier, and a coil panel is designed to realize the division frequency operation of TENG and EMG. Mechanism analysis uncovers that the magnetic multiplier facilitates the regulation of transmission ratio and assists in achieving a faster operating frequency of EMG than TENG. The parameters affecting RTENG (grating degree of electrode, frequency, and tribo-layer) and EMG (transmission ratio and frequency) outputs are systematically studied. The proposed strategy allows for an effective energy utilization efficiency of EMG comparable to that of RTENG at the same driving frequency, which is a substantial improvement in comparison to the hybrid generator's operation at cofrequency. Moreover, the magnetic-multiplier-enabled hybrid generator (MMHG) can be used to build self-powered water quality monitoring and fishing alarm systems benefiting from a power management circuit (PMC). This work renders a feasible approach toward low-frequency rotational energy harvesting with high energy utilization efficiency by means of division frequency strategy.",
"discussion": "Discussion In this work, we designed a MMHG to facilitate the 2 parts of MMHG to function simultaneously at divided frequency. TENG moves at the same frequency as the low-speed rotor of the magnetic multiplier, while the output of EMG relies on the working frequency of the high-speed rotor, which is achieved through internal harmonic magnetic field coupling. Based on this structure design, systematically investigations about the factors affecting both RTENG and EMG outputs are implemented. Under the action of the rectification bridge, the effective energy utilization efficiency of EMG reaches the same level of RTENG. Furthermore, incorporating with a PMC, the MMHG can continuously supply power for acidimeter, TDS tester, and fish alarm, establishing a self-powered water quality and fishing monitoring system. This frequency division strategy demonstrated in this work provides an effective way for hybridized generators to achieve high energy utilization efficiency in low-frequency rotational energy harvesting, which paves stepping stones for further development of the self-powered systems."
} | 1,619 |
33668634 | PMC7918521 | pmc | 8,190 | {
"abstract": "Revealing the relationship between taxonomy and function in microbiomes is critical to discover their contribution to ecosystem functioning. However, while the relationship between taxonomic and functional diversity in bacteria and fungi is known, this is not the case for archaea. Here, we used a meta-analysis of 417 completely annotated extant and taxonomically unique archaeal genomes to predict the extent of microbiome functionality on Earth contained within archaeal genomes using accumulation curves of all known level 3 functions of KEGG Orthology. We found that intergenome redundancy as functions present in multiple genomes was inversely related to intragenome redundancy as multiple copies of a gene in one genome, implying the tradeoff between additional copies of functionally important genes or a higher number of different genes. A logarithmic model described the relationship between functional diversity and species richness better than both the unsaturated and the saturated model, which suggests a limited total number of archaeal functions in contrast to the sheer unlimited potential of bacteria and fungi. Using the global archaeal species richness estimate of 13,159, the logarithmic model predicted 4164.1 ± 2.9 KEGG level 3 functions. The non-parametric bootstrap estimate yielded a lower bound of 2994 ± 57 KEGG level 3 functions. Our approach not only highlighted similarities in functional redundancy but also the difference in functional potential of archaea compared to other domains of life.",
"conclusion": "5. Conclusions Our results suggest a limited contribution of archaea to the total functional potential of the microbiome, with most archaeal functions already identified as of today. However, the existence of archaea-specific functions must be validated by novel and more sophisticated methods. The accumulation curve describing the increase of functional categories with the number of sequenced genomes in archaea was closer to the asymptote than in bacteria [ 13 ] and fungi [ 12 , 13 , 14 ]. This made the estimate of archaeal contribution to the total microbiome functionality more precise, although it is still uncertain if the functional diversities of different domains can easily be compared. Noteworthy and similar to fungi, only one quarter of all genes in archaeal genomes on average were affiliated with a KEGG function, which demonstrates the limitations of the annotation because the prediction of microbiome functionality technically excluded three quarters of the entire functional potential in archaea. Different ortholog databases such as COG or Pfam could further improve our understanding of functional diversity, especially in archaea, as those covered three times more genes than KEGG did. Still, different approaches and definitions of functions are necessary to estimate the actual functional diversity of the microbiome.",
"introduction": "1. Introduction The biochemical transformations conducted by a community of microbes from all domains of life mediate ecosystem functioning [ 1 ]. Even though ecological studies tend to focus on bacteria and fungi, archaea as a major part of global ecosystems [ 2 ] are ubiquitous in both terrestrial and aquatic environments [ 3 , 4 ]. Particularly, archaea make up between 20% and 30% of the total prokaryotes in marine environments [ 5 ] and between 0% and 10% in soil environments [ 4 , 6 ]. Increases in the proportion of archaea were found in extreme habitats such as acidity and low temperature [ 7 ]. Functionally, archaea play key roles in global carbon (e.g., methanogenesis or CO 2 fixation) and nitrogen (e.g., N 2 fixation or oxidation of ammonia) cycles [ 8 ], but they also have complex relationships with both bacteria and fungi [ 9 ]. In a microbial community, multiple organisms with different taxonomy may have similar if not identical roles in ecosystem functionality, the so-called functional redundancy [ 10 ]. Indeed, interspecies redundancy was reported to be very high with several hundreds to thousands of different taxa to express the same function in a habitat [ 11 ]. These functions can be statistically inferred based on the homology to experimentally characterized genes and proteins in specific organisms to find orthologs present in a microbiome [ 12 , 13 , 14 ]. This ortholog annotation is used by KEGG Orthology [ 15 , 16 ], which covers a wide range of functional classes (level 1 of KEGG) comprising cellular processes, environmental information processing, genetic information processing, human diseases, metabolism, organismal system, brite hierarchies, and functions not included in the annotation of the two databases pathway or brite. KEGG level 2 functions provide more detail, i.e., differentiating glycoside hydrolases and glycosyltransferases within carbohydrate active enzymes (level 1), whereas KEGG level 3 is the enzyme itself, i.e., the glycogen phosphorylase (K00688, EC: 2.4.1.1.) that belongs to the glycosyltransferases (more information can be obtained under https://www.genome.jp/kegg/kegg3.html , accessed on 17 July 2020). However, the bottleneck of reporting microbiome functions is the low number of fully sequenced and annotated genomes as only organisms are captured that have undergone isolation and extensive characterization [ 12 , 13 , 14 ] with respect to the expected total diversity. Hence, the lower the share of known species or the higher the predicted total diversity, the weaker the prediction itself. Problematically, the vast majority of organisms have not yet been studied [ 17 , 18 ] which is why the annotation is based on the similarity to the genomes of the very few studied model organisms [ 12 , 13 , 14 ]. Consequentially, microbiome functionality is inferred based on the taxonomic composition of the community and its relation to functional parameters [ 19 ], indicated by the frequent use of the 16S rRNA gene metabarcoding to describe the prokaryotic community. Even though the description of microbial communities is important to assess the drivers of the occurrence of individual taxa and the composition of their communities [ 12 ], the mere taxonomic composition itself did not provide detailed answers about its functional diversity [ 20 ]. The functional diversity for both bacteria [ 13 ] and fungi [ 12 , 13 , 14 ] were recently predicted to comprise millions of different functions using meta-analyses of proteins [ 13 ] and genomes [ 12 , 14 ], most of which are unknown today. However, our understanding of functional redundancy in archaea and their contribution to the total microbiome functionality is still scarce. Here, we used both parametric and non-parametric estimators of functional richness with the aim to predict the total archaeal functionality on Earth and to unveil the relationship between taxonomy and function in the archaeal domain. To do so, we obtained all completely annotated genomes of taxonomically unique archaeal species (n = 417) from the integrated microbial genomes and microbiomes (IMG) of the Joint Genome Institute (JGI) ( https://img.jgi.doe.gov/ , accessed on 17 July 2020) with taxonomic annotation on the species level and functional annotation of KEGG on level 3 (referred to as KEGG function). We used a parametric estimation based on accumulation curves (AC) [ 21 ] that are characterized by the increasing number functions with increasing species. The AC was fitted to saturated, unsaturated, and logarithmic models, and the best fitting model was chosen based on its fitness in comparison to the other models. As a non-parametric estimator, Chao-1 was used for every 50 randomly picked species of all 417 in the database each with 20 replicates. The precision of both the parametric and the non-parametric approach generally depends on the proximity to the asymptote of the model, with greater extrapolation to the total count resulting in greater error [ 22 ]. We therefore hypothesized more precise estimates of the contribution of archaea to the total microbiome functionality than previously proposed for both bacteria [ 13 ] and fungi [ 12 , 13 , 14 ] due to the higher coverage of the predicted taxonomic diversity in archaea.",
"discussion": "4. Discussion 4.1. Genome Content The genome size of an organism generally reflects its developmental and ecological needs [ 32 ]. Larger genomes are directly related to increases in both cellular and nuclear volumes [ 33 ] that help to cushion fluctuations in concentrations of regulatory proteins or to protect coding DNA from spontaneous mutation [ 34 ]. Variation in genome size is therefore a result of the adaptive needs or of natural selection in different organisms [ 32 ]; the so-called adaptive theory of genome evolution. The smaller genomes of archaea could be directly related to a higher evolutionary rate [ 35 ]. Indeed, the 417 archaeal species generally comprised smaller genomes compared to bacteria [ 36 ], but with statistically significant differences among phyla, habitats, and temperature ranges that were mirrored by the number of KEGG level 3 functions in each genome. Particularly, archaea inhabiting extreme habitats such as deep sea or hot springs characteristic with high local temperatures not only had significantly fewer total genes but also fewer KEGG functions. Otherwise, environments of higher complexity and diversity such as soils or sediments contained archaea with a larger functional potential that may have allowed them more options for the competition for or the utilization of a wider range of nutrients. 4.2. Functional Redundancy A limited set of metabolic pathways found in a variety of taxonomic groups drive most biogeochemical reactions [ 37 ] which is why the diversity in the community is correlated strongly with its functional diversity [ 38 ]. Functions are classified into two groups [ 12 , 13 , 14 ]: (i) Highly redundant across different species present in more than 90% of all species or (ii) unique to only a few species present in less than 10% of all species. Here, intergenome redundancy was either high or low for roughly two thirds of all the KEGG functions; fewer than the 77.3% were found in fungal genomes [ 12 , 14 ]. However, the presence of a higher share of functions of intermediate redundancy that are present between the two thresholds suggested the presence of more than two groups [ 12 , 13 , 14 ] that could be particularly important for organisms with smaller genomes such as archaea and bacteria. A set of functions present in a quarter of all archaea indicated that the presence of a driving phylotype or environment may drive intergenome redundancy. Indeed, most functions (151/194) with an intergenome redundancy between 22% and 26% belonged to the phylum Crenarchaeota , the habitat hot springs, and the temperature range of hyperthermophilic archaea, mainly affiliated with amino acid utilization, fermentation, methanogenesis, and nucleic acid metabolism. The median intergenome redundancy was twice as high as found for fungi before [ 12 , 14 ], implying a higher share of functions shared among archaea on average. However, only half the gene copies (1.02 in archaea compared to 2.0 in fungi) were present, highlighting the close relationship between intergenome and intragenome redundancy. Indeed, the archaeal genomes revealed that low intergenome redundancy is generally related to high intragenome redundancy and vice versa. Presumably, every organism must choose between additional copies of functionally important genes or a higher number of different genes, especially in reduced genomes. Similarly to the pattern found in fungi before [ 12 , 14 ], functions belonging to the maintenance apparatus such as S-adenosylmethionine synthetase (EC 2.5.1.6, K00789) involved in the biosynthesis of amino acids were with both high intergenome and high intragenome redundancy, allowing for more complex regulation of the gene, i.e., when more transcripts are needed. Otherwise, functions with low intergenome and low intragenome redundancy are highly specialized processes only performed by a few archaea such as the drug transporter MFS transporter, DHA1 family, multidrug resistance protein (K19578) found in the crenarchaeote Thermofilum adornatus . 4.3. Archaeal Contribution to the Total Microbiome Functionality The parametric approach estimated the archaeal contribution to the total microbiome functionality to roughly 4200 KEGG functions; a magnitude less than predicted for both bacteria [ 13 ] and fungi [ 12 , 13 , 14 ]. The lower bound estimate of functional richness derived from the non-parametric approaches yielded roughly 3000 KEGG functions. A plateau of functional richness with higher species richness made the predictions for archaea more reliable as the errors decreased with proximity to the asymptote [ 22 ]. Theoretically, a higher number of species must be sequenced until no additional functions are unveiled and the accumulation curve reaches the actual asymptote [ 39 ]. However, practically, this is nearly impossible as a prohibitively large number of species are needed to be sampled in order to reach an asymptote [ 40 ]. In our meta-analysis, admittedly, the 417 genomes of distinct archaeal species only spanned three archaeal phyla from all 21 proposed phyla [ 41 , 42 ] and covered only a small part of the predicted taxonomic diversity in archaea; with databases containing up to 13,159 archaeal species [ 26 ], the prediction of 5000 archaeal genera [ 43 ], and the finding of 669 distinct archaeal species among 10,575 prokaryotic genomes [ 44 ]. Hence, the addition of genomes from novel archaeal species with potentially new KEGG functions could change both the parametric and the non-parametric estimates of functional richness. However, the differences in the estimates are likely not as tremendous as the potential differences in the estimates for both bacteria [ 13 ] and fungi [ 12 , 13 , 14 ] as the accumulation curve already plateaued with 417 taxonomically distinct archaeal species. Noteworthy, it is unclear how well new functions are recovered in archaea. As there is notably less interest in archaea compared to bacteria, functional annotations might generally miss archaea-specific functions to a larger extent than bacteria-specific functions missed in bacteria. As of today, our understanding of the contribution of archaea to the total microbiome functionality covers the majority of the KEGG functions, but many as-yet unknown archaea-specific functions could exist."
} | 3,616 |
35415879 | PMC9323439 | pmc | 8,191 | {
"abstract": "Summary Coastal waters are a major source of marine methane to the atmosphere. Particularly high concentrations of this potent greenhouse gas are found in anoxic waters, but it remains unclear if and to what extent anaerobic methanotrophs mitigate the methane flux. Here we investigate the long‐term dynamics in methanotrophic activity and the methanotroph community in the coastal oxygen minimum zone (OMZ) of Golfo Dulce, Costa Rica, combining biogeochemical analyses, experimental incubations and 16S rRNA gene sequencing over 3 consecutive years. Our results demonstrate a stable redox zonation across the years with high concentrations of methane (up to 1.7 μmol L −1 ) in anoxic bottom waters. However, we also measured high activities of anaerobic methane oxidation in the OMZ core (rate constant, k , averaging 30 yr −1 in 2018 and 8 yr −1 in 2019–2020). The OPU3 and Deep Sea‐1 clades of the Methylococcales were implicated as conveyors of the activity, peaking in relative abundance 5–25 m below the oxic–anoxic interface and in the deep anoxic water respectively. Although their genetic capacity for anaerobic methane oxidation remains unexplored, their sustained high relative abundance indicates an adaptation of these clades to the anoxic, methane‐rich OMZ environment, allowing them to play major roles in mitigating methane fluxes.",
"introduction": "Introduction Microbial processes are the main source of the potent greenhouse gas methane, CH 4 , to the atmosphere, where it accounts for about one‐quarter of the increase in radiative forcing since the preindustrial era (IPCC, 2013 ; Etminan et al ., 2016 ). Marine methane emissions originate mainly from coastal and shelf environments, yet despite recent modelling and meta‐data analysis efforts, estimates of methane emissions from such systems remain uncertain (4–27 Tg CH 4 yr −1 ; Rosentreter et al ., 2021 ; Weber et al ., 2019 ), due to sparse measurements, high spatial and temporal variability, and changing conditions as a result of human influence (Rosentreter et al . 2021; Weber et al ., 2019 ). Notable areas of marine methane accumulation (tens of nanomolar to tens of micromolar) include anoxic basins and fjords as well as coastal upwelling systems, particularly those characterized by oxygen‐depleted subsurface waters known as oxygen minimum zones (OMZs; Capelle et al ., 2019 ; Kessler et al ., 2006 ; Naqvi et al ., 2010 ; Sansone et al ., 2001 ; Thamdrup et al ., 2019 ). Here, benthic archaeal methanogenesis is suggested to be the major methane source (Sansone et al ., 2001 ; Chronopoulou et al ., 2017 ), but molecular data have additionally revealed the potential for pelagic methanogenesis (Padilla et al ., 2016 ). It is, however, unclear to what extent methane fluxes from anoxic waters are mitigated by microbial oxidation, and the microorganisms and metabolic pathways potentially involved remain elusive. Nonetheless, to understand these processes is imperative given the predicted expansion of marine oxygen‐depleted waters (Breitburg et al ., 2018 ). Microbial methane oxidation can be performed both aerobically and anaerobically. The latter uses alternative oxidants such as sulfate, iron, manganese, nitrate and nitrite, with nitrate and nitrite being the most energetically favourable (see Guerrero‐Cruz et al ., 2021 and references therein). Overlapping distributions of oxidized nitrogen compounds and methane in oxygen‐depleted waters, for example in the eastern tropical North Pacific (ETNP) OMZ and Golfo Dulce, Costa Rica, suggest a niche for microbes coupling anaerobic methane oxidation to reductive nitrogen transformations (Padilla et al ., 2016 ; Padilla et al ., 2017 ; Thamdrup et al ., 2019 ). Measuring rates of anaerobic metabolisms in these waters requires extensive protocols to minimize oxygen contamination that could inhibit anaerobic pathways (De Brabandere et al ., 2012 ). Thamdrup et al . ( 2019 ) utilized procedures developed to minimize oxygen contamination and were able to quantify rates of anaerobic methane oxidation in the ETNP OMZ between 0.014 and 0.12 nmol L −1 d −1 . Highest rates were located between the nitrite and methane maxima and rates were consistently inhibited with 0.5 μmol L −1 of oxygen (Thamdrup et al ., 2019 ). Whether anaerobic methane oxidation plays a comparable role in other OMZ systems is still unknown as are the temporal dynamics of the process. Molecular analysis performed in anoxic marine waters indicates that diverse microbes may be responsible for linking the methane and nitrogen cycles. In the OMZs of the Golfo Dulce and ETNP, close relatives of ‘ Candidatus Methylomirabilis oxyfera’ of the NC10 clade have been shown to be present and transcriptionally active (Padilla et al ., 2016 ; Thamdrup et al ., 2019 ). Members of the NC10 are hypothesized to oxidize methane aerobically by producing intracellular oxygen through the dismutation of nitric oxide (NO) from nitrite reduction, which allows them to thrive in anoxic waters despite the obligate need for oxygen (Ettwig et al ., 2010 ). A coupling of anaerobic methane oxidation to NO in the ETNP OMZ was further indicated by rate measurements being inhibited by the presence of the NO scavenger PTIO (Thamdrup et al ., 2019 ). Further links to the nitrogen cycle in the ETNP OMZ have been indicated by the recovery of transcripts affiliated with ‘ Candidatus Methanoperedens nitroreducens’ of the ANME‐2d clade, known to couple anaerobic methane oxidation to nitrate reduction (Haroon et al ., 2013 ; Thamdrup et al ., 2019 ). Another group of methanotrophs detected in anoxic marine waters are members of the Gammaproteobacterial order Methylococcales, which are traditionally considered as aerobic methanotrophs (Hayashi et al ., 2007 ; Padilla et al ., 2017 ; Thamdrup et al ., 2019 ). Diverse marine clades of Methylococcales have been identified based on the phylogenetic marker gene particulate methane monooxygenase ( pmo ) and are referred to as operational pmo units (OPUs, Tavormina et al ., 2008 ). Among these, the clade OPU3 has been shown to be abundant in the ETNP OMZ, peaking in abundance at <4 μM of oxygen (Tavormina et al ., 2013 ), and in the anoxic core of the Golfo Dulce OMZ (Padilla et al ., 2017 ), the latter hinting at a potential role in anoxic waters. A metagenomic study performed in Golfo Dulce showed that an abundant member of the OPU3 clade could perform partial denitrification to NO (Padilla et al ., 2017 ) and the analysis of a related isolate strain likewise demonstrated the ability to link methane oxidation with denitrification to nitrous oxide (N 2 O) when oxygen concentrations were <50 nmol L −1 (Kits et al ., 2015 ). However, neither of the studies demonstrated methane oxidation completely independent of oxygen, since oxygen still appears to be required for the initial methane oxidation step. It, therefore, remains to be determined how members of OPU3 either obtain oxygen for methane oxidation or otherwise metabolize within anoxic waters. Here, we present data from a 3‐year (2018–2020) study of methane oxidation in the coastal OMZ of Golfo Dulce. Our aim was to explore the temporal and spatial dynamics of anaerobic methane oxidation in a biogeochemical and microbial context. To do so, each year we quantified rates of methane oxidation in the anoxic core of the OMZ, measured relevant biogeochemical parameters, and investigated the methanotroph community through 16S rRNA gene amplicon sequencing. We found that (i) the biogeochemical zonation in the OMZ, which had high concentrations (up to 1.7 μmol L −1 ) of methane, was stable between years. (ii) The OMZ supported high rates of anaerobic methane oxidation, (iii) most likely performed by members of Methylococcales, although the exact pathway remains unclear.",
"discussion": "Discussion The 3 years of data on biogeochemical zonation, methane oxidation rates and microbial community composition demonstrate that the anoxic waters of Golfo Dulce serve as a stable niche for Gammaproteobacterial methanotrophs and support high rates of anaerobic methane oxidation. Below we discuss methane dynamics within the bay and how the biogeochemical setting may shape the structure and function of the methanotrophic community, and thereby ultimately its role in mitigation of methane emissions in the coastal OMZ. Long‐term stability of the redox zonation in Golfo Dulce The three expeditions to the Golfo Dulce OMZ (2018–2020) revealed a consistent vertical redox zonation across years (Fig. 2 ), as well as along the bay (Fig. 1 ). The accumulation of nitrite at anoxic depths indicated active nitrate reduction. We detected nitrate at all anoxic depths in 2018 and 2019, while it was depleted (LOD 0.5 μM) at 180 m at the innermost four stations in 2020 (Fig. 1E , Sta. 1‐2a). The lack of detectable sulfide suggested that nitrate and nitrite were the preferred electron acceptors over sulfate at the depths examined. These conditions are similar to those found in open‐ocean OMZs (Ulloa et al ., 2012 ). The redox zonation observed in 2018–2020 was broadly similar to patterns shown in previous studies in Golfo Dulce since the first surveys along the bay in March 1969 (Richards et al . ( 1971 ). This prior work further includes a mapping of the transect in January 1994 (Thamdrup et al ., 1996 ; Ferdelman et al ., 2006 ) as well as profiles from selected stations in November 2001 and January 2015 (Dalsgaard et al ., 2003 ; Padilla et al ., 2016 ). One notable difference is that the previous studies have typically detected low levels of sulfide (≤7 μmol L −1 ) in deeper waters at the head of the bay with the most recent report from 2015 finding sulfide at and below 150 m at Sta. 1 (up to 6.6 μmol L −1 ; Padilla et al ., 2016 ). However, in 1994, sulfide was only detected in one profile from Sta. 1 and then disappeared a few days later (Ferdelman et al ., 2006 ). Thus, the differences between prior results and those from our study may potentially reflect minor short‐term variability rather than a long‐term trend. Overall, the similarity of data obtained over more than 50 years indicates a remarkable stability of the Golfo Dulce OMZ. Distribution and potential sources of methane Methane consistently accumulated with depth in the OMZ with highest concentrations near the sediment–water interface (Figs 1H, I and 2D, H ). The concentrations of methane in Golfo Dulce (≤1.7 μmol L −1 ) are high compared to those of open ocean OMZs (typically ≤0.1 μmol L −1 , Chronopoulou et al ., 2017 ; Jayakumar et al ., 2001 ; Sansone et al ., 2001 ) but similar to concentrations observed in the seasonally anoxic fjord of Saanich Inlet (1.3 μmol L −1 ; Capelle et al ., 2019 ) and anoxic basins such as the Black Sea and Cariaco (up to 17 μmol L −1 ; Kessler et al ., 2006 and references therein). The single previous study of methane concentrations in Golfo Dulce from January 2015 also reported an increase with depth, but only to 80 nmol L −1 at 180 m (Padilla et al ., 2016 ). At that time, bottom waters contained 6.6 μmol L −1 sulfide, in contrast with our study where sulfide was not detected. The higher concentrations of methane in 2018–2020 can therefore not be explained by more reduced conditions in the OMZ. We see no obvious explanation for this difference and further measurements are needed to determine whether such variability is a recurring phenomenon. The accumulation of methane towards the bottom waters could suggest that the sediments are the main methane source to the OMZ. No studies to date have analysed the biogeochemistry of methane in the sediments of Golfo Dulce. In general, upper layers of marine sediments contain low concentrations of methane, as the supply of sulfate from seawater typically favours sulfate reducers over methanogens in shallower sediment layers, and strongly attenuates the upward flux of methane from deeper layers through anaerobic oxidation in the sulfate methane transition zone (Iversen and Jorgensen, 1985 ; Knittel and Boetius, 2009 ). During the survey of Golfo Dulce in 1994 (Thamdrup et al ., 1996 ) high concentrations of sulfate (>24 mM) were measured in the upper 30 cm of the sediments, with no sign of decline with depth (B. Thamdrup unpubl. res.). A large benthic methane flux, therefore, seems unexpected considering that, in continental shelf sediments, methane concentrations in the sulfate zone are usually <0.01 mmol L −1 and sediments represent a negligible methane source to the water column (Iversen and Jorgensen, 1985 ; Niewöhner et al ., 1998 ; Beulig et al ., 2019 ). Other methane sources that could contribute to the high concentrations observed in the Golfo Dulce OMZ are seeps/vents and methanogenesis in the water column. Seeps and vents are common on the Costa Rican Pacific continental margin (Mau et al ., 2006 ; Sahling et al ., 2008 ) and methane seeps have been observed in Golfo Dulce. The currently documented seeps are located at water depths ≤10 m (Berrangé, 1987 ; Wild et al ., 2015 ), yet undiscovered methane seeps could be located deeper in the bay. Another possibility is that methane is produced in situ in the water column. This hypothesis is supported by the previous recovery of transcriptionally active methanogenic archaea in the Golfo Dulce OMZ (Padilla et al ., 2016 ) and our current observation of the putative methyl‐reducing methanogens ‘ Ca . Methanofastidiosa’ (Nobu et al ., 2016 ; Zhang et al ., 2020 ) in our 16S rRNA gene dataset (Figs S3 and S4). While the descriptions of ‘ Ca . Methanofastidiosa’ are from an anaerobic digester and mangrove sediment, we are not aware of published reports of their occurrence in OMZ waters. However, screening of the NCBI database identified a ‘ Ca . Methanofastidiosa’ sequence recovered from the Eastern Tropical South Pacific OMZ (Belmar et al ., 2011 ). Phylogenetic analysis subsequently revealed a clade within the ‘ Ca . Methanofastidiosa’ that exclusively contained sequences from pelagic habitats, including those from OMZs (Fig. S3). In Golfo Dulce, methylotrophic methanogens would likely have to compete for substrates against nitrate/nitrite respiring methylotrophs in the water column and might therefore be confined to reduced micro‐environments such as inside sinking particles (Beck et al ., 2014 ), including faecal pellets from anoxic gut environments of zooplankton (Stief et al ., 2017 ; Wäge et al ., 2020 ). There was a notable difference in the relative abundance of ‘ Ca . Methanofastidiosa’ between 2018 and 2020 (Fig. S4). Since the biogeochemical conditions remained similar and the sequencing procedure and analysis were the same between these 2 years, we see no obvious cause for the abundance difference. Nonetheless, the presence of ‘ Ca . Methanofastidiosa’ in the water column, and especially its high abundance in 2020, suggests this taxon may play a role in methane production in the OMZ. We suggest that both benthic and pelagic sources should be evaluated further through biogeochemical experiments and analysis of molecular capabilities in order to understand the origin of methane in the Golfo Dulce OMZ. A role for anaerobic methane oxidation The consistent observation of a steep increase in rate constants and rates below the oxic–anoxic interface indicates that the anoxic, nitrate‐, nitrite‐ and methane‐rich environment in the OMZ of Golfo Dulce is a niche for anaerobic methane oxidation. In our incubations, which were carried out with minimal oxygen contamination (<100 nmol L −1 oxygen, see Experimental procedures ), the activity from above the interface was far lower ( k < 1 yr −1 ) than in the OMZ (1–57 yr −1 ), with the exception of Sta. 1 in 2018 (HR1a), where rate constants in the oxycline ranged from 9 to 15 yr −1 (Fig. 2A ). These higher values might be due to greater mixing around the oxic–anoxic interface, as suggested by the 13‐m deepening in interface depth between sampling dates 4 days apart, which could have transported methanotrophs from deeper depths to the oxycline. Unfortunately, 16S rRNA gene sequence data were not available for HR1a in 2018, leaving us unable to explore features of the methanotroph community that might further explain the rate variation. Rate constants are assumed to scale with the population size of microbes involved in methane oxidation. The observed depth distribution of rate constants, therefore, suggests a community of methanotrophs specialized for OMZ conditions. The activity observed near the oxic–anoxic interface might still be influenced by occasional oxygen supply from mixing or photosynthesis (e.g. Tiano et al ., 2014 ; Garcia‐Robledo et al ., 2017 ). Based on the 16S rRNA gene sequence data, Cyanobacteria were present below the oxic–anoxic interface in Golfo Dulce at relative abundances between 0.5% and 1% (Fig. S5d). However, light (>0.1 μmol photons m −2 s −1 ) was never observed to penetrate below the interface (Fig. S5C and G). The highest relative abundance of Cyanobacteria occurred in the shallowest samples, suggesting that the Cyanobacteria observed in the OMZ were sinking and likely inactive, as no chlorophyll was detected below the oxic–anoxic interface either (Fig. S5B andF). Therefore, photosynthesis is unlikely to serve as an internal oxygen source for methane oxidizers in this OMZ (incubations were also carried out in the dark). While mixing is likely to supply oxygen occasionally to microbes in the upper OMZ, the likelihood decreases with the distance to the oxic–anoxic interface, and we never detected oxygen deeper than 110 m at our experimental stations. Thus, our finding of the highest rate constants measured at depths between 130 and 190 m suggests an anaerobic metabolism for methane oxidizers. Contamination with low levels of oxygen appears inevitable when performing laboratory incubations with OMZ waters (Ganesh et al ., 2015 ; Thamdrup et al ., 2019 ). Continuous monitoring of oxygen in our incubations constrained oxygen to low nanomolar amounts (<100 nmol L −1 ; see Experimental procedures ), which could potentially sustain microaerophilic methane oxidation as observed in lakes (Blees et al ., 2014 ; Oswald et al ., 2015 ) and coastal waters (Steinle et al ., 2017 ). Still, the methane oxidation rates of up to 243 nmol L −1 d −1 observed in our incubations could in several cases not be sustained by the maximum oxygen concentrations in incubations, given the expected oxygen to methane ratio of 2:1 for aerobic methane oxidation (Naguib, 1976 ), or even with a decreased ratio if parts of the metabolism are carried out anaerobically (Dam et al ., 2013 ; Kalyuzhnaya et al ., 2013 ). Moreover, the methanotrophs would face competition for any available oxygen from other members of the microbial community, such as oxygen‐respiring heterotrophs. Previously measured oxygen consumption rates from the Golfo Dulce OMZ reached as high as 1000 nmol L −1 d −1 (Garcia‐Robledo et al ., 2016 ), which in our incubations would largely deplete an oxygen pool of <100 nmol L −1 within the incubation period. In the kinetic experiments rates were linear throughout the incubation period, thus we saw no indications of a shift from aerobic to anaerobic methane oxidation. Taken together these lines of evidence strongly indicate that methane oxidation activity observed in the OMZ core of the Golfo Dulce was not dependent on oxygen and was anaerobic, as also concluded for similar experiments in the ETNP OMZ (Thamdrup et al ., 2019 ). Rate constants of anaerobic methane oxidation from the OMZ core in 2018 (6–57 yr −1 ) were high compared to those measured the subsequent 2 years (1–23 yr −1 , Fig. 3B, F, J ). The difference might in part be driven by the steeper methane gradients observed in 2018 (Fig. 2D ) that suggest a higher substrate flux (the density gradient in the OMZ core did not change between years, Fig. S6) that would be expected to sustain a more active community (Brune et al ., 2000 ). Notably, incubations from the OMZ core in 2018 were carried out with lower methane concentrations than in situ , while in situ concentrations were re‐established in the incubations in 2019 and 2020 (see Experimental procedures ). However, the methane concentration does not affect the calculation of k (see Experimental procedures ) and based on the first‐order kinetics observed in the OMZ (Fig. 4 ; S1) and typically assumed for methane oxidation (e.g. Valentine et al ., 2001 ), this difference should not affect the determination of rates. Alternatively, as the radiolabel constituted a larger fraction of the methane in the incubations from 2018 than in subsequent years, a faster turnover of the tracer relative to the unlabelled pool could potentially have led to higher rates in 2018. While we cannot fully exclude such a kinetic effect, we note that previous comparisons of methane oxidation rates obtained with 3 H and 14 C labelled methane respectively, report similar rates with the two tracers in systems with rates in the range observed in Golfo Dulce (Mau et al ., 2013 ; Pack et al ., 2015 ), arguing against such a bias. In addition to the inter‐annual variability, the activity exhibited some variability within years, despite relatively stable biogeochemical conditions (Fig. 3B, F, J ). We see no clear explanation for this variability, which suggests that unresolved chemical, physical or biological dynamics play a role in modulating methane oxidation rates. Overall, rate constants from the Golfo Dulce OMZ (1–57 yr −1 ) are similar to estimates from vent/seep influenced waters (≤73 yr −1 ; Chan et al ., 2019 , ≤25 yr −1 ; Heintz et al ., 2012 ) and coastal systems (1–31 yr −1 , Steinle et al ., 2017 ), which are environments characterized by high methane turnover rates. Values were 1–2 orders of magnitude higher than those estimated for the open ocean ETNP OMZ (0.2–3.3 yr −1 ; Pack et al ., 2015 , ~1 yr −1 ; Thamdrup et al ., 2019 ). This observation is similar to relative differences in rates observed between the coastal OMZ of Golfo Dulce and the oceanic OMZs for the anaerobic processes denitrification and anammox (Dalsgaard et al ., 2003 ; Bulow et al ., 2010 ; Dalsgaard et al ., 2012 ) and likely reflects the overall higher areal productivity of coastal zones (Sigman and Hain, 2012 ) and the resulting higher input of substrates that sustain efficient elemental cycling relative to the open ocean (Kalvelage et al ., 2013 ). The high rate constants of anaerobic methane oxidation suggest this process could be a substantial methane sink in Golfo Dulce. An additional sink is mixing, which has been suggested to be driven primarily by the horizontal advective exchange of water that occurs due to the inflow of intermediate Pacific water (Richards et al ., 1971 ; Ferdelman et al ., 2006 ). Ferdelman et al . ( 2006 ) using a salt balance model estimated the replacement time of water in the deep basin of Golfo Dulce to be 1–2 months. Averaging measured rate constants over anoxic depths, we estimate a turnover time for methane in the OMZ core of ~12 days in 2018 and between 1 and 4 months in 2019–2020, indicating that anaerobic methane oxidation and mixing serve as near equal methane sinks in the OMZ. Our kinetics experiment showed that anaerobic methane oxidation in Golfo Dulce followed first‐order reaction kinetics (Fig. 4 ), which is generally assumed to hold for methane oxidation in marine waters (Reeburgh et al ., 1991 ; Valentine et al ., 2001 ; Thamdrup et al ., 2019 ). First‐order kinetics allowed us to use rate constants and in situ methane concentrations to calculate methane oxidation rates. The steep increase in methane concentration with depth in the OMZ resulted in a similar trend in anaerobic methane oxidation rates, which were typically highest near the sediment–water interface (Fig. 3C, G, K ). The rates measured in the OMZ spanned a wide range (0.02–243 nmol L −1 d −1 ) but nonetheless fell in the upper half of an even broader set of aerobic methane oxidation rates estimated from various marine waters (10 −5 to 10 3 nmol L −1 d −1 ; Mau et al ., 2013 ), where the variability is largely attributed to variations in methane concentrations. Thus, anaerobic methane oxidation appears at least as efficient as the aerobic equivalent. Methylococcales as the dominant methane oxidizers Members of Methylococcales were the only known methane oxidizers recovered from the Golfo Dulce water column in all years sampled (Fig. 5 ). Relative abundances of this group peaked each year (6%–11% of amplicons) below the oxic–anoxic interface between 115 and 130 m (Fig. 3D, H, L ). In 2018 we observed a second peak in relative abundance at 180 m (9%), which could potentially support the particularly high methane oxidation activity observed at that depth in 2018, although relative abundances cannot be translated directly to absolute abundances. In 2019 and 2020, the highest proportional representation of Methylococcales (115–130 m, 6%–9%) roughly corresponded to the primary peak in rate constants (110–130 m, 5–8 yr −1 ; Fig. 3 ). In contrast, the secondary peak in methane oxidation activity (3–8 yr −1 ) at 180–190 m depth was not reflected in Methylococcales abundance, suggesting that other taxa may have contributed to methane oxidation at that depth in 2019/2020. The high relative abundance of Methylococcales and their dominance of the aerobic methanotroph community in Golfo Dulce align with work from the same site in 2015 by Padilla et al . ( 2017 ). In contrast, in 2015 members of the nitrite‐dependent NC10 clade were also detected in Golfo Dulce through qPCR (up to 0.006% of total 16S rRNA sequences; Padilla et al ., 2016 ) and may have escaped detection in this study due to low abundance. We did not identify any other groups of methanotrophs in our 16S rRNA gene datasets, but the possibility for overlooked players cannot be ruled out. The Methylococcales community each year was dominated by a single ASV from the OPU3 clade (Fig. 5 ). The three sequences were highly similar between years and closely related to the OTU (GD_7) that dominated methanotroph sequences in Golfo Dulce in 2015 (Padilla et al ., 2017 ). Although these members of OPU3 typically peaked in abundance between 5 and 25 m below the oxic–anoxic interface, they had a high proportional representation (50%–100% of methanotrophs) across all depths and on both sides of the interface. In contrast, the distribution of OPU1 and Deep Sea‐1 ASVs showed redox driven niche separation, where OPU1 was confined to samples from the oxycline, while members of Deep Sea‐1 only appeared in samples from anoxic depths (with one exception; Fig. 5 ). Similar redox driven niche separation of Gammaproteobacterial methanotrophs has been observed in stratified lakes (Mayr et al ., 2020 ). A niche separation between the two OPU groups is consistent with earlier observations of the two clades in the region. In the OMZ off the Costa Rican coast, OPU3 was observed to primarily inhabit the centre and edges of the OMZ, while OPU1 dominated in the oxycline below the OMZ (Tavormina et al ., 2013 ). Previously in Golfo Dulce, OPU3 peaked in abundance inside the OMZ while OPU1 was only present in the oxycline (Padilla et al ., 2017 ). Members of Deep Sea‐1 are commonly found in methane seep‐influenced surface sediments and as endosymbionts in seep‐associated fauna (Redmond et al ., 2010 ; Raggi et al ., 2013 ; Tavormina et al ., 2015 ), but have also been observed in OMZ waters of the Eastern Tropical Pacific (Hayashi et al ., 2007 ) and in the anoxic water column of the Black Sea (Glaubitz et al ., 2010 ). Our molecular and experimental evidence point to Methylococcales as the main, if not only, conveyors of anaerobic methane oxidation in the Golfo Dulce OMZ. This aligns with evidence from other anoxic aquatic settings, including transcriptional activity of OPU3 members in the anoxic core of the ETNP OMZ (Thamdrup et al ., 2019 ) and repeated observations of Gammaproteobacterial methanotrophs in deep anoxic lakes (Blees et al ., 2014 ; Oswald et al ., 2016 ; Naqvi et al ., 2018 ). The chemical zonation indicates nitrate and nitrite as the most favourable electron acceptors in the OMZ (Fig. 2 ). Indeed, Padilla et al . ( 2017 ) showed that the dominating OPU3 OTU in Golfo Dulce in 2015, GD_7, contained and transcribed genes for dissimilatory nitrate and nitrite reduction to NO and suggested that this group could link partial denitrification to methane oxidation (Padilla et al ., 2017 ), an ability that the closely related OPU3 members identified in 2018–2020 likely share with GD_7. A coupling between methane oxidation and denitrification (to N 2 O) has been demonstrated experimentally and genetically for the strain Methylomonas denitrificans of the Methylococcales (Kits et al ., 2015 ). However, a means for members of Methylococcales, including the OPU3 clade and Methylomonas denitrificans , to circumvent the need for oxygen for the initial methane activation step by the oxygen‐requiring particulate methane monooxygenase ( pmo ) has not been identified. While Padilla et al . ( 2017 ) could not exclude the possibility that the OPU3 population was sustained to at least some degree via oxygen through a fluctuating oxic–anoxic interface or transient inputs, our finer spatial resolution, demonstration of long‐term stability, and rate measurements all point to an anaerobic lifestyle. The detection of the Deep Sea‐1 clade at anoxic depths in the Golfo Dulce (Fig. 5 ) suggests that this group may be even less likely to rely on oxygen, although the ability for anaerobic methane metabolism has not been investigated in this clade. It thus seems likely that these groups possess the ability for anaerobic methane oxidation, and further investigation of the genetic potential of these clades should have a high priority. Taken together, our work indicates that the Golfo Dulce OMZ contains a stable redox zonation with high concentrations of methane compared to open‐ocean OMZs, and an active methane cycle that includes high rates of methane oxidation and, potentially, also pelagic methanogenesis. Based on the consistent steep increase in methane oxidation rate constants below the oxic–anoxic interface measured from incubations containing less than 100 nmol L −1 oxygen, our results strongly indicate that methane oxidation in the OMZ is anaerobic. Our data also suggest that methane oxidation is primarily conducted by specific members of the OPU3 and Deep Sea‐1 clades of the Methylococcales, with the former being present at particularly high relative abundances, though the exact pathway for anaerobic methane oxidation in the two clades remains to be described. Contributions from other unidentified taxa are also possible, including the NC10 clade reported in a previous investigation (Padilla et al ., 2016 ), although this clade seems to be in too low abundance to explain the high rates seen in bottom waters, where the relative abundance of Methylococcales is comparatively low. Thus, from our 3 years of investigation, we hypothesize that members of the OPU3 and Deep Sea‐1 clades are adapted to the stable, anoxic, methane‐rich environment in the OMZ, and substantially reduce the methane flux from the OMZ."
} | 7,885 |
26539175 | PMC4609892 | pmc | 8,192 | {
"abstract": "The minimal cell concept represents a pragmatic approach to the question of how few genes are required to run a cell. This is a helpful way to build a parts-list, and has been more successful than attempts to deduce a minimal gene set for life by inferring the gene repertoire of the last universal common ancestor, as few genes trace back to this hypothetical ancestral state. However, the study of minimal cellular systems is the study of biological outliers where, by practical necessity, coevolutionary interactions are minimized or ignored. In this paper, we consider the biological context from which minimal genomes have been removed. For instance, some of the most reduced genomes are from endosymbionts and are the result of coevolutionary interactions with a host; few such organisms are “free-living.” As few, if any, biological systems exist in complete isolation, we expect that, as with modern life, early biological systems were part of an ecosystem, replete with organismal interactions. We favor refocusing discussions of the evolution of cellular systems on processes rather than gene counts. We therefore draw a distinction between a pragmatic minimal cell (an interesting engineering problem), a distributed genome (a system resulting from an evolutionary transition involving more than one cell) and the looser coevolutionary interactions that are ubiquitous in ecosystems. Finally, we consider the distributed genome and coevolutionary interactions between genomic entities in the context of early evolution.",
"introduction": "Introduction The minimal genome concept is a theoretical idea that has been considered in two different arenas. One is synthetic biology, an area of biological engineering where there is interest in establishing the minimal machinery for a cell ( Peterson and Fraser, 2001 ; Dewall and Cheng, 2011 ; Juhas et al., 2011 ; Acevedo-Rocha et al., 2013 ). The other is in cell origins, where there has been interest in establishing the nature of early cellular systems ( Mushegian, 1999 ; Koonin, 2003 ). Synthetic biology has made strong technical advances, with key proof-of-principle results such as systematically screening for essential genes ( Hutchison et al., 1999 ; Glass et al., 2006 ), synthetic genome assembly ( Gibson et al., 2008 ), transformation of a cell with a chemical genome ( Gibson et al., 2010 ), and the development of computational cellular models derived from genomic knowledge ( Karr et al., 2012 ). The synthetic biology approach to minimal cells is pragmatic, though of unclear value to evolutionary questions pertaining to cellular origins. As an engineering project it is very productive and has a straightforward definition: it is simply the microbe with the smallest genome that is able to grow in axenic culture. There may of course be numerous minimal genomes for growth media, and hence there need not be a definitive minimal genome ( Smalley et al., 2003 ; Dewall and Cheng, 2011 ; Juhas et al., 2014 ). On the criterion of growth in axenic culture, an organism can be designated “free-living” if it can be cultured in the absence of other organisms. Through this criterion, Mycoplasma genitalium is a good candidate for a minimal genome ( Hutchison et al., 1999 ), though growing it in this way has the effect of removing it completely from its natural context, where, as an obligate intracellular pathogen, it is not in the least bit free-living ( Dewall and Cheng, 2011 ). In contrast, the use of comparative genomics to reconstruct the hypothetical last universal common ancestor (LUCA) has been less successful. A range of studies indicate that few genes are common to all three domains of life (Bacteria, Archaea, and Eukaryotes), and fewer still can be said to have an evolutionary history consistent with placement in some hypothetical common ancestor ( Harris et al., 2003 ; Koonin, 2003 ; Hoeppner et al., 2012 ; Goldman et al., 2013 ). Horizontal gene transfer ( Koonin, 2003 ), secondary gene losses ( Becerra et al., 1997 ), and loss of evolutionary signal ( Penny and Poole, 1999 ; Penny and Zhong, 2014 ) all obscure or erase early evolutionary history. It is not clear how to establish which, if any, of the many processes for extracting a living from the environment is the most ancient (though opinions abound), and it seems there is little to be gained from revisiting this question with ever larger genomic datasets. Moreover, there is no reason to expect that the LUCA was in any way a minimal cell, and it is difficult to equate the two, other than to assess the core of processes common to known biological systems ( Goldman et al., 2013 ). The common ground between these efforts is that both the LUCA and the minimal cell concept focus on the internal parts-list of the genome: the gene set. The point of this piece is to begin thinking about early evolution against the backdrop of biological interactions. We think that this is helpful for several reasons. First, systems that exist in isolation are probably the exception rather than the rule. Second, evolutionary transitions theory has provided the means by which to understand the emergence of complex systems, from replicators to cells to eukaryote cells with organelles, and is prefaced on interactions. As a way to navigate this topic, we briefly summarize three concepts: 1. The pragmatic minimal genome concept 2. The evolutionarily stable distributed genome 3. Coevolutionary ecosystem interactions We will then consider how biological interactions may help inform our understanding of early evolution."
} | 1,390 |
35502577 | PMC9249325 | pmc | 8,193 | {
"abstract": "Summary Pseudomonads play crucial roles in plant growth promotion and control of plant diseases. However, under natural conditions, other microorganisms competing for the same nutrient resources in the rhizosphere may exert negative control over their phytobeneficial characteristics. We assessed the expression of phytobeneficial genes involved in biocontrol, biostimulation and iron regulation such as, phlD , hcnA , acdS , and iron‐small regulatory RNAs prrF1 and prrF2 in Pseudomonas brassicacearum co‐cultivated with three phytopathogenic fungi, and two rhizobacteria in the presence or absence of Brassica napus , and in relation to iron availability. We found that the antifungal activity of P. brassicacearum depends mostly on the production of DAPG and not on HCN whose production is suppressed by fungi. We have also shown that the two‐competing bacterial strains modulate the plant growth promotion activity of P. brassicacearum by modifying the expression of phlD , hcnA and acdS according to iron availability. Overall, it allows us to better understand the complexity of the multiple molecular dialogues that take place underground between microorganisms and between plants and its rhizosphere microbiota and to show that synergy in favour of phytobeneficial gene expression may exist between different bacterial species.",
"conclusion": "Conclusions Our results suggest that it is difficult to predict the behaviour of an introduced strain, since the expression of its phytobeneficial traits will depend on several factors, including pedoclimatic factors not addressed here, interaction with other microorganisms, the plant and availability of nutritive resources including essential elements such as iron. The expression of antifungal genes was modulated by interspecies interactions, confirming that rhizosphere bacteria can compete with each other through a range of antibacterial agents and quorum‐sensing (as well as presumably unknown signalling molecules). The mechanisms by which one bacterial population interferes with the gene expression of another population are not yet fully understood. Interkingdom and interspecies interactions, as well as external factors, may affect the success of introduced beneficial microorganisms. Future research must continue to characterize the mechanisms by which these interactions occur and evolve in the rhizosphere. The introduction of a bacterial strain into an ecological niche may have an impact on the assembly of the microbiota and in particular on the interaction networks within the microbial community. These interactions can be synergistic by acting positively directly or indirectly on the expression of genes of inoculated strain or conversely prevent the inoculated strain (biointrant) to express its phytobeneficial effect. In conclusion, the successful development of smart biocontrol not only requires that bacteria carry the appropriate genes, but also that they are expressed in a coordinated and dynamic manner.",
"introduction": "Introduction The microbiota is thought to provide the host with valuable abilities that influence its physiology and improve its fitness (Zilber‐Rosenberg and Rosenberg, 2008 ). The rhizosphere microbiota may play an important role in plant nutrition and protection against pathogens. Many bacteria are able to lower the level of the ethylene precursor, 1‐aminocyclopropane‐1‐carboxylate (ACC) and consequently the production of ethylene by plants, thanks to the bacterial enzyme ACC deaminase (Glick, 2014 ). Bacteria producing acdS are capable of increasing plant tolerance to salinity (Heydarian et al ., 2021 ) and to improve plant root growth (Penrose et al ., 2001 ). Many fluorescent pseudomonads with biocontrol abilities protect plants from soil‐borne diseases by producing antimicrobial secondary metabolites such as hydrogen cyanide (HCN) and 2,4‐diacetylphloroglucinol (DAPG), which are important biocontrol determinants (Vincent et al ., 1991 ; Fenton et al ., 1992 ; Keel et al ., 1992 ; Haas and Défago, 2005 ). The phl cluster that encodes DAPG synthesis is involved in the biocontrol of a broad spectrum of diseases by many antagonistic bacterial strains (Haas and Défago, 2005 ). The involvement of the phl cluster in biocontrol has been determined from studies of root‐colonizing Pseudomonas in disease‐suppressive and conducive soils (Raaijmakers and Weller, 1998 ; Weller et al ., 2002 ). The presence of DAPG‐producing strains does not necessarily guarantee disease suppression, as these strains have also been found in conducive soils (Ramette et al ., 2003 ; Almario et al ., 2013 ). Indeed, Rezzonico et al . ( 2007 ) reported a lack of correlation between the amounts of DAPG produced in vitro by bacteria having the phl cluster and their biocontrol efficacy in planta . The reason for the discrepancy between these studies has not yet been established. Several studies have reported the benefit of introducing particular bacterial strains with in vitro antifungal activity in field conditions, as a strategy for disease control (Nelson, 2004 ). Effective application of this approach in the field requires knowledge of the biotic and abiotic factors that influence the functions of the introduced strain, especially under natural conditions. Biocontrol is the exploitation of disease‐suppressive microorganisms to improve plant protection (Oconnell et al ., 1996 ). From this perspective, disease suppression through biocontrol involves plant–bacteria, bacteria–phytopathogen and bacteria–bacteria interactions, as well as interactions with their physicochemical environment. Although there is an increasing demand for biocontrol in the context of sustainable agriculture, biocontrol in field conditions still faces challenges before its practices can be widely accepted and optimally used (Meyer and Roberts, 2002 ; Bashan et al ., 2014 ). The success of these biocontrol assays also depends on field conditions, where inter‐ and intra‐species interactions can also impact the type and concentration of compounds produced by a microorganism (van Agtmaal et al ., 2018 ). The physicochemical soil properties and the interactions of biocontrol agents with root‐colonizing microbiota may explain why many microorganisms suppressed diseases successfully under laboratory conditions, but failed in the field. Introduced biocontrol agents can be affected by other microbial communities, as well as inducing a change in the assembly of the microbiota leading to changes in interactions within the microbial community and, in some cases, they can work in synergy to suppress plant diseases. In other cases, multiple partners interact to regulate a single phytobeneficial trait, so that these complex relationships can have important ecological consequences (Hussa and Goodrich‐Blair, 2013 ). Abiotic factors may also be involved in the establishment of inoculated PGPR strains and in the expression of their phytobeneficial genes (Lim et al ., 2012 ). Almario et al . ( 2013 ) hypothesized that the clay mineral composition of soils may impact on iron availability, which may confer disease suppressiveness in the rhizosphere, allowing the expression of biocontrol relevant genes in antagonistic Pseudomonas protegens . Few studies, under controlled conditions, have reported interspecies interactions that could be based on quorum sensing that coordinates interactions both within a species and between species (Abisado et al ., 2018 ), or in particular on competition for iron that plays a central role in microbe–microbe and microbe–host interaction. Ho et al . ( 2021 ) studied the interaction between pyochelin‐producing bacteria and the plant pathogen Phellinus noxius . P. noxius converts pyochelin and ent‐pyochelin from Pseudomonas and Burkholderia species to pyochelin‐GA (and ent‐pyochelin‐GA), impairing their antifungal and iron chelation activities (Ho et al ., 2021 ). Interspecies interactions occur between microorganisms occupying the same ecological niche, such as Pseudomonas aeruginosa and members of the Burkholderia cepacia complex, which often coexist in both the soil and the lungs of cystic fibrosis patients. Weaver and Kolter ( 2004 ) showed that ornibactin, a siderophore produced by nearly all B. cepacia strains, can induce the expression of P. aeruginosa PA4467 gene, indicating that ornibactin can be produced by B. cepacia and detected by P. aeruginosa when the two species coexist. \n Pseudomonas brassicacearum was described as the major root‐associated bacterium in the rhizosphere of Arabidopsis thaliana and Brassica napus (Achouak et al ., 2000 ; Fromin et al ., 2001 ) and displayed biological control, plant growth promotion and for some strains pathogenic traits (Belimov et al ., 2007 ). Mutations in the gacS‐gacA system have been shown to lead to drastic pleiotropic changes in P. brassicacearum (Lalaouna et al ., 2012 ). The expression of secondary metabolites ( e.g . the antifungal compounds DAPG and hydrogen cyanide), auxin, exoenzymes ( e.g . lipase and protease), three different N ‐acyl‐homoserine lactone molecules, the type VI secretion machinery and alginate synthesis was downregulated in variants with mutations in gacS or gacA genes, and biofilm formation ability was greatly reduced (Lalaouna et al ., 2012 ). This indicates that the expression of the phl and hcn genes is under the positive control of the GacS/GacA system through the synthesis of rsmX‐1 , rsmX‐2 , rsmY and rsmZ (Lalaouna et al ., 2021 ). To study in situ gene expression of phytobeneficial bacteria, Haichar et al . ( 2013 ) investigated the expression of phlD by developing mRNA‐stable isotope probing (mRNA‐SIP). Their results have demonstrated that the phlD gene was expressed by bacteria inhabiting the rhizosphere soil that derive nutrients from the breakdown of organic matter and root exudates, whereas phlD gene expression appeared to be repressed on A. thaliana roots. In vitro expression of the phlD gene in P. brassicacearum was strongly activated by wheat and Medicago truncatula and to a lesser extent by B. napus , whereas it was downregulated by A. thaliana (Haichar et al ., 2013 ). This regulation is probably mediated by plant root exudates (Haichar et al ., 2013 ). In addition to other microorganisms, the plant can also be involved in modulating the expression of phytobeneficial genes. Little is yet known about the activity of the genes underpinning biocontrol and biostimulation under conditions of competition for scarce nutrients and essential elements such as iron, the lack of which has major consequences for the nature of interactions between organisms in the three kingdoms of life. The objective of this work was therefore to better understand the behaviour of biocontrol agents by assessing the impact of biological factors on the expression of the phlD , hcnA and acdS genes, which encode two major biocontrol and a biostimulation traits. Here, we investigated the ability of the plant root‐associated P. brassicacearum NFM421 to compete with two other rhizobacteria, Kosakonia sacchari NO9, a root‐associated diazotroph (Bloch et al ., 2020 ) and Rhizobium alamii YAS34, known for its ability to produce exopolysaccharide and to contribute to soil stability in the rhizosphere of Helianthus annuus and B. napus (Alami et al ., 2000 ; Tulumello et al ., 2021 ), under conditions of iron depletion and iron repletion, and in the presence (or not) of B. napus . Under these different experimental conditions, we analysed the expression of the phytobeneficial genes phlD , hcnA and acdS and to assess the intracellular status of iron in P . brassicacearum NFM421, we analysed the expression of iron‐regulatory RNAs prrF in this strain. We also tested the antagonism of P. brassicacearum NFM421 wt and the knockout mutants ∆ phlD and ∆ gacA against the soil‐borne plant pathogens Fusarium culmorum , Fusarium graminearum and Microdochium nivale , to compare the antifungal activity of DAPG and HCN.",
"discussion": "Discussion Although certain pseudomonads play a crucial role in promoting plant growth and controlling plant diseases, other living organisms such as bacteria, fungi and plants may interfere with their biocontrol efficiency. Most fluorescent pseudomonads promote plant growth and produce antimicrobial secondary metabolites including hydrogen hydrogen cyanide (HCN) and 2,4‐diacetylphloroglucinol (DAPG). Here, we assessed the expression of hcnA , phlD and acdS in P. brassicacearum during co‐culture with phytopathogenic fungi and during co‐culture with non‐pathogenic plant root‐associated bacteria. This work demonstrates the modulation of phytobeneficial genes by other microorganisms, depending on the presence of competing bacteria as well as plant and iron availability. Host colonization by biocontrol and biostimulant agents is essential for biocontrol but not sufficient for plant protection against pathogens and plant growth promotion. In addition to competing for C and energy resources in the rhizosphere and of the rhizoplane sites to be colonized, the biocontrol agent must be able to express antimicrobial genes and produce the corresponding compounds. Genes that are modulated during interspecific bacterial interactions often include genes involved in the production of antibiotics, as a defensive or offensive strategy in microbial interactions (Fong et al ., 2001 ; Harrison et al ., 2008 ; Garbeva and de Boer, 2009 ). Rhizosphere‐associated bacteria are always encountered as mixed populations of numerous species in the environment, and they have to cope with other microorganisms, the host and environmental cues. These associations are the result of coevolution leading to the implementation of strategies for adaptation to specific ecological niches. Antifungal activity of P. brassicacearum is DAPG‐dependent When iron is scarce, bacteria produce siderophores that may act as antifungals by depriving fungi of iron. In the presence of iron, other antifungal metabolites intervene like DAPG and phenazines among others. Interaction assays of P. brassicacearum NFM421 with Fusarium head blight fungi, carried out without addition of iron to determine the role of DAPG and HCN, clearly show the role of DAPG in inhibiting the growth of the tested fungi. Using the two mutants, ∆ gacA and ∆ phlD , we demonstrated that DAPG, but not HCN, seems to be primarily responsible for the antifungal activity. We observed a repression of hcnA expression in the presence of the three phytopathogenic fungi F. culmorum , F. graminearum and M. nivale , whereas Barret et al . ( 2009 ) reported an increase in the expression of hydrogen cyanide synthesis genes with Pseudomonas fluorescens strain Pf29Arp, in the presence of the phytopathogenic fungus Gaeumannomyces graminis var. tritici . Co‐culture of P. brassicacearum NFM421 with M. nivale species did not alter phlD expression, while we noticed a significant increase in phlD expression in the presence of F. culmorum and F. graminearum , whereas these fungi are known to produce fusaric acid, which has been reported to repress phlD expression (Notz et al ., 2002 ). These bacterium–fungus interactions do not appear to be generalizable, and these interactions may be species‐ or even strain‐dependent. The expression of phlD was one thousand times higher than that of hcnA , indicating that even if HCN is known to be effective against pathogenic fungi, the levels of its expression by P. brassicacearum NFM421 are too low to inhibit fungal growth. Therefore, the mode of biocontrol of plant pathogenic fungi used in this study, which seems plausible in P. brassicacearum NFM421, most likely involves DAPG. \n Pseudomonas brassicacearum outcompetes K. sacchari and R. alamii \n Our results showed the competitiveness of P. brassicacearum NFM421 when grown with K. sacchari NO9 and R. alamii YAS34, under iron‐depleted and iron‐rich conditions via a higher growth rate. The total population size evolved from 3 10 7 cells ml −1 to at least 10 9 cells ml −1 , suggesting that the three bacterial strains could proliferate, and their population size increased at least tenfold. P . brassicacearum produces at least two types of siderophores, which facilitate the recovery of small amounts of iron in aerobic environments (Matthijs et al ., 2016 ), and at least two antimicrobial compounds, DAPG and HCN (Lalaouna et al ., 2012 ). These observations are consistent with previous studies showing that batch cultures of Pseudomonas aeruginosa grown in a defined medium produce one or more secreted factors that are stimulated by iron limitation and inhibits Agrobacterium tumefaciens biofilms (Hibbing and Fuqua, 2012 ). Surprisingly, R. alamii YAS34 was more efficient in planta at competing with P. brassicacearum NFM421 under iron‐depleted conditions than under iron‐rich conditions, where they reached up to 18 and 4% of the total population, respectively. This finding suggests that the growth rate of P . brassicacearum NFM421 is probably much higher at the expense of root exudates and under iron‐rich conditions. Expression of phlD , hcnA and acdS is not exclusively controlled by iron Both competitors ( K. sacchari NO9 and R . alamii YAS34) mostly downregulated phlD expression under iron‐rich conditions when co‐cultured with P . brassicacearum . These results suggest that these competing bacterial strains may produce specific signals that are perceived by P. brassicacearum , which act directly or indirectly at the transcriptional or post‐transcriptional level. They may produce certain metabolites that might activate the repressor phlF . Bacterial extracellular metabolites such as salicylate and pyoluteorin as well as fusaric acid, a toxin produced by the pythopathogen fungus Fusarium , strongly repressed DAPG synthesis in Pseudomonas protegens strain CHA0 (Schnider‐Keel et al ., 2000 ). The expression of phlD and hcnA genes was remarkably increased by the addition of iron, unlike P. fluorescens Pf‐5 in which Lim et al . ( 2012 ) reported overexpression of DAPG‐encoding genes and downregulation of hydrogen cyanide‐encoding genes under iron‐depleted conditions. In contrast to the interaction with fungi, competition with bacteria led to a significant decrease in phlD expression in the presence of iron and conversely a significant increase in hcnA expression. Analysis of the prrf sRNAs expression indicated that it was likely that the presence of competitors allowed P . brassicacearum NFM421 to increase its intracellular iron level. It is difficult to imagine how Fur could be activated other than by complexing iron, as it is the only regulator known to repress the expression of the small prrF regulatory RNAs in the presence of iron (Wilderman et al ., 2004 ; Liu et al ., 2016 ). Even genes that are finely regulated seemed to be modulated by the co‐culture of bacteria. In addition to the indirect competition for iron involving siderophores, it has been reported that P. aeruginosa was able to kill Staphylococcus aureus and utilized the released iron (Mashburn et al ., 2005 ). Competitive behaviours between bacteria are partly controlled by iron availability (Andrews et al ., 2003 ). Numerous mechanisms may be involved in bacterial exchange, such as secondary metabolites and quorum‐sensing quenching system. Bacteria are known to communicate with each other and modulate their gene expression through quorum‐sensing signalling molecules (Lowery et al ., 2008 ). Contrary to phlD , whose expression is tenfold higher in CAA medium enriched with iron than in planta , acdS expression is tenfold higher in planta than CAA medium enriched with iron (Fig. 3A and C vs Fig. 4A and C ). The production of ACC by the plant must probably activate the expression of acdS , as indicated by genes transcript level normalized to 16S rRNA transcription level. We previously reported the impact of the plant via root exudates on the expression of the non‐coding RNAs rsmZ , acdS gene encoding 1‐aminocyclopropane‐1‐carboxylate (ACC) deaminase and nosZ gene encoding nitrous oxide reductase, and evidenced their expression in the root‐adhering soil and on the roots of A. thaliana by using mRNA‐SIP approach (Haichar et al ., 2012 ). However, phlD gene expression was shown to be highly activated by root exudates of wheat and that of Medicago truncatula and to a lesser extent by that of B. napus , while it was strongly suppressed by root exudates of A. thaliana suggesting that the signals for downregulation of phlD gene expression may originate from B. napus and A. thaliana root exudates (Haichar et al ., 2013 ). We observed no effect of iron on phlD or hcnA expression, with the exception of co‐inoculation with K. sacchari NO9 which induced a strong increase in phlD and hcnA expression in the absence of iron and a significant decrease of phlD in the presence of iron. The triple inoculation also led to an increase of phlD expression in the absence of iron. During in planta interspecies interaction, P. brassicacearum NFM421 antifungal genes were positively modulated by K. sacchari NO9 denoting a synergy between these two strains in favour of the use of P. brassicacearum NFM421 as a biocontrol agent, in combination with K. sacchari NO9. However, acdS expression was reduced in P. brassicacearum NFM421 grown alone on CAA or in monoxenic condition in planta , whereas the presence of K. sacchari NO9 seems to induce an increase in acdS expression, in planta under iron‐depleted condition. This increase in acdS expression did not support the need for increased ethylene production by the plant under Fe deficiency, which, in the roots of several plant species, led increased ethylene production contributing to the regulation of subapical root hairs and Fe acquisition genes (Angulo et al ., 2021 ). Our results show an in planta modulation of the expression of phytobeneficial genes in P. brassicacearum NFM421, essentially in the presence of K. sacchari NO9 suggesting a signalling exchange between these strains on the plant root surface. For this reason, we colocalized the competing strains in planta . Colocalization imagery showed that K. sacchari NO9 strain occupied the same sites as P . brassicacearum NFM421 on the rhizoplane, as we consistently observed their colocalization (Fig. 6B ). Figure 6B shows a strong colonization by the NO9 strain at the apical part of the roots; however, they are less abundant at the basal part (Fig. S1 ), while R . alamii YAS34 did not exhibit a significant impact on the expression of antifungal genes, probably because they inhabited niches on the roots that did not overlap with those of P . brassicacearum NFM421 (Fig. 6A ). It is likely that inhabiting separate niches may lead to a weaker interaction and exchange. The type and quantity of compounds produced by one bacterial strain can probably vary when it interacts directly with another, especially when they compete for space and resources in the same ecological niche. It is this proximity and exchange that should shape microbial assemblages and interaction networks within microbial communities. The interactions, we observed in a very simplified in vitro system with only three strains, hint at the complexity of what could happen in situ in real soil with thousands of bacterial species. Scheme for multiple dialogues that may modulate phytobeneficial gene expression To illustrate our results, we propose a scheme targeting the modulation of phl gene expression by other microorganisms and the plant, as well as the role of iron and its regulation by prrF regulatory non‐coding RNAs (Fig. 7 ). The expression of hcnA and acdS also seems to depend on the biotic and abiotic environment of the bacterium. This scheme of interactions in a simplified in vitro system of bacteria with their biotic and abiotic environment and their potential impact on the expression of phytobeneficial traits hints at the complexity of these relationships under natural conditions. Fig. 7 Scheme for multiple dialogues that may modulate phytobeneficial gene expression. This model illustrates the modulation of phl gene expression by other microorganisms, the plant and iron, and the role of iron and its regulation by prrF regulatory non‐coding RNAs. These interactions are most likely even more complex under natural conditions in the rhizosphere in the presence of a very diverse microbiota producing a multitude of metabolites. This is also valid for the other genes hcnA and acdS . ↓, positive effect; ┴, negative effect; dotted lines, indirect effects. The conclusion of this study is that the regulation of phytobeneficial gene expression is under the control of different biotic and abiotic factors."
} | 6,274 |
37746168 | PMC10512219 | pmc | 8,195 | {
"abstract": "Plant communities and fungi inhabiting their phyllospheres change along precipitation gradients and often respond to changes in land use. Many studies have focused on the changes in foliar fungal communities on specific plant species, however, few have addressed the association between whole plant communities and their phyllosphere fungi. We sampled plant communities and associated phyllosphere fungal communities in native prairie remnants and post-agricultural sites across the steep precipitation gradient in the central plains in Kansas, USA. Plant community cover data and MiSeq ITS2 metabarcode data of the phyllosphere fungal communities indicated that both plant and fungal community composition respond strongly to mean annual precipitation (MAP), but less so to land use (native prairie remnants vs. post-agricultural sites). However, plant and fungal diversity were greater in the native remnant prairies than in post-agricultural sites. Overall, both plant and fungal diversity increased with MAP and the communities in the arid and mesic parts of the gradient were distinct. Analyses of the linkages between plant and fungal communities (Mantel and Procrustes tests) identified strong correlations between the composition of the two. However, despite the strong correlations, regression models with plant richness, diversity, or composition (ordination axis scores) and land use as explanatory variables for fungal diversity and evenness did not improve the models compared to those with precipitation and land use (ΔAIC < 2), even though the explanatory power of some plant variables was greater than that of MAP as measured by R 2 . Indicator taxon analyses suggest that grass species are the primary taxa that differ in the plant communities. Similar analyses of the phyllosphere fungi indicated that many plant pathogens are disproportionately abundant either in the arid or mesic environments. Although decoupling the drivers of fungal communities and their composition – whether abiotic or host-dependent – remains a challenge, our study highlights the distinct community responses to precipitation and the tight tracking of the plant communities by their associated fungal symbionts.",
"introduction": "Introduction Aerial plant photosynthetic tissues – the phyllosphere – are among the most extensive microbial habitats on Earth ( Morris et al., 2002 ). This habitat can be oligotrophic and exposed to rapid fluctuations in environmental conditions including shifts in temperature, humidity, and radiation ( Lindow and Brandl, 2003 ). Yet, the phyllosphere represents a diverse ecosystem ( Lindow and Leveau, 2002 ), colonized by hyperdiverse communities of bacteria, archaea, virus, protists, and fungi all living on (epiphytes) and within (endophytes) the leaves ( Jumpponen and Jones, 2009 ; Martiny et al., 2011 ; Vorholt, 2012 ; Laforest-Lapointe & Whitaker, 2019 ). These diverse communities drive ecosystem function ( Song et al., 2017 ; Laforest-Lapointe and Whitaker, 2019 ) and can contribute to nitrogen cycling by fixing nitrogen in situ ( Furnkranz et al., 2008 ). Phyllosphere communities can also affect plant fitness and productivity ( Davison, 1988 ; Schauer and Kutchera, 2011 ) through their modulation of stress tolerance ( Vorholt, 2012 ) or pathogen resistance ( Innerebner et al., 2011 ). Further, the phyllosphere communities may drive plant community dynamics ( Aschehough et al., 2014 ; Whitaker et al., 2017 ; Laforest-Lapointe and Whitaker, 2019 ) thereby linking the phyllosphere communities to plant communities and their productivity ( Laforest-Lapointe et al., 2017 ). Phyllospheres are clearly important for ecosystem function and as a hotspot for microbial diversity ( Arnold and Lutzoni, 2007 ; Laforest-Lapointe et al., 2017 ). Foliar fungi are among the most diverse members that can impact plant productivity and physiology within the phyllosphere ( Saikkonen et al., 1998 ; Rodriguez et al., 2009 ; Meyer and Leveau, 2012 ; Zahn and Amend, 2019 ). These fungi presumably occupy photosynthetic tissues of all species and in all divisions of land plants ( Bacon and White, 2000 ). While present in the foliage, these communities include taxa that are directly and functionally associated with the plant tissues ( e.g. , pathogens, foliar parasites or endophytes) as well as those that may be observable on these tissues but neither penetrate the cuticle nor directly functionally interact with the host plant ( i.e. , epiphytes that may utilize nutrients available on the foliar surfaces but never cross the cuticular barrier) (see Gomez et al., 2018 ). The foliar fungal communities may be more sensitive to environmental factors than those of bacteria ( Bernard et al., 2021 ) or ectomycorrhizal fungi ( Bowman and Arnold, 2021 ). A recent study of Hibiscus tiliaceus trees in Hawaii ( Bernard et al., 2021 ) reported that while bacterial community composition was better explained by the plant organ macrohabitat, location within a steep environmental gradient better predicted variation in fungal community composition (see also Zimmerman and Vitousek, 2012 ). Similarly, Bowman and Arnold (2021) concluded that while the distribution of ectomycorrhizal fungi was mainly constrained by dispersal, foliar fungi were more constrained by climate factors such as mean annual precipitation and mean annual temperature. These studies exemplify the value of studying steep environmental gradients as a means to better understand how environmental variation influences the composition and assembly of fungal communities ( Fraser et al., 2015 ; Rudgers et al., 2021 ). In addition to the environment, communities can be impacted by a variety of human factors. The anthropogenic conversion of natural ecosystems presents a substantial threat to biodiversity ( Foley et al., 2005 ; Newbold et al., 2015 ; Perreault and Laforest-Lapointe, 2021 ). Human land-use, including agriculture and silviculture, can have long-lasting legacies wherein the altered ecosystem attributes persist after cessation of human land-use ( Dupouey et al., 2002 ; Foster et al., 2003 ; Flinn et al., 2005 ; McLauchlan, 2006 ; Cramer et al., 2008 ). These systems struggle with the establishment of native plant communities after the human land-use abandonment ( Kuussaari et al., 2009 ; Moreno-Mateos et al., 2017 ). For example, compared to systems that have no history of human use, former agricultural lands may possess altered soils, non-native plant communities, and other distinct ecosystem properties for decades and even millennia following farm abandonment ( Bellemere et al., 2002 ; Dupouey et al., 2002 ; Flinn and Marks, 2007 ). Similarly, agricultural land-use history can reduce soil-inhabiting fungal diversity and result in communities distinct from those in native remnants that have never been used for production agriculture ( Oehl et al., 2003 ; Wagg et al., 2018 ; Turley et al., 2020 ). Phyllosphere communities may be less responsive to edaphic factors as they do not directly interact with the soil matrix, whose biogeochemical attributes may strongly influence soil-inhabiting communities. Consistent with this, community composition of the foliar fungi often reflects climatic factors ( Bowman and Arnold, 2021 ) such as mean annual temperature and precipitation ( Oita et al., 2021 ), rather than variation in soil properties. This is particularly true if phyllosphere communities are assessed broadly and include casual epiphytes that may only utilize readily available resources on the leaf surfaces. Even within the phyllosphere, controls of communities in foliar compartments may differ, as the communities of leaf epiphytes and endophytes may be shaped by distinct environmental controls ( Gomes et al., 2018 ). Although plant and fungal communities and their responses to environmental gradients have been targets of many studies, analyses to better establish linkages among them are still rare. Large-scale studies have reported correlations between plant and fungal richness ( Arnold and Lutzoni, 2007 ; Tedersoo et al., 2020 ) that may often stem from collinearities and/or correlations between plants and associated fungal communities. In this contribution, we attempt to concurrently dissect plant communities as well as those fungal communities that occupy their photosynthetic tissues. Many studies thus far have focused on diversity at the local scales ( Allan et al., 2014 ; Newbold et al., 2015 ) but neglected changes at larger spatial scales. We utilized the steep precipitation gradient in the state of Kansas (USA) located in the Great Plains to assess how plant communities and their foliar fungal communities may respond to this precipitation gradient, how the plant and fungal communities may differ across two distinct historic land uses (post-agricultural sites and native remnant prairies), whether the communities within these two historic land uses respond differently to precipitation, and how the plant communities and their foliar communities may be linked to each other. Agricultural systems that have a history of intensive human land-use are a common focus of restoration efforts but how post-agricultural fields compare to native prairie remnants remains unclear particularly for fungal communities that occupy photosynthetic tissues. We hypothesized that 1) plant and their phyllosphere fungal communities increase in richness, evenness, and diversity with increasing the mean annual precipitation; 2) post-agricultural sites – as a result of their previous intensive agricultural use – have a lower richness and diversity as well as distinct communities when compared to native prairie sites; 3) native prairie remnants and post-agricultural sites differ in their responses to the precipitation gradient such that richness, diversity, and evenness in the remnant prairie sites respond more strongly to precipitation than post-agricultural prairies; and 4) fungal communities correlate with plant communities in diversity and composition. We emphasize that the approaches linking aboveground plant diversity with fungal richness and diversity are rare ( Cho et al., 2017 ) and that studies across land-use systems and plant diversity are required to enable sound recommendations for sustainable land-use ( Monkai et al., 2017 ).",
"discussion": "Discussion We sampled the steep precipitation gradient in the central United States to better understand how plant and fungal communities vary with MAP, among native prairie remnants and post-agricultural sites, and how these communities may be linked. Our data indicate that both plant and fungal communities shift compositionally and increase in their diversity with MAP and had greater diversity in native remnant prairies than in post-agricultural sites. Further, although plant community richness also increased with MAP, fungal community richness did not. This lack of fungal richness response to MAP is surprising, given that the plant and fungal communities were correlated in composition. Although it is impossible to decouple MAP and other potential correlates, our analyses suggest the importance of MAP gradient and land-use history in controlling plant and fungal communities. Our data supported our hypotheses that plant communities change in composition and increase in richness and diversity with MAP. Temperate grasslands in central North America range from 200 to 1200 mm·y –1 in MAP ( Lauenroth et al., 1999 ) resulting in distinct ecosystems ranging from the shortgrass steppes with very low annual net primary productivity to the highly productive tallgrass prairies ( Sala et al., 1988 ; Lauenroth et al., 1999 ). Our study covered a substantial proportion of this gradient (455.7–1040.5 mm yr -1 ) and our results are consistent with the transition from shortgrass steppes and mixed grass prairies to tallgrass prairies along the west-east precipitation gradient. The broad variability in MAP not only affects ecosystem annual net primary productivity, but also plant community composition, cover, and diversity ( Lauenroth et al., 1978 ; Watson et al., 2021 ). Our results are congruent with Watson et al. (2021) and suggest that MAP is an important plant diversity predictor for regionally distinct plant communities. Our indicator taxon analyses highlighted that it is indeed the dominant graminoids that define these grassland communities, particularly so in the mesic tallgrass prairies. Interestingly, our data suggested that floristic quality response to MAP depended on the land-use such that FQI adj increased with MAP in native remnants but not in the post-agricultural fields. When we excluded the potential outliers, these responses became even more obvious and indicated an actual decline in FQIadj with MAP in the post-agricultural fields. The stochastic niche hypothesis ( Tilman, 2004 ) predicts that plant communities with greater species richness would be less subject to establishment of new species – in our case also non-native species – than communities that have low species richness. This resistance to invasion is posited to stem more from resource exhaustion by the large number of potentially competing species with differing niches than from community diversity itself ( McKane et al., 2002 ; Reich et al., 2012 ; Lannes et al., 2020 ). Plant species richness increased with MAP in both native prairie and post-agricultural sites in our analyses. As a result, our FQIadj results in the native prairie remnants seem consistent with this hypothesis but not in the post-agricultural sites. In contrast, in the post-agricultural sites, the decline in the FQI adj in sites with greater species richness suggests that the agricultural land use legacy results in communities that are increasingly of lesser floristic quality and include a greater proportion of non-native species the greater the richness of comparable native sites is. It remains an open question whether the post-agricultural sites differ from the native prairies as a result of differences in available soil resources that reflect the past anthropogenic inputs during row crop production. Plant communities and their shifts along gradients have been extensively studied (see Watson et al., 2021 ), whereas similar studies on fungal communities and/or their diversity are less common (but see e.g. , Tedersoo et al., 2014 ; Glynou et al., 2016 ; Rudgers et al., 2021 ). Factors that may affect fungal communities include latitude ( Arnold et al., 2000 ; Arnold and Lutzoni, 2007 ; Tedersoo et al., 2014 ), climate ( McGuire et al., 2012 ; U’ren et al., 2012 ; Zimmerman and Vitousek, 2012 ; Eusemann et al., 2016 ; Oita et al., 2021 ; Rudgers et al., 2021 ), soil ( Tedersoo et al., 2020 ; Bowman and Arnold, 2021 ; Rudgers et al., 2021 ), plant host ( Hoffman and Arnold, 2008 ; U’ren et al., 2012 ; Lau et al., 2013 ; Kembel and Mueller, 2014 ; Tedersoo et al., 2020 ; Rudgers et al., 2021 ), and disturbance ( Delgado-Baquirizo et al., 2021 ). Some studies highlight strong host species and/or climatic/edaphic effects ( e.g ., Hoffman and Arnold, 2008 ; Tedersoo et al., 2020 ; Rudgers et al., 2021 ), whereas others find no support for correlations between plant community diversity and fungal communities ( e.g. , McGuire et al., 2012 ; Tedersoo et al., 2014 ). While soil- and root-inhabiting fungal communities may be buffered against climatic drivers ( Rudgers et al., 2021 ) or correlate with plant diversity ( Shen et al., 2021 ), phyllosphere communities may be particularly sensitive to climatic drivers whilst buffered against edaphic factors ( Bowman and Arnold, 2021 ; Oita et al., 2021 ). Consistent with our hypotheses and predictions, our data strongly suggest that phyllosphere fungal communities respond to MAP. These conclusions agree with others who have concluded that climatic factors strongly influence the phyllosphere fungal communities and their assembly ( Carroll and Carroll, 1978 ; Zimmerman and Vitousek, 2012 ; U’Ren et al., 2012 ; Oita et al., 2021 ). In addition to environmental factors, fungal communities respond to host species ( Rudgers et al., 2021 ), although not necessarily to plant diversity or richness ( McGuire et al., 2012 ; Tedersoo et al., 2014 , but see Hooper et al., 2000 ; Shen et al., 2021 ). Our data clearly indicate that plant and fungal communities correlate, even though fungal richness neither strongly correlated with MAP nor was well predicted by climatic or plant community variables. Differences in plant metabolities and plant physiology may control phyllosphere community diversity and composition ( Bailey et al., 2005 ; Rajala et al., 2014 ; Eusemann et al., 2016 ), resulting in greater fungal diversity in systems with greater plant diversity. We hypothesize that our observed compositional correlations likely stem from the niche heterogeneity provided by diverse plant communities that then may host diverse and distinct phyllosphere fungi. Indeed, some of our most common fungal indicator taxa were directly linked to their hosts, exemplified by foliar plant pathogens ( e.g ., Phyllosticta sorghina and Blumeria sp.). In sum, as host species communities shift, so does the probability of distinct fungal associates in the phyllosphere. Ranking factors for their importance in structuring fungal communities is not simple. Some studies have suggested that edaphic factors can override the influence of host plant identity ( Glynou et al., 2016 ), whereas others have suggested that the importance of edaphic factors varies among host species ( Rudgers et al., 2021 ). In our study, MAP and plant community composition or diversity are inherently collinear and evaluating their relative importance in phyllosphere community assembly is therefore challenging. The controls may also differ among fungal guilds. McGuire et al. (2012) targeted lowland tropical rain forests with high plant richness in Panama and concluded that the compositionally distinct communities in soil and leaf litter differed in their compositional controls. (2012) targeted lowland tropical rain forests with high plant richness in Panama and concluded that the compositionally distinct fungal communities in soil and leaf litter differed in their compositional controls. Although the former correlated with MAP but not with plant richness, the latter correlated with neither MAP nor plant diversity. Further experiments that manipulated litter richness suggested that plant diversity may be less important in determining fungal richness than MAP as the fungal richness did not track the plant richness. In contrast to those studies, Shen et al. (2021) manipulated herbaceous plant community richness in a greenhouse experiment and concluded that the soil fungal richness correlated with that of the plant communities. Clearly, experimental systems, targeted fungal guilds and included host taxa appear essential controls of fungal communities. To better understand the relative importance of environmental factors and plant community estimators in the current experiment, we compared models using the main and interactive effects of land use and either MAP or plant estimators (richness, diversity, evenness, FQI adj , or PCoA axis 1). These simple model comparisons suggested that MAP is usually a superior predictor for fungal diversity and evenness. Although our studies emphasize the importance of climatic factors (see also Oita et al., 2021 ), further and more detailed studies may be needed to better resolve these issues. Understanding how climatic or edaphic variables can influence host-associated fungal communities is becoming increasingly important as the ongoing environmental change has the potential to disrupt host-microbe interactions ( Ranelli et al., 2015 ; Glynou et al., 2016 ; Vetrovsky et al., 2019 ; Steidinger et al., 2020 ). Analysis of environmental gradients, such as MAP here, is a powerful approach to dissect such patterns ( Rudgers et al., 2021 ). Contrary to our hypotheses and predictions, we observed no strong evidence for differences in community composition and dispersion of plants or their phyllosphere fungi among the post-agricultural fields and native prairie remnants. However, our data indicate that native remnant prairies harbor greater plant richness and diversity as well as greater phyllosphere fungal diversity and evenness. Land-use and particularly its intensification have been posited as major drivers of biodiversity loss ( Sala et al., 2000 ; Foley et al., 2005 ; Gossner et al., 2016 ; Brinkmann et al., 2019 ) and biotic and ecological homogenization ( Gossner et al., 2016 ; Brinkmann et al., 2019 ; Delgado-Baquirizo et al., 2021 ). Some have suggested that the communities in post-agricultural sites remain distinct from those in native sites because of fungal dispersal limitations from native remnants ( Turley et al., 2020 ), as has been reported for plants ( Turley et al., 2017 ). The establishment of fungal communities in post-agricultural sites may also be a result of poor recovery of soil conditions after intensive agriculture ( Bellemare et al., 2002 ; Dupouey et al., 2002 ; Flinn and Marks, 2007 ). Although the phyllosphere fungal communities correlate with phyllosphere chemistry and have been reported to differ among land-use types (e.g., Jumpponen and Jones, 2010 ), they may be less affected directly by the altered post-agricultural soil conditions than the soil- or root-inhabiting fungal communities are. Dispersal limitations for the phyllosphere communities may also be less restrictive than they are for soil-dwelling fungi ( Bowman and Arnold, 2021 ). Our results are congruent with those of many others that emphasize agricultural legacy effects on bacterial and fungal communities decades after agricultural abandonment ( Lauber et al., 2008 ; Upchurch et al., 2008 ; Jangild et al., 2011 ; Hui et al., 2018 ; Turley et al., 2020 ) as well as those that report strong biotic and ecological homogenization by anthropogenic land-use ( e.g. , McKinney and Lockwood, 1999 ; Groffmann et al., 2014 ; Gossner et al., 2016 ; Delgado-Baquirizo et al., 2021 ; Kotze et al., 2021 ). Our data indicate that land-use is an important driver of phyllosphere communities across broad environmental gradients such as the steep precipitation gradient sampled here. Taken together, our study suggests that the phyllosphere communities in these systems closely track plant communities whose diversity has been impacted by the land-use legacies. We simultaneously analyzed plant communities and their phyllosphere fungal communities to assess responses to MAP and land-use history across a precipitation gradient extending much of the known range of the temperate grasslands in the central Great Plains. Our data indicate strong climatic controls of both the plant and phyllosphere fungal communities and the lesser impact of the historic land-uses on community composition. Interestingly, these data highlight the resilience of the species-rich tallgrass prairies and comparatively lesser floristic quality of post-agricultural sites in the more mesic regions of this MAP gradient. The phyllosphere fungal communities also responded strongly to MAP, whereas the historic land-use appeared to have minimal to no effects on the compositionof these communities. However, our data indicate greater plant richness and diversity as well as greater fungal diversity and evenness in native remnant prairies than in post-agricultural sites. Although our model comparisons highlighted that MAP was commonly a stronger predictor of phyllosphere fungal community metrics than plant richness or community composition, the fungal communities closely tracked plant community composition suggesting that plant communities likely serve as a key driver for foliar fungal communities."
} | 5,975 |
26067952 | PMC4463517 | pmc | 8,196 | {
"abstract": "Anoxybacillus geothermalis strain GSsed3 is an endospore-forming thermophilic bacterium isolated from filter deposits in a geothermal site. This novel species has a larger genome size (7.2 Mb) than that of any other Anoxybacillus species, and it possesses genes that support its phenotypic metabolic characterization and suggest an intriguing link to metals."
} | 90 |
36286193 | PMC10100480 | pmc | 8,198 | {
"abstract": "Abstract Toxic metal pollution requires significant adjustments in plant metabolism. Here, we show that the plant microbiota plays an important role in this process. The endophytic Sporobolomyces ruberrimus isolated from a serpentine population of Arabidopsis arenosa protected plants against excess metals. Coculture with its native host and Arabidopsis thaliana inhibited Fe and Ni uptake. It had no effect on host Zn and Cd uptake. Fe uptake inhibition was confirmed in wheat and rape. Our investigations show that, for the metal inhibitory effect, the interference of microorganisms in plant ethylene homeostasis is necessary. Application of an ethylene synthesis inhibitor, as well as loss‐of‐function mutations in canonical ethylene signalling genes, prevented metal uptake inhibition by the fungus. Coculture with S. ruberrimus significantly changed the expression of Fe homeostasis genes: IRT1 , OPT3 , OPT6 , bHLH38 and bHLH39 in wild‐type (WT) A. thaliana . The expression pattern of these genes in WT plants and in the ethylene signalling defective mutants significantly differed and coincided with the plant accumulation phenotype. Most notably, down‐regulation of the expression of IRT1 solely in WT was necessary for the inhibition of metal uptake in plants. This study shows that microorganisms optimize plant Fe and Ni uptake by fine‐tuning plant metal homeostasis.",
"introduction": "1 INTRODUCTION Plant metal management depends on the plant genotype, physicochemical characteristics of the soil and the presence or absence of other environmental cues, such as drought and light conditions (Verbruggen et al., 2009 ). In recent years, it has become clear that plant‐associated microorganisms (both soil microorganisms and plant endophytes) also play a significant role in this process (reviewed in Domka et al., 2019a ; Trivedi et al., 2020 ). The best‐described aspects of microorganism‐dependent impacts on plant metal tolerance and essential metal uptake are microorganism‐driven metal transformation processes taking place in the soil. As a result of microorganisms, metal availability may increase or decrease (Han et al., 2020 ; Janoušková & Pavlíková, 2010 ). Significantly less is known about how/if symbiotic microorganisms impact plant metabolism during adaptation to particular (high or low availability) metal conditions. Metals such as Fe, Zn and Cu are essential nutrients, deficiencies of which cause severe reductions in plant productivity. Simultaneously, both essential and nonessential metals in excess can be toxic to plants. Excess concentrations of iron and other metals in soils cause yield reductions in the world's leading crop species, wheat and rice, of up to 100%; thus, metal and particularly iron toxicity is regarded as one of the most formidable crop management and research challenges (Aung & Masuda, 2020 ; Briat et al., 2015 ). Our understanding of plant metal uptake and distribution control is scarce; thus, understanding the mechanisms of plant Fe tolerance and uptake seems essential. Iron availability and toxicity are determined by multiple factors, including soil aeration, organic matter content, pH, soil redox potential and microbial populations (Moreno‐Jiménez et al., 2019 ). Soil pH is one of the main factors affecting Fe solubility and availability. It was reported that an increase in soil pH above 7 can cause a ca. 50% reduction in the availability of Fe to plants (Rengel, 2015 ; Zhang et al., 2019 ). Therefore, plants do not encounter excess of Fe in calcareous, saline and alkaline soils. (Mori et al., 1991 ). Conversely, in acidic soils, Fe is widely available in excess (Vigani et al., 2019 ; Zhang et al., 2019 ). Another important factor that affects Fe availability is the organic matter content in the soil. The deposition of bicarbonates and phosphates abundantly present in organic matter prevents Fe uptake by plants (Moreno‐Jiménez et al., 2019 ). The primary cause of Fe toxicity is its ROS generation potential (in the Fenton reactions) and resultant interference in plant redox balance (Halliwell & Gutteridge, 1992 ). To counteract the deleterious effects of iron, plants have evolved strategies to prevent Fe accumulation and toxicity. These strategies include ion uptake selectivity, exclusion of Fe at the root surface, oxidation and formation of ‘iron plaques’ in the rhizosphere, sequestration in apoplasts and vacuoles, occlusion in ferritin proteins and enzymatic detoxification (reviewed in Kobayashi & Nishizawa, 2012 ; Thomine & Vert, 2013 ). The plant hormone ethylene was shown to play a central role in regulating the plant response to metals with respect to both deficiency and excess. Even exposure to mild levels of nonessential Cd induces ethylene synthesis (Schellingen et al., 2014 ). The analysis of ethylene production in different cultivars of rice and maize revealed a relationship between ethylene production and metal tolerance. Metal tolerance was attributed to higher ethylene production (Lu & Kirkham, 1991 ; Peng & Yamauchi, 1993 ; Yamauchi & Peng, 1995 ). Plant hormone balance, including ethylene homeostasis, plays a vital role in plant–microorganism interactions. Plants deficient in ethylene signalling were shown to be unable to benefit from interacting with plant growth‐promoting microorganisms (Camehl et al., 2010 ). Additionally, several lines of evidence have shown that the beneficial impact of the plant microbiota is related to microorganism modulation of ethylene balance and signalling (reviewed in Ravanbakhsh et al., 2018 ). The plant response to Fe starvation/deficiency is the best‐described aspect of plant Fe homeostasis. Under Fe deficiency (Strategy I), plants, including species from the Brassicaceae family ( Arabidopsis , Cardaminopsis ), solubilize inaccessible Fe 3+ by secreting H + into the rhizosphere. Simultaneously, the transcription of genes involved in Fe acquisition, such as ferric reductase 2 ( FRO2 ), ferric reductase 1 ( FRO1 ), iron‐regulated transporter 1 ( IRT1 ), iron‐regulated transporter 2 ( IRT2 ) and fer‐like iron deficiency‐induced transcription factor ( FIT ), is up‐regulated. IRT1 is the major root iron transporter and, concurrently, an important metal sensor. Its expression in root epidermal cells is controlled by Fe availability in the rhizosphere and the plant Fe status. Signalling molecules, such as auxin, ethylene and gibberellins, were shown to control the expression of IRT1 and other Fe homeostasis proteins under Fe deficiency. Under such conditions, the expression of IRT1 is strongly and rapidly up‐regulated, whereas under excess Fe availability, its expression remains unchanged or is down‐regulated, as shown by different authors (Ravet et al., 2009 ; Tissot et al., 2019 ; Vert et al., 2003 ). Apart from transcriptional control of IRT1 expression, IRT1 is controlled by protein degradation triggered by excess accumulation of noniron divalent metals (Zn, Co, Mn), which are sensed by the transceptor histidine‐rich motif in an intrinsically disordered cytosolic loop (Dubeaux et al., 2018 ). Under excess metal concentration, polyubiquitination by UB‐conjugating enzymes and IRT1 targeting into the endosome for degradation is induced, limiting the intake of noniron metals by the plant and avoiding toxicity Dubeaux et al., 2018 ). Due to the high similarity of non‐iron divalent metal ions, they are massively taken up under metal deficiency. The aim of this work was to identify how endophytic microorganisms facilitate plant adaptation to metal toxicity. We hypothesized that plant inherent microbiota induces changes in plant metabolism that result in improved metal tolerance. One of the processes that is affected by endophytic microorganisms is plant metal uptake and distribution. We also hypothesized that microorganism interference in plant ethylene homeostasis plays a central role in this process. In this study, we used the endophytic yeast Sporobolomyces ruberrimus isolated from Arabidopsis arenosa to test its ability to improve plant growth under excess of toxic metals (TMs). This basidiomycete yeast is known for its ability to synthesize important carotenoids for the food and pharmaceutical industry: torularhodin, torulene, γ‐carotene and β‐carotene (Cardoso et al., 2016 ). It has also been shown to be widespread among different species of plants in both mono‐ and dicotyledon phyla (Khunnamwong et al., 2018 ; Longa et al., 2022 ). For the purpose of this study, we tested the ability of the fungus to protect its plant host under different metal regimes, including conditions mimicking the mine dump ‘Bolesław’ in southern Poland.",
"discussion": "4 DISCUSSION Although numerous reports show that the plant microbiota plays a significant role in the optimization of metal uptake by the plant, the mechanism of this process is largely unknown. In the present study, we described microorganism‐dependent alterations in plant metabolism and transport that lead to metal: Fe and Ni uptake inhibition by the plant. The results of our investigations clearly show that the endophytic yeast S. ruberrimus affects plant ethylene homeostasis, which eventually leads to down‐regulation of the expression of the plant iron (and other divalent metals) transporter IRT1 and metal uptake inhibition. Plants cocultured with the endophytic yeast accumulated significantly less Fe and Ni and exhibited less metal stress symptoms. Plant tolerance against TMs depends on a complex network of mechanisms that can be classified into two general categories: metal avoidance and metal detoxification (reviewed in Podar & Maathuis, 2022 ). In the present study, metal uptake inhibition by the fungus was selective: Fe and Ni uptake by the plant was strongly limited, whereas Zn and Cd uptake was not changed. This finding can be partially explained by the immobilization of Fe in the substrate/medium by the fungus (Jędrzejczyk et al., under review), making it unavailable for plant uptake. Ni, however, does not form insoluble deposits in the soil. It is, however, incorporated into the crystal or amorphous aggregates such as carbonates, especially iron‐derived compounds, due to immobilization and coprecipitation with other oxides, such as Fe and Mn. The nature of the Ni‐compound texture has a significant role in metal transformation in soil (Rajaie et al., 2008 ). The chemical and physical forms of metals in soil depend on many factors, such as pH, water content and the cation exchange capacity (Han & Banin, 1999 ); hence, further detailed analysis of Ni, Cd and Fe transformation in the soil is needed. Additionally, as expected, we did not observe any precipitates in the fungal liquid culture supplemented with excess Ni, which indicated that Fe and Ni uptake inhibition by the fungus might be caused not only by metal immobilization in the medium but also by the fungus interference/effect on plant metal uptake. Plant uptake of essential metals depends on metal availability in the soil and on tightly regulated intrinsic metal homeostasis maintenance machinery. The functioning of this molecular network is necessary for optimal plant growth and development. Symbiotic microorganisms, together with host plants, coadapt to the environment they inhabit, optimizing the holobionts' metabolism to particular environmental contexts (Ravanbakhsh et al., 2018 ). The S. ruberrimus used in the investigations was isolated from the serpentine population; thus, this strain was adapted to excess Ni and Fe. We cannot exclude the possibility that the specificity of the metal species uptake inhibition was related to the origin of the fungus; this, however, requires further investigation. Previous reports have shown that ethylene signalling is necessary for A. thaliana growth promotion by Serendipita indica , most likely for balancing beneficial and nonbeneficial traits in plant–fungus symbiosis (Camehl et al., 2010 ). In the present study, inoculation with the fungus activated plant ethylene signalling and expression of ethylene effector genes. Simultaneous metal treatment and colonization by the endophyte induced a synergistic effect in regard to ethylene signalling, manifested as GUS activity in the A. thaliana EBS:GUS reporter line. This result indicated that ethylene signalling was activated in plants by both metal toxicity cues and interactions with microorganisms independently, however, simultaneous treatment resulted in significantly higher activity than the individual treatments. Thus, we can expect that ethylene homeostasis in plants under metal toxicity is established by the integration of inputs from multiple environmental stimuli, including interaction with microorganisms. The outcome of this new equilibrium is an adaptation to the environment. One aspect of this adaptation may be the regulation of metal uptake and within plant distribution. According to Cao et al. ( 2009 ), the expression of EIN2 was up‐regulated in Arabidopsis exposed to excess Pb, and the ein2‐1 mutant was found to accumulate higher concentrations of the metal. Additionally, the expression of genes encoding ET signalling proteins was up‐regulated in the metal‐tolerant A. thaliana ecotype Col‐0 in relation to the metal‐sensitive Wassilewskija (Fu et al., 2014 ), which indicates that the ethylene response is necessary for plant tolerance to excess metals. Here, we showed that ethylene was necessary to inhibit plant Fe uptake by the endophytic fungus. After inhibiting ethylene production in plants by pharmacological inhibition of the ethylene synthesis rate‐limiting ACS, E+ plants did not take up less Fe from the medium than E‐ plants. A similar relationship was observed in A. thaliana signalling mutants. The fungus was unable to inhibit Fe uptake in ethylene‐insensitive ein2‐1 and etr1‐1 mutants, indicating that for the endophyte to exert its metal‐inhibiting effect on its host, ethylene synthesis and signalling were necessary. Interestingly, the ethylene signalling mutants ein2‐1 , ein3‐1 and etr1‐1 took up significantly less Fe from medium than WT plants (Figure 5b ). This suggested that ethylene insensitivity or even deficiencies in ethylene signalling (due to high redundancy between EIN3 and EIL1) resulted in Fe uptake inhibition. According to available reports, etr1‐1 and ein3‐3 are more resistant to excess Li (Fu et al., 2014 ) than WT Arabidopsis . The application of ACC deaminase‐producing bacteria inhibits metal uptake by plants (Sun et al., 2022 ), indicating that the deactivation of ethylene biosynthesis/signalling can be beneficial for plants under metal toxicity. Conversely, Li et al. ( 2015 ) showed that activation of ethylene synthesis under excess Fe plays an important role in Arabidopsis Fe tolerance. Ethylene‐overproducing eto1‐1 mutants were found to accumulate less Fe than WT plants. Additionally, in plants treated with ACC, Fe‐induced inhibition of root growth was alleviated in eto1‐1 plants. All of these examples indicate that the role of ethylene in plant metal uptake and metal tolerance is not unequivocal; both up‐ and down‐regulation of ethylene production synthesis may result in plant metal tolerance. However, our results show that for endophyte‐dependent inhibition of metal uptake, ethylene is essential. Another interesting observation was that inhibition of ethylene production and loss‐of‐function mutations in essential ethylene perception and signalling molecules resulted in increased Fe uptake in plants cocultured with the fungus. Ethylene was shown to activate IRT1 transcription under Fe deficiency (García et al., 2015 ; Lingam et al., 2011 ; Lucena et al., 2006 ; Waters et al., 2007 ) and to contribute to the regulation of IRT1 expression along the root axis (Ivanov et al., 2014 ). Under iron excess, the role of ethylene in the regulation of IRT1 expression is unknown. In general, information concerning IRT1 turnover in conditions other than deficit or optimum is limited. Here, inoculation with the endophyte triggered the down‐regulation of IRT1 expression in an ethylene‐dependent manner, which may explain the reduced accumulation of Fe by E+ plants. A similar phenomenon was reported by (Fan et al., 2014 ; Xu et al., 2018 ). The authors showed that application of abscisic acid (ABA) or inoculation with ABA‐producing microorganisms suppressed Cd root uptake by inhibiting the expression of IRT1. In the present study, no metal uptake inhibition was observed in E+ irt1 insertional mutants and 35S‐ irt1 transgenic plants. Both the loss‐of‐function mutation in the IRT1 gene and the constitutive expression of IRT1 from its nonnative promoter 35S prevented the fungus from inhibiting Fe uptake by the plant. This finding confirmed that alterations in metal uptake in plants inoculated with the yeast were IRT1 dependent and that transcriptional control of IRT1 expression was required for Fe uptake inhibition by the endophyte. It is worth noting that overexpression of IRT1 in 35S‐ irt1 transgenic plants was not sufficient to confer an enhancement of Fe uptake, which is in line with the results of Barberon et al. ( 2011 ), Connolly et al. ( 2002 ) and Kerkeb et al. ( 2008 ). This suggests that metal uptake inhibition by the fungus might not be related to IRT1 transporter function but to other unknown functions. The regulation of IRT1 expression in E+ plants was also examined under Fe starvation. E+ Arabidopsis exhibited fewer stress symptoms after transfer to Fe‐deficient medium, and despite no differences at the IRT1 messenger RNA level, the protein content was up‐regulated in comparison to E− plants. This finding provides further evidence that the endophytic microorganism interferes with IRT1 expression regulation (not necessarily at the transcriptional level) and possibly plays an important role in plant Fe homeostasis. Even though we were not able to provide evidence that E+ plants accumulate more Fe than E− Arabidopsis , we showed that S. ruberrimus can improve plant nutrition by siderophore production and increasing Fe availability. This result suggests that in the rhizosphere of the plant–fungi consortium, Fe availability may be increased by the microorganism, and simultaneously, the plants' ability to take up Fe is also improved. The relationship between IRT1 expression and metal availability may be altered in plants by microorganisms; our results show that IRT1 expression in plants cultured in conditions with possibly higher Fe availability (compared with E− plants) is activated. According to the canonical model, the expression of this gene is dependent on Fe availability; IRT1 expression was shown to be positively correlated with decreasing concentrations of available Fe. The presence of the endophytic microorganism seems to change this relationship. The results presented in this study show that endophytic microorganisms play an important role in plant metal homeostasis. The basidiomycete yeast S. ruberrimus possesses the ability to control Fe availability in the soil by siderophore production under Fe deficiency and by Fe immobilization under excess Fe. Additionally, the microorganism optimizes plant Fe and Ni uptake by fine‐tuning plant metal homeostasis mechanisms. Our results shed new light on the importance and complexity of plant–microorganism interactions under metal toxicity."
} | 4,862 |
36814252 | PMC9948338 | pmc | 8,200 | {
"abstract": "Background Energy is the basis and assurance for a world's stable development; however, as traditional non-renewable energy sources deplete, the development and study of renewable clean energy have emerged. Using microalgae as a carbon source for anaerobic bacteria to generate biohydrogen is a clean energy generation system that both local and global peers see as promising. Results Klebsiella pneumonia , Enterobacter cloacae , and their coculture were used to synthesize biohydrogen using Oscillatoria acuminata biomass via dark fermentation. The total carbohydrate content in O. acuminata was 237.39 mg/L. To enhance the content of fermentable reducing sugars, thermochemical, biological, and biological with magnesium zinc ferrite nanoparticles (Mg-Zn Fe 2 O 4 -NPs) pretreatments were applied. Crude hydrolytic enzymes extracted from Trichoderma harzianum of biological pretreatment were enhanced by Mg-Zn Fe 2 O 4 -NPs and significantly increased reducing sugars (230.48 mg/g) four times than thermochemical pretreatment (45.34 mg/g). K. pneumonia demonstrated a greater accumulated hydrogen level (1022 mLH 2 /L) than E. cloacae (813 mLH 2 /L), while their coculture showed superior results (1520 mLH 2 /L) and shortened the production time to 48 h instead of 72 h in single culture pretreatments. Biological pretreatment + Mg-Zn Fe 2 O 4 NPs using coculture significantly stimulated hydrogen yield (3254 mLH 2 /L), hydrogen efficiency)216.9 mL H 2 /g reducing sugar( and hydrogen production rate (67.7 mL/L/h) to the maximum among all pretreatments. Conclusion These results confirm the effectiveness of biological treatments + Mg-Zn Fe 2 O 4 -NPs and coculture dark fermentation in upregulating biohydrogen production. Supplementary Information The online version contains supplementary material available at 10.1186/s12934-023-02036-y.",
"conclusion": "Conclusions Based on the current results, it can be concluded that biological pretreatment maximizes reducing sugar release, avoids generating fermentation-inhibiting compounds and reduce chemical inputs compared to thermochemical pretreatments. Mg-Zn Fe2O4-NPs enhanced Trichoderma harzianum cellulolytic activity to give the bacteria with the largest easily fermentable sugar (230.48 mg/g) to produce the highest H 2 evolution among other pretreatments owing to their high hydrolase content and therefore ability to hydrolyze O. acuminata biomass. Furthermore, the coculture of the two studied bacteria, K. pneumoni and E. cloacae, produced more hydrogen (3254 mLH 2 /L), reflecting synergistic consumption of available resources. Therefore, biological pretreatment + Mg-Zn Fe 2 O 4 -NPs and coculture dark fermentation are helpful for biohydrogen production using microalgae biomass.",
"discussion": "Results and discussion Isolation and identification of microalgal species Oscillatoria was genetically identified using 16S rRNA gene, respectively. The sequence analysis was done with the neighbor-joining algorithm based on the parameter distance (NJ-PD) by aligning the 16S gene sequence with 16S nucleotide sequences of 16 cyanobacteria species plus four 16S sequences of Planktothrix agardhii as an outgroup (Fig. 1 ). Each Oscillatoria sp. formed monophyletic subclades with bootstrap support, ranging from 87 to 100. The sequences were subjected to BLAST homology searches of the 16S sequence, indicating that the closest match was Oscillatoria acuminata (Fig. 1 ). Oscillatoria sp showed very high similarity (~ 98–100%) to Oscillatoria acuminata (MK014210, NR_102463, and CP003607). The sequences of the cyanobacteria have been submitted to Genbank (OP277605). Fig. 1 Neighbor-Joining (NJ) dendrograms showing the isolated Oscillatoria acuminata based on 16S rRNA nucleotide sequences, respectively. Bootstrap values higher than 70 are shown below the branches of the trees Estimation of growth and biochemical composition of Oscillatoria acuminata The algal growth was determined by optical density and dry weight. The results in Fig. 2 indicated that the growth curve of O. acuminata showed no lag phase. The exponential phase was from zero to the 14th day. These cultivation periods are shorter than those recorded by El-Sheekh et al. [ 28 ], who denoted that O. acuminata could grow till the 22nd day of cultivation. This difference in biomass may be due to the differences in culture medium, cultivation condition, and metabolic activity [ 29 , 30 ]. Fig. 2 Growth curve of O. acuminata using dry weight and optical density (OD 750) The biochemical composition was estimated at the end of the log growth phase on the 16th day of culture (Table 1 ). Chl a and carotenoid content extracted from O. acuminata recorded maximum value at 16th (18.1 µg/mL, 3.3 µg/mL respectively). O. acuminata showed the highest carbohydrate content of 237.39 mg/L, while the total protein content was 180.78 mg/L. These results are less or more than those recorded in recent studies [ 29 ]. This could be attributed to the difference between the isolates, biomass yield, and growth conditions [ 31 ]. Microalgal biomass was harvested on the 16th day for use in further experiments. Table 1 Metabolic constituents of O. acuminata Biochemical constituents Contents Chlorophyll a (µg/mL) 18.1 ± 0.3 Carotenoid (µg/mL) 3.3 ± 0.07 Total soluble protein (mg/L) 180.78 ± 2.21 Total soluble carbohydrate (mg/L) 237.39 ± 4.12 Values represent means of three replicates ± standard error Influence of different pretreatment methods on biomass hydrolysis Pretreatment is crucial in releasing fermentable sugars to be used in microbial fermentation. Thermochemical pretreatment of microalgal biomass As shown in Fig. 3 , the highest value of reducing sugar was recorded at 1.5% H 2 SO 4 of 45.34 mg/g dry weight DW, while any increase or decrease in H 2 SO 4 concentrations reduces sugar content reduced. Among various acids hydrolysis, Hessami et al. [ 32 ] proved that H 2 SO 4 is the best hydrolysis acid compared to HCl, HClO 4 and CH 3 COOH. Thus, acid hydrolysis using sulphuric was performed in the current study as an efficient, fast, appropriate and cost-effective technique [ 33 ]. Furthermore, El-Souod et al. [ 34 ] demonstrated that hydrolysis of microalgal biomass using H 2 SO 4 pretreatment showed the best results, especially at 1.5%. These results run with Ashour et al. [ 35 ], Elshobary et al. [ 33 ] and Li et al. [ 36 ], who found that dilute acid pretreatment was extensively used since strong acid would induce the excessive breakdown of the complex material, resulting in a release of fermentable sugars. However, acid pretreatment produces fermentation-inhibiting compounds for example, furfural, hydroxymethylfurfural, and levulinic acid [ 37 ]. In contrast, biological pretreatment, such as fungal treatment, may be a further practical alternative than typical acid pretreatment processes [ 38 ]. Fig. 3 Effect of different concentrations of acid pretreatments on O. acuminata biomass for reducing sugar production. Different capital letters of the plotted series indicate significant differences at p ≤ 0.05 using Duncan's test Biological treatment of microalgal biomass This study studied biological pretreatment as an eco-friendly tool to maximize reducing sugar release, avoid generating fermentation-inhibiting compounds and reduce chemical inputs. Fungi produce several hydrolysis enzymes such as cellulase, filter paper cellulase and β-glucosidase. These enzymes break down the algal cell wall and convert complex polysaccharides to fermentable monosaccharides. T. harzianum (OP264067) showed the highest cellulolytic activity compared to 20 isolated fungi [ 39 ] and naturally produces hydrolysis enzymes such as cellulase, β-glucosidase and xylanase [ 40 ]. Screening of different substrates for cellulolytic activity produced by T. harzianum using solid-state fermentation (SSF) To further increase the hydrolysis process and reducing sugar content, optimization of biological pretreatment was conducted using different substrates (rice straw, wheat straw, and wheat bran) Fig. 4 . Results demonstrated that wheat bran was the most efficient substrate that improved the cellulolytic activity of T. harzianum to the maximum. The maximum activities of CMCase, βGase, Fpase were 1320.57, 940.59 and 402.41 U/g DW substrate, respectively and total cellulolytic activity was 2663.56 U/g DW using wheat bran, followed by rice straw (1582.52 U/g DW), and wheat straw (1345.06 U/g dry weight substrate). Our results agreed with El-Shishtawy et al. [ 41 ], who reported that wheat bran substrate inhanced the hydrolysis enzymes, total proteins, and carbohydrates of T. virens to the maximum using base pretreated. Moreover, wheat bran revealed its potential as a cheap substrate for more remarkable enzyme synthesis by Penicillium citrinum [ 42 ]. One unit of FPase, βGase, and CMCase was described as µg of reducing sugars produced per minute per gram of substrate dry weight. Fig. 4 Screening of different substrates for cellulolytic activity produced by T. harzianum using solid-state fermentation (SSF). Different capital letters of the plotted series indicate significant differences at p ≤ 0.05 using Duncan's test Effect of Mg-Zn Fe 2 O 4 -NPs on the hydrolytic enzymes of T. harzianum Cellulolytic action of T. harzianum using wheat bran as substrate was further improved by Mg-Zn Fe 2 O 4 -NPs as shown in Fig. 5 . The results concluded a gradual increase in all cellulolytic activities (CMCase, βGase, Fpase and total cellulolytic activity) by applying different Mg-Zn Fe 2 O 4 -NPs concentrations until reaching the maximum activities at 60 ppm of Mg-Zn Fe 2 O 4 -NPs for CMCase, βGase, Fpase (2162.40, 877.65 and 659 U/g respectively) with the highest total cellulolytic activity (4319.37 U/g). Interestingly, 60 ppm of Mg-Zn Fe 2 O 4 -NPs improved the total cellulolytic activity about two times over the untreated control. It has been observed that the presence of metal oxides is essential for the production of cellulase enzymes and growth of microorganisms in fermentative media. This is due to the fact that these micronutrients play a crucial role in the synthesis of cellulase enzymes and microbial growth. Therefore, by incorporating Mg-Zn Fe 2 O 4 -NPs as a catalyst, it enhanced the production of cellulase enzymes and biohydrogen [ 20 ]. Moreover, the stability of enzymes can be significantly improved through protein adsorption on nanomaterials, due to the unique properties provided by high surface area to volume ratios [ 43 ]. Furthermore, the high surface area of nanoparticles offers an efficient matrix for immobilizing enzymes, resulting in enhanced stability. The large surface area of these materials allows for multiple points of attachment for enzyme particles, which prevents protein unfolding and ultimately leads to improved enzyme stability [ 44 ]. Srivastava et al. recorded that nickel ferrite nanoparticles stimulated the yield of total cellulase enzyme using remaining cyanobacteria biomass of Lyngbya limnetica as feedstock [ 18 ]. Asar et al . reported a comparable increase in cellulase and sugar yield following the application of iron oxide magnetic nanocomposites [ 45 ]. Fig. 5 Effect of Mg-Zn Fe 2 O 4- NPs for cellulolytic activity produced by T. harzianum Ps-2 using solid-state fermentation (SSF). All data represented means of 3 replicas ± standard deviation (SD). Different capital letters in each plotted series indicate significant differences at p ≤ 0.05 using Duncan's test Effect of enzyme dose and incubation duration on cyanobacterial biomass One of the most common and straightforward experiments used to improve the fermentation process is single-factor optimization [ 46 , 47 ]. Effect of enzyme dose and incubation duration were determined and optimized by a series of single-factor experiments to provide optimal conditions for the fermentation process. The results illustrated in Fig. 6 showed that the highest yield (106.48 mg/g) of reducing sugar and cellulolytic activity (5958.50 U/g) were recorded at 1:2 crude enzyme: algal suspension (33%) concentration after 12 h. Cellulolytic activity at an enzyme concentration of 33% increased by 52.31%. These results may be explained by increasing algal suspension increases the substrate and available carbon source for the enzyme activity. In this context, Vishwakarma and Malik investigated the enzymatic pretreatment effects using crude enzymes from T. reesei increased by twofold in protein efflux and a 41% increase in released sugars after 12 h of enzymatic pretreatment at 33% (v/v) enzyme concentration [ 48 ]. Fig. 6 Effect of enzyme dosage and incubation time on biodegradation of O. acuminata biomass . All data represented means of 3 replicas ± standard deviation (SD). Different capital letters in each plotted series indicate significant differences at p ≤ 0.05 using Duncan's test Cellulolytic activity and reducing sugar in the presence of Mg-Zn Fe 2 O 4 -NPs by solid state fermentation Optimized conditions of 1:2 crude enzyme: algal suspension at 12 h incubation were subjected to Mg-Zn Fe 2 O 4 -NPs to maximize the cellulolytic activity and reducing sugar content of O. acuminata biomass. As a result, Mg-Zn Fe 2 O 4 -NPs significantly induced the production of reducing sugar and cellulolytic activity by 20.1% (7158.50 U/g) and 1.16-fold of reducing sugar (230.48 mg/g, 23.05 g/L) compared to biological pretreatment without NPs. Fewer studies have been conducted on the use of nanoparticles to treat enzymatic conversion of biomass for sugar production, but the results of this study showed a higher production of sugar and higher cellulolytic activity compared to previous research. For example, Srivastava et al. recorded enhanced cellulolytic activity to 222 U/g by using nickel ferrite nanoparticles using residual biomass of Lyngbya limnetica [ 18 ]. Zanuso et al. obtained 21.84 g/L reducing sugar production using corn cob biomass using magnetic nanoparticles [ 49 ], and Jiang et al. showed an improvement in sugar production using corn stalk with the use of cellulase enzyme immobilized on nanocomposite compared to using the crude enzyme [ 50 ]. It is important to note that the type and form of magnetic carboxymethyl chitosan/calcium alginate–cellulase nanomaterials used can greatly impact the efficiency of hydrolysis enzymes, along with the type of fungal strain used. As previously discussed, the type and form of nanomaterials have a significant impact on the effectiveness of enzymes, in addition to fungal species used. Fourier Transform Infrared Spectroscopy (FTIR) of the algal biomass before and after pretreatment FTIR was used to demonstrate the structural and chemical variations in different pretreatments (chemical, biological, and biological pretreatment + Mg-Zn Fe 2 O 4 -NPs) using O. Acuminata biomass. The Hydrogen bond band at 3600–3250 cm −1 was diminished after the hydrolysis due to the destruction of many H-bonds in the cellulose molecules [ 51 ]. Substantial reduction in the intensity at 3600–3250 cm −1 in biological + Mg-Zn Fe 2 O 4 -NPs pretreatment, indicating the breaking down of hemicellulose by cracking of H-bond in the amorphous cellulose [ 52 , 53 ]. Variations in band intensity and position may imply a reduction in structural component concentration as well as the development of different kinds of groups from free radical groups created during treatments [ 54 ]. The expansion of the band also shows the bond's weakening as a result of pretreatment. Because these peaks reflect proteins, secondary amines (proteins, lipids), saccharides, and carbohydrates, a drop in intensity correlates to a reduction in their concentration. The peak at 2860 cm −1 was nearly gone after biological treatment, indicating that the carbohydrate and lipid complexes had been broken down [ 55 ]. The functional groups of methylene (-CH 2 -), methyl (-CH 3 ), and glycosidic bond dominated the spectrum area at 1460–1160 cm −1 . These groups are extensively dispersed and generated in numerous monosaccharides of the hemicellulose molecule, such as xylose, arabinose, and mannose. [ 53 , 56 ]. The glycosidic link (C–O–C) breaking down in amorphous cellulose causes an obvious vibration at wavenumbers of 600–1050 cm −1 . The intensities of the typical cellulose peaks were found to diminish after treatments (Allard and Templier, 2001). The shift in peak intensity at wavenumbers 700–600 cm −1 was related to the cis–trans structural alteration in the cell wall carbohydrates during biomass processing. The band at 600–900 cm-1 was ascribed to the vibration of ether groups or glycosidic bonds, which were the primary connections in polysaccharide compounds and pectin [ 57 ]. These results verified that Mg-Zn Fe 2 O 4 -NP S enhanced the efficiency of biological pretreatment of O. acuminata biomass by sequentially cracking the cell wall structural components (Fig. 7 ). Fig. 7 The diagnostic fingerprint regions of the FT-IR spectra of raw and pretreated O. acuminata biomass. A, untreated biomass; B, Chemical pretreatment; C, Biological pretreatment, and D, Biological pretreatment enhanced by Mg-Zn Fe 2 o 4 -NPs Scanning Electron Microscopic (SEM) analysis of the cyanobacterial biomass before and after treatment In order to get insights into the structural and morphological variations in O. acuminata biomass during pretreatments (thermochemical, biological, and biological pretreatment + Mg-Zn Fe 2 O 4 -NPs). SEM microscopy analysis was used. Figure 8 reveals that the untreated biomass had a normal, smooth, and compact surface structure without degradation (Fig. 8 A). After chemical treatment, the cell wall was partially broken (Fig. 8 B). In biological treatment, obvious damage observed, resulting in exposure of the inner structure (Fig. 8 C) [ 58 ]. The structure and integrity of O. acuminata cell walls might be efficiently disrupted with biological (fungal) pretreatment. The structure was utterly damaged after applying the biological pretreatment enhanced by Mg-Zn Fe 2 O 4 -NPs. This biomass treatment led to severe cell morphology alterations, including severe cell breaking and visible cellular debris (Fig. 8 D) [ 59 ]. Fig. 8 Scanning Electron Microscopy (SEM) observation of O. acuminata biomass at 2000 × magnification. Where, A , untreated biomass; B , Chemical pretreatment; C , Biological pretreatment and D , Biological pretreatment enhanced by Mg-Zn Fe 2 o 4 -NPs"
} | 4,633 |
31418553 | PMC6764109 | pmc | 8,201 | {
"abstract": "We\ndescribe the triggered assembly of a bioinspired DNA origami\nmeshwork on a lipid membrane. DNA triskelia, three-armed DNA origami\nnanostructures inspired by the membrane-modifying protein clathrin,\nare bound to lipid mono- and bilayers using cholesterol anchors. Polymerization\nof triskelia, triggered by the addition of DNA staples, links triskelion\narms to form a mesh. Using transmission electron microscopy, we observe\nnanoscale local deformation of a lipid monolayer induced by triskelion\npolymerization that is reminiscent of the formation of clathrin-coated\npits. We also show that the polymerization of triskelia bound to lipid\nbilayers modifies interactions between them, inhibiting the formation\nof a synapse between giant unilamellar vesicles and a supported lipid\nbilayer.",
"conclusion": "Conclusion We have demonstrated the controlled formation of extended DNA origami\ntriskelion arrays on lipid bilayers by electron and optical microscopy.\nWe have observed that polymerized networks of triskelia can induce\nsub-micrometer deformation of a lipid monolayer, which is reminiscent\nof the formation of clathrin-coated pits. Triskelia also modulate\nlipid interfaces: they mediate an attractive interaction between free\nbilayers but can inhibit the formation of an interface when one bilayer\nis bound to a rigid substrate. These results help demonstrate the\npotential of biomimetic membrane-associated nanostructures as tools\nto control the dynamic behavior of lipid membranes and their shapes\nand interactions. We anticipate that the exploration of the design\nof membrane-modifying nanostructures will lead both to a greater understanding\nof natural processes and to biomimetic systems for signaling, synthesis,\nand reproduction based on membrane-bound compartments.",
"discussion": "Results\nand Discussion We designed a three-arm DNA origami nanostructure\nwhose shape resembles\nthat of the clathrin triskelion ( Figure 1 , Supporting Figure S1 ). 21 The DNA origami is approximately\n20 times more massive than its natural counterpart. 22 Each triskelion arm consists of 28 parallel DNA helices\norganized on a honeycomb lattice to create a 13 nm diameter bundle\nthat is 30 nm long. The angles between the arms of the DNA triskelion\nare constrained by rigid three-helix bundles that form bridges connecting\nthe midpoints of each arm ( Figure 1 A, Supporting Figure S1 ).\nWe sought to control the shape of the triskelia by using bridges of\ndifferent lengths: the flat triskelion has arms of 92 base pairs (bp)\n(angle between arms approximately 120°, Figure 1 A,B), and the curved triskelion has arms\nof 84 bp (forming a triangular pyramid with a height of approximately\n18 nm, Figure 1 C,D).\nArms are linked to each other, where they meet near the center of\nthe structure by single-stranded DNA links, formed by routing the\nscaffold strand between arms, and by the bridges between arms (through\nwhich the scaffold also runs). Triskelia can be programmed to assemble\ninto extended arrays on addition of staples that link arms end-to-end.\nAn edge in this network is approximately 60 nm long, twice the dimension\nof the natural clathrin lattice. 23 Figure 1 Triskelion\nmonomers, dimers, and arrays. (A–D) Designed\nstructure and transmission electron microscopy (TEM) micrographs of\n(A) flat triskelion monomer; (B) dimer formed by linking arms 1 end-to-end;\n(C) curved triskelion monomer; and (D) curved dimer. Cylinders represent\nDNA helices. White arrows point to bridges, visible on some of the\nelectron micrographs. (E) TEM micrograph of lipid monolayer to which\nflat triskelion dimers were attached before their polymerization was\ntriggered by addition of DNA polymerization staples with 6 nt sticky\nends linking arms 2 and 3. A discrete, approximately circular membrane\nstructure is visible, covered by a partially ordered triskelion array\nwith hexagonal and pentagonal cells. (F) as in E but using curved\ntriskelion dimers. Magnified images in the right-hand panels of E\nand F are of structures similar to those shown on the left. We attribute\nthe formation of discrete, triskelion-coated membrane structures (E\nand F) to local deformation (budding) of the membrane induced by the\nnanostructure. We functionalized the triskelion\nwith three cholesterol groups\non the broader “bottom” surface of each arm to enable\ndirect attachment to lipid membranes. Six Alexa647 fluorophores are\nattached to the narrower upper layers of DNA helices ( Supporting Figure S1 ) for visualization by fluorescence microscopy.\nIn the right-hand electron micrograph in Figure 1 C the honeycomb cross-section of the bundles\nof helices in the origami arms is clearly resolved: the “top”\nof each arm, a layer of four DNA helices, is at the center of the\n3-fold structure, indicating that the triskelion has been forced into\na conformation in which the cholesterol anchors lie on the convex\nside of the distorted structure, as shown in the diagram ( Supporting Figure S2 ). Other images indicate that\ntriskelia also fold with the cholesterols on the inner side ( Supporting Figure S3 ). Clathrin self-assembles\ninto arrays as a result of weak attractive\ninteractions distributed along its arms. 22 , 24 For design purposes it was easier to localize the linking sites\nof our artificial triskelion at the extremities of its arms, doubling\nthe distance between two attached origami centers compared to that\nof clathrin. Individual DNA triskelia can be linked through the addition\nof DNA staples that bind to the origami scaffold at the ends of the\narms. Most of the experiments described used preformed triskelion\ndimers, which we found to give better-formed arrays when bound to\na membrane and polymerized. Dimers are formed during origami assembly\nusing six dimerization staples, each of which binds to scaffold domains\nat the ends of both of the arms 1 of the two component triskelia,\nforming six parallel connections between them ( Figure 1 B,D and Supporting Figures\nS4 and S5 ). Dimers can be linked into arrays by adding two\nsets of six DNA polymerization staples (12 strands in total), each\nof which hybridizes to the scaffold at the ends of one of the free\narms (arms 2 and 3), creating 10 overhanging 6-nucleotide “sticky\nends” at the end of each arm ( Supporting\nFigures S4 and S6 ). These sticky ends are designed such that\nhybridization of the sticky ends displayed on arm 2 of one dimer to\nthose on arm 3 of another links the two arms together: this connectivity\nis consistent both with a hexagonal array and with the formation of\npentagonal cells ( Supporting Figure S4D ),\nallowing curvature. The pattern of connections between pairs of arms\nis such that the two origamis are aligned with the membrane-binding\nfaces orientated in the same direction. Triskelion networks\nwere observed by transmission electron microscopy\n(TEM). Networks formed by polymerization in solution, in the absence\nof lipids, are generally extended and poorly ordered; occasional polygons\nare observed ( Supporting Figure S7 ). Triskelia\ninserted into the membranes of small unilamellar vesicles (SUVs) formed\ndense coatings around the vesicles ( Supporting\nFigure S8 ). However, distortion of the vesicles by the relatively\nharsh staining and drying protocol required for TEM ( Materials and Methods ) precluded clear identification of the\neffect of the DNA origami on membrane shape. 7 Polymerization of triskelion dimers on supported lipid monolayers\nwas imaged by TEM using an apparatus adapted from that developed in\nthe group of McMahon ( Supporting Figures S9, S10 and Materials and Methods ). 25 , 26 Addition of a small excess of 1,2-dioleoyl- sn -glycero-3-phosphocholine\n(DOPC, 10–20% more than monolayer coverage) to a pool of buffer\nin a Teflon well leads to the formation of a lipid monolayer, which\ncan be transferred to a gold TEM grid, stained with uranyl acetate,\nand observed by TEM ( Supporting Figure S11 ). Flat triskelion dimers injected into the well attach readily to\nthe lipid surface that covers the grid and polymerize into an extended\nnetwork reminiscent of clathrin assembly ( Figure 1 E). Curved triskelia form denser, less regular,\nnetworks ( Figure 1 F).\nIn both cases distinct, isolated, clusters, approximately circular\nin projection, are observed after, but not before, triskelion polymerization\n( Figure 1 E and F, Supporting Figures S11–S13 ). These structures\nare consistent with local deformation (budding) of the monolayers\ninduced by the formation of triskelion arrays and are similar to TEM\nimages of clathrin-coated pits on lipid monolayers. 25 − 27 In the case\nof flat triskelia, the circular clusters are frequently partly circumscribed\nby high-contrast crescent-shaped regions, characteristic of the projection\nimage of a partially collapsed bleb. 28 This\nis particularly clear in Supporting Figure S11E , in which two such structures are superimposed. In the case of curved\ntriskelia, the clusters are more densely stained with little evidence\nof collapse. Observation of on-edge triskelia at the edges of the\ncircular structures supports our interpretation that the membrane\nis deformed ( Supporting Figure S13 ). The membrane blebs induced by the two triskelion variants are qualitatively\nsimilar, despite that fact that one triskelion is designed without\nintrinsic curvature and the surface of the other is much more curved\nthan the blebs themselves. In neither case can our triskelia be forcing\nthe membrane to conform to an intrinsic array curvature. This is consistent\nwith the observation that the clathrin protein can form planar as\nwell as curved arrays, yet clathrin alone (if bound to the membrane)\nis sufficient to induce the formation of spherical buds. 29 It has been suggested that natural membrane\nbending, including the formation of clathrin-coated pits, is not directly\ndependent on the details of protein structure but is driven by crowding\nof membrane-anchored proteins: 30 protein\ncrowding alone is even sufficient to drive membrane fission. 31 We suggest that a similar mechanism drives the\nmembrane deformation that we observe: triskelion polymerization induces\nlocally dense membrane coverage and (in contrast to better-ordered\nand more rigidly connected arrays of membrane-bound DNA nanostructures) 8 the disordered triskelion lattice provides an\nentropic drive for the membrane to curve away to relieve crowding. The induction of membrane curvature through polymerization of DNA\ntriskelia is qualitatively different from the effects of membrane-bound\nDNA nanostructures reported previously. Most published studies of\nDNA nanostructure arrays on membranes are of planar arrays on planar\nmembranes. Indeed, tightly packed and well-ordered DNA arrays have\nbeen shown to induce planar deformations of naturally curved GUV membranes. 8 Where increases in membrane curvature have been\nachieved by membrane-bound DNA nanostructures, it has been through\nstrong interactions that force the membrane to conform to the intrinsic\ncurvature of the nanostructure. 9 − 11 , 13 It has been suggested that the natural role of the similarly shaped\nproteins that inspired these nanostructures is to sense, rather than\nto induce, membrane deformation. 32 − 34 In order to study\nof the dynamics of triskelion–membrane\ninteractions, we used fluorescence microscopy to observe triskelion\nassembly on lipid bilayers, using both giant unilamellar vesicles\n(GUVs) and supported lipid bilayers (SLBs). These experiments do not\nresolve the sub-micrometer membrane deformations described above but\ndo enable investigation of the effects of triskelia on membrane interactions.\nThe two triskelion types (flat and curved) behaved similarly in these\nexperiments. GUVs, comprising DOPC with 0.1 mol % fluorescently\nlabeled 1,2-dioleoyl- sn -glycero-3-phosphoethanolamine\n(Atto488-DOPE), were prepared\nby electroformation in 52 mM sucrose. 16 , 17 Triskelion\ndimers with Alexa647 labels ( Supporting Figure\nS1 ) were assembled in equiosmolar TE-MgCl 2 buffer,\nincubated for at least 10 min with the GUV suspension diluted 20-fold\nin the same TE-MgCl 2 buffer ( Supporting\nFigure S14 ), and observed using confocal microscopy. Dimers\nwere observed to bind to membranes homogeneously and diffuse freely\n( Figure 2 A,B and Supporting Movie M1 ). When polymerization of the\ndimers was triggered by addition of polymerization staples, DNA triskelia\nassembled into dense networks on the GUVs ( Figure 2 C,D, Supporting Movies\nM1–M4 ), observable as discrete, diffusing aggregates.\nWe do not see evidence of planar deformations similar to those induced\nby the densely packed nanostructure arrays reported by Czogalla et al . 8 Membrane-bound\nDNA triskelion dimers, before and after polymerization, are enriched\nat GUV–GUV interfaces ( Figure 3 ). Figure 2 DNA triskelia interacting with giant unilamellar vesicles.\n(A,\nC) Inferred distributions of DNA triskelion dimers on GUVs. Unpolymerized\nDNA triskelia are homogeneously distributed on the GUV surface; polymerization,\ntriggered by addition of polymerization staples, causes triskelia\nto assemble into arrays in mesoscopic domains. (B, D) Confocal micrographs\ncorresponding to a 500 nm thick section through the top of the GUV.\nThe formation of large clusters of curved triskelia on polymerization\nis evident but has no significant effect on the lipid distribution.\nScale bar: 10 μm. Figure 3 Confocal micrographs\nof GUV–GUV interfaces. (A) DNA triskelia\n(curved) are enriched at interfaces between GUVs both before and after\nthe addition of polymerization staples. Scale bar: 20 μm. (B)\nIntensity profiles along the lines indicated in A show the accumulation\nof both unpolymerized and polymerized triskelia at GUV–GUV\ninterfaces. We performed similar experiments\nto examine the assembly of fluorescently\nlabeled triskelia on SLBs using epifluorescence and total internal\nreflection fluorescence (TIRF) microscopy. In contrast to their behavior\non GUVs, dimer diffusion and triskelion polymerization on the SLB\nwere inhibited: fluorescence recovery after photobleaching (FRAP)\nconfirmed the presence of mobile lipids but immobile triskelion dimers\n( Figure 4 A). GUV–SLB\ninterfaces were formed by sedimentation of sucrose-containing GUVs\non an SLB washed in less-dense TE-MgCl 2 buffer. Triskelia\ndiffusing freely on the GUV were excluded from this interface and\naccumulated at its edge ( Figure 4 B i; see also Supporting Figure\nS15 i ). Triskelion dimers added after GUV–SLB interface\nformation also accumulated at the junction between the two bilayers\n( Figure 4 B iv and Supporting Figure S15 iv ). In each of these experiments,\nthe initial contact between a GUV and an SLB expands to form a planar,\ncircular, interface within seconds. However, when the triskelia bound\nto GUVs are polymerized by addition of polymerization staples before\nthe two bilayers were brought into contact, formation of the contact\ninterface is inhibited for at least several minutes ( Figure 4 B ii, Supporting\nFigures S15 ii and S16 ), When triskelion dimers are initially\nbound to the SLB rather than to the GUV, triskelia are partially excluded\nfrom the interface ( Figure 4 B iii and Supporting Figure S15 iii ). The difference in behavior between triskelia initially bound to\nthe GUV (unpolymerized triskelia escape the interface) and to the\nSLB (slowly diffusing triskelia remain at the interface) confirms\nthat the interactions between triskelia and the lipid bilayers are\ndirectional, consistent with stable insertion of cholesterol into\nthe first membrane encountered. Figure 4 DNA triskelia on supported lipid bilayers\nand at GUV–SLB\ninterfaces. (A) Fluorescence recovery after photobleaching (FRAP)\nanalysis of fluorescently labeled curved DNA triskelia (Alexa647,\nred) and lipid (Atto488, green) shows that the lipids in the SLB are\nmobile, whereas the DNA triskelia are immobile. (B) TIRF micrographs\nof fluorescently labeled triskelia at GUV–SLB interfaces: (i)\ntriskelia added to GUV before synapse formation are excluded from\nthe interface; (ii) triskelia added to GUV and polymerized before\nsynapse formation inhibit interface formation; (iii) triskelia added\nto SLB before synapse formation remain at the interface; (iv) triskelia\nadded after synapse formation are excluded from the interface. Triskelia\nused (see Supporting Figure S15 for the\nother type): curved (i, iii); flat (ii, (iv). Scale bars: 10 μm. Accumulation of triskelia at the interfaces between\nGUVs is consistent\nwith passive diffusion-mediated trapping. It implies that triskelia\nmediate an attractive interaction between the membranes by binding\nto both, an important mechanism in cell adhesion; 35 , 36 it is consistent with the observation that DNA origamis can bind\nto lipid membranes even in the absence of cholesterol anchors. 37 In contrast, the exclusion of triskelia from\nan SLB–GUV interface (where GUV-bound triskelia are able to\ndiffuse away from the interface) and the inhibition of interface formation\nby polymerized triskelia (whose escape is hindered) suggest that there\nis a repulsive interaction between GUV and SLB mediated\nby triskelia. The marked asymmetry between triskelion-mediated GUV–GUV\nand GUV–SLB interactions is intriguing. All membranes have\nthe same lipid composition: the most obvious difference between them\nis that the SLB is planar and closely bound to a glass surface, whereas\nthe GUV bilayer is constrained only by its natural elasticity and\nany residual difference in osmotic pressure across it. We hypothesize\nthat the stable incorporation of a triskelion array at a GUV–GUV\ninterface requires significant distortion of the membranes to conform\nto the far-from-planar triskelia: the SLB is incapable of this distortion."
} | 4,425 |
28405363 | PMC5383820 | pmc | 8,202 | {
"abstract": "DNA has been used to construct a wide variety of nanoscale molecular devices. Inspiration for such synthetic molecular machines is frequently drawn from protein motors, which are naturally occurring and ubiquitous. However, despite the fact that rotary motors such as ATP synthase and the bacterial flagellar motor play extremely important roles in nature, very few rotary devices have been constructed using DNA. This paper describes an experimental study of the putative mechanism of a rotary DNA nanomotor, which is based on strand displacement, the phenomenon that powers many synthetic linear DNA motors. Unlike other examples of rotary DNA machines, the device described here is designed to be capable of autonomous operation after it is triggered. The experimental results are consistent with operation of the motor as expected, and future work on an enhanced motor design may allow rotation to be observed at the single-molecule level. The rotary motor concept presented here has potential applications in molecular processing, DNA computing, biosensing and photonics.",
"conclusion": "4. Conclusion This paper has described an experimental study of the putative mechanism of a prototype synthetic rotary motor made from DNA. The motor was designed to be driven by strand displacement and to be capable of autonomous operation after the brake was released. Results were also reported from experiments in which the phenomenon underlying the motor mechanism was investigated extensively, building on previous studies of strand displacement in surface-immobilized DNA nanomachines. Ensembles of DNA rotary motors were examined using two complementary techniques: gel electrophoresis and QCM-D. The former was used to probe static motors in solution before and after operation, whereas the latter provided time-resolved data on surface-immobilized motors. Single-molecule measurements were beyond the scope of this study, but future work on an enhanced rotary motor design could involve a combination of the biophysical approaches of single-molecule fluorescence microscopy and structural imaging methods such as atomic force microscopy and cryoelectron microscopy. For the investigation described here, QCM-D and gel electrophoresis had a number of advantages. The data in this paper represent promising experimental results that were consistent with the operation of the rotary motor as designed, and further work could enable the properties of the motor or an enhanced model to be exploited for real-world applications, such as manipulation of molecules, translocation of cargo through a nanopore, nanoswimmer propulsion, biosensors, hybrid photonic-biomolecular systems [ 39 ], and biomolecular computing.",
"discussion": "3. Results and discussion 3.1. Sequential strand displacement in a linear complex To test the hypothesis that a single long invader could displace multiple strands from an immobilized target complex, the system depicted in figure 2 a was studied using QCM-D. The first stage of the experiment involved immobilization of a ‘capture complex’ via chemical bonds made between the gold surface and a thiol modification at one end of the complex. The capture complex itself comprised two strands, called CS and Block-3. As the capture complex was immobilized, the frequency decreased ( figure 2 b ), owing to the increase in the mass attached to the sensor. In the second step, the DNA molecule X was added, where X had the correct sequence to bind to the capture complex, leaving a long single-stranded overhang, as illustrated. Next, two more DNA strands were added, and these bound to the overhang, leaving only a short single-stranded toehold. Further decreases in frequency were observed ( figure 2 b ), corresponding to additional increases in mass. The final step was the addition of an invading strand, which bound to the toehold and displaced all three of the blocking strands from the long strand in the immobilized complex, resulting in loss of the strands X, Block-2 and Block-1 from the surface. The frequency change observed during the displacement reaction was approximately 55% of that measured during binding of the strands X, Block-2 and Block-1. This implies that the sequential strand displacement process was not 100% efficient, as expected—surface-specific phenomena interfere with processes that occur in immobilized molecules, and decrease the efficiency of the reaction. However, the final baseline was higher than the value observed (at around 70–90 min) for the plateau corresponding to the construct comprising CS, Block-3 and X, which would not be possible if only Block-1 had been displaced from the immobilized constructs. The data were therefore consistent with the hypothesis that a single invader can displace multiple targets within the same surface-immobilized molecule, justifying the subsequent experiments and development of a motor based on this concept. 3.2. Sequential strand displacement in a folded structure QCM-D experiments were also performed on a geometrically constrained structure, formed by using two 23 nt oligonucleotides to fold a 73 nt oligonucleotide (T) into a triangular configuration ( figure 3 a ). The shorter strands were named ‘staples’, following the convention of DNA origami [ 5 ]. Surface-immobilization of this structure was achieved via hybridization of the triangle strand with a pre-immobilized capture strand (CS), as shown in figure 3 a .\n Figure 3. Observing strand displacement in a folded nanostructure. ( a ) Schematic diagram of the triangle structure. The long strand T had six domains (t, d, c, b, a, RC-CS), separated with T triplets where necessary. The long strand T was folded into a triangular conformation by means of two shorter strands (bd-staple) and (ac-staple) and the triangle could be immobilized through attachment to the CS strand. ( b ) QCM-D data which illustrate immobilization of strand CS and either isothermal on-surface assembly of triangles ( T + S ) or attachment of pre-folded triangles (F(T)). Control strands were used to test whether the triangle had assembled correctly. ( c ) Strand displacement in the nanotriangles. The immobilized triangles were exposed to unfolded or folded reverse complements, which bound to their targets and induced displacement. Red lines: smoothed data (50 point adjacent average filter). ( d ) Dissipation changes as a function of frequency changes for the time period shown in part ( c ). A 20 point adjacent-averaging smoothing filter was used. Figure 3 b presents the overtone-normalized frequency shifts observed with QCM-D during the assembly of a triangle in situ on the surface (in part 1) and immobilization of pre-folded triangles (in parts 2 and 3), followed by the application of control strands designed to bind to free domains in the immobilized structures. The bottom two panels of figure 3 b show the application of the same molecules in parallel. The first phase of all three experiments performed was the immobilization of CS alone, corresponding to a frequency decrease of 3–5 Hz. The variation observed is attributable to uncontrollable minor differences between the sensors. The top panel of figure 3 b relates to the experiment in which the triangle was folded isothermally on the surface. After immobilization of CS, the strand T was supplied, leading to a decrease in frequency as T hybridized to CS. For rigid layers immobilized on the surface of a QCM-D sensor, the Sauerbrey equation (commonly used in analysis of QCM data [ 23 ]) predicts that the overtone-normalized frequency shift should be directly proportional to the change in the surface mass density. However, it will be seen that Δf 13 /13 did not scale with the molecular mass of the oligonucleotides added; the strand T contains 73 nt and its addition induced a shift of approximately 8.5 Hz, which is just over half of what would be expected based on the shift of approximately 3.5 Hz observed for the 16 nt strand CS. This implies either that not all of the CS oligonucleotides captured a T or that the Sauerbrey equation does not apply here, which would be unsurprising owing to the non-rigid nature of the surface-immobilized layer of biomolecules. The addition of the staple strands resulted in a further frequency decrease, corresponding to an additional increase in immobilized mass. This was associated with an increase in dissipation, indicative of conformational change. Control experiments (described below) suggested that no triangle domains were left single-stranded and no staples were half-bound, consistent with proper folding of the triangle. These results are interesting in their own right because they suggest that it is possible to assemble a complex structure on the surface isothermally. It is likely that the staples cooperate during folding, because the binding of one staple will bring into closer proximity the binding sites for the next, which will then be able to bind more easily [ 25 , 26 ]. After folding, the dissipation values were comparatively high, approaching two units for the CS-T complex, and this implies that the Sauerbrey equation should not be used here because the layer is viscoelastic rather than rigid [ 27 ]. In fact, the shift in frequency resulting from addition to the staples was higher than would be predicted from the Sauerbrey equation. If a linear relationship existed between frequency changes and mass, the results would imply that 1.25 copies of each staple had bound to each T molecule, and because that is impossible, this result represents a further indication that the Sauerbrey equation does not apply in this regime. This is to be expected for a layer of biological molecules, which have a complex three-dimensional structure and dissipate acoustic energy, displaying viscoelastic behaviour. The data presented in the bottom two panels of figure 3 b show immobilization of a pre-folded triangle by hybridization with CS. The frequency shift will be seen to be greater in magnitude than the combined shifts induced by application of T and S in the top panel, suggesting that the use of pre-folded triangles may lead to a higher-density molecular layer. In all cases, it was necessary to test whether the triangle structures had folded completely. Two primary folding ‘errors’ are conceivable. Either a staple could be ‘half-bound’, where one domain is attached to the triangle and the other is single-stranded, or it could be missing completely. In all three experiments, to test for these errors, control strands were applied after the supposed formation of a surface-immobilized layer of folded triangles ( figure 3 b , all graphs, section marked ‘controls’). The sensor was supplied in turn with a series of control samples containing strands that should either bind to a free domain of a staple, or a free domain of the T strand. In all cases, minimal frequency shifts were observed, attributable to changes in the mixture of molecules passing over the sensors rather than the binding of any strand to the surface-immobilized constructs. This suggested that all structures were indeed completely folded, including the triangle folded isothermally in situ on the surface. Use of the online analysis package NUPACK [ 28 ] confirmed that 96% or more of the control strands should hybridize to their target domains, if available, and although this simulation uses the energy parameters established for solution-phase reactions, it provides a good indication that this control experiment is sound. Strand T* was complementary to strand T, and the exposure of folded triangles to T* resulted in some unfolding of T ( figure 3 c , left and centre panels). The data show a drop in frequency that corresponds to binding of T*, followed by an increase as staples are displaced and T unfolds. The same effect was observed for the application of T* strands that had been folded up into triangles in the same way as T. In all cases, the loss of mass resulting from unfolding was significant although comparatively small, and the timescale was long (many tens of minutes). The amount of mass lost appeared to vary between the three cases. The data presented here do indicate that strand displacement is possible in a structure that is geometrically constrained, an important result for the rotary motor concept. However, the results also suggest that the efficiency of the reaction can be quite low. This may be associated with molecular crowding effects, which are known to have an effect on strand displacement on surfaces [ 14 ]. Importantly, it was necessary for T* to displace at least three staple domains before any staples could be lost, and this means that at least some of the triangles were partly or entirely unfolded by T*. It would not have been possible for displacement of any domain to be initiated prematurely by the interaction between AAA and TTT domains in the T and T* strands, because AT-rich toeholds 3 nt in length do not permit displacement in surface-immobilized DNA machines [ 14 ]. Figure 3 d shows that dissipation and frequency changes are not perfectly correlated. This is an interesting result because it implies that the structure of the molecules is changing independently of the changes in mass, which is exactly what would be expected from unfolding of triangles. It has been demonstrated that the acoustic ratio ( ΔD / Δf ) can be used as a measure of the conformation of DNA molecules on surfaces [ 24 ], but it is difficult to use this a priori without any reference structures. 3.3. Rotary motor The rotary motor prototype is shown in figures 1 c and 4 a . It was assembled in stages in solution from a number of oligonucleotides, as described in the electronic supplementary material. It consisted of two squares (A and B), made from ‘tape’ strands and staples that held the tape strands in place. The design principles are the same as for the triangle presented in the previous section. The two squares were linked, as shown. The domains adjacent to the linked sections were not complementary and were hybridized to blocking strands. Release of the blocking strands allowed the hybridized domains to act as a remote toehold for initiation of strand displacement, driving rotation of the motor. Each tape actually consisted of a series of three oligonucleotides, joined together by hybridization of connecting domains that protruded from the wheels like spokes. For clarity, these spokes are not shown in the figures because they are irrelevant to the reaction mechanism, but they are illustrated in the electronic supplementary material. The spokes remained double-stranded and did not change partners throughout motor operation. They were not expected to interfere with rotation because each spoke could pass over or around the other wheel.\n Figure 4. The rotary motor. ( a ) Schematic diagram of the motor which shows how it was constructed. The staple strand names begin with ‘St’ or ‘Str’. The tape for square A is made from three strands (SquareA_1, SquareA_2 and SquareA_3) as is the tape for square B (SquareB_1 or SquareB_1mmwA, SquareB_2 and SquareB_3). In each square, these strands are connected through connection domains (not shown here) that protrude like spokes from the motor. As described in the text, the connection domains are irrelevant to the motor mechanism because they remain inert throughout operation, but they are shown explicitly in the electronic supplementary material. Apart from the connection domains and the domains marked in red, the tapes are complementary to each other. ( b ) The structure used to immobilize a square. ( c ) Agarose gel which shows the assembly of the surface connector illustrated in ( b ), the two folded squares and an unfolded square (made from only the tape strands, without any staples). ( d ) Polyacrylamide gel which shows the most important bands for the indicated samples: unfolded square of type B, folded squares of both types and assembled motor—‘before’ and ‘after’ operation. ( e ) QCM-D data which show strand displacement in motor components or the operation of the motor. Note that the graphs have different y -axis scales but all show the same time window. The labels at the bottom of the figure show what structure was immobilized on the surface, and the coloured bands above the plots indicate what was applied to the sensors, where ‘Unbl’ refers to an unblocking strand. The sections of the plot shaded in yellow correspond to mass loss from the surface, resulting from displacement. Binding and unblocking events are marked. Red lines: smoothed data (100-point adjacent average filter). ( c – e ) Samples are described using simplified sketches and abbreviations—F denotes ‘folded’, U denotes ‘unfolded’, open square denotes ‘square’, ‘Unbl’ denotes ‘unblocking strand’. The squares were immobilized on a surface via the structure shown in figure 4 b , which includes the thiolated strand CS. The assembly of the tapes from their constituent parts, the surface-immobilization connector and the squares themselves was confirmed prior to immobilization using agarose gel electrophoresis ( figure 4 c ). It was not feasible to test assembly using QCM-D in the same way as for the triangle, because, in this case, the incorporation of one of the staples was necessary for immobilization, and the greater complexity of the design would have significantly increased the required number of control strands. The surface-immobilization connector produces one clear band in the gel, which indicates that this structure has assembled correctly. Similarly, there is one dominant band in the lane corresponding to the unfolded square (assembled tape), although some smearing of the band is apparent. The squares were folded as described in the electronic supplementary material. The staples were present in excess, which means that the presence of the lower band in the lanes marked F□A and F□B does not imply poor assembly. However, it is not possible to distinguish between the unfolded and folded constructs using agarose gel electrophoresis, and a polyacrylamide gel was therefore used to provide higher resolution ( figure 4 d ), showing both assembly and operation of the motor. In this gel, there is a distinct difference between the position of the band identified as unfolded square B (lane 5), and that identified as folded square B (lane 6), indicating that folding was successful. Folded square A (lane 7) runs slightly further than folded square B, owing to the asymmetry in the positions of the spokes mentioned above. Lane 8 shows the motor before operation, obtained by hybridization of folded square A and folded square B. At this stage, the brake of the motor was applied, which means that the blocking strands were attached, preventing the unrolling of the tapes. The highest molecular weight band in lane 8 is postulated to be the assembled motor. Some of the squares did not connect with another square. Lane 9 shows the motor after incubation with the unblocking strands, which was intended to remove ‘SqBlockA’ and ‘SqBlockBmmwA’ by displacement. This was expected to release the brake, allowing strand displacement to proceed. Comparing lanes 8 and 9 reveals that the unblocking process induced a significant structural rearrangement in the motor, which is the expected result. However, gel electrophoresis is not conclusive and provided information only on static structures. By contrast, the subsequent QCM-D experiments allowed structural changes to be measured in real time. In addition to probing the operation of the motor, these measurements provided a further indication that folding had been successful, because the observation of any changes corresponding to operation of the motor or unfolding of squares would have been highly unlikely if folding had been incomplete. QCM-D experiments were performed to investigate motor operation in more detail. In contrast to the previous QCM-D measurements, DNA was co-immobilized with the backfilling agent mercaptohexanol, which competes with the thiolated DNA for access to the surface. The aim of this was to reduce the density of machines on the surface and consequently decrease any inhibitory effects arising from intermolecular interactions. The results of the QCM-D experiments are shown in figure 4 e . The first panel shows the effect of supplying an unfolded square of type B to an immobilized square of type A. This is directly analogous to the experiments in which T* was applied to folded triangles. In the same way, the binding of the unfolded square gave rise to a decrease in frequency. This was followed shortly thereafter by a slow increase, suggesting that the immobilized square unfolded and lost mass. The second panel shows the result of a similar experiment, but in this case the immobilized square was initially blocked with a blocking strand, and was unblocked before the unfolded square was applied. Removal of the blocking strand is expected to cause a small reduction in mass, and the data do show a corresponding frequency increase. The result of applying the unfolded square was qualitatively similar to the case in which the immobilized square had not been blocked, but the frequency shift corresponding to unfolding was small. This may be attributable to incomplete unblocking in a surface-immobilized molecular layer consisting of only one type of square. To investigate whether one folded square could induce unfolding in another, a fully folded square of type B was applied to an immobilized folded square of type A. This resulted in a frequency decrease, followed by a significantly delayed frequency increase. The former was attributable to binding of the square, whereas the latter was consistent with unfolding of the squares and release of staples. These experiments indicated that strand displacement reactions occurred as intended in the square constructs. The full motor was assembled by the connection of two squares, and immobilized on the surface with mercaptohexanol. Unblocking strands were supplied to release the brake, and a slow frequency increase was observed thereafter, as expected to result from strand displacement, rotation and release of waste strands. This result is therefore consistent with the correct operation of the motor on the surface. The data suggest that rotation is initiated before the second unblocking strand has been supplied. Alternative displacement pathways may exist, and it is not clear what effect this could have on the operation of the motor. It is possible that these pathways could be eliminated with an alternative motor design based on components made using DNA origami methods, as discussed below. 3.4. Further work For further study, single-molecule biophysical experiments could be used to directly observe the dynamic functional operation of prototype rotary machines, but these measurements are beyond the scope of this paper, which represents a proof-of-concept study. Further work on the motor would comprise the development of an enhanced model (as described below), which would enable the use of techniques such as atomic force microscopy to visualize intermediate states. Direct tracking of rotation could be achieved using time-resolved super-resolution microscopy [ 29 ], which would involve tracking the position of a single fluorescent molecule (or group of them). Single-molecule fluorescence microscopy on this system poses significant challenges, such as: attaching the fluorescent label to the motor, compensating for microscope drift over the long timescale of this experiment, correcting for two-dimensional diffusion of the motor at the end of its tether and avoidance of excessive fluorophore photobleaching on the timescale of the rotation. Pursuit of such challenging single-molecule experiments could form the basis of a subsequent study, following on from the work presented here. In future studies, the design of the rotary motor could also be enhanced, in three main respects. First, in the present implementation, the motor is only capable of performing 1.5 rotations, and this could be increased by a straightforward extension of the design. Second, the operation of the motor leads to disassembly of the motor, and the structure remaining on the surface is comparatively insubstantial. Third, it would be difficult to use the existing design in a single-molecule experiment because it is difficult to identify where a fluorescent label could be placed. These last two issues could be addressed simultaneously by constructing the wheels of the motor using DNA origami, a very popular technique for assembling DNA nanostructures [ 5 , 30 , 31 ]. Recent studies have explored the DNA origami folding process [ 25 , 26 , 32 , 33 ], and the results will allow more sophisticated origami structures to be assembled [ 34 ], including devices such as the enhanced motor suggested here. Alternatively, other DNA nanostructure assembly methods could be used [ 35 – 38 ]."
} | 6,239 |
36135298 | PMC9498466 | pmc | 8,203 | {
"abstract": "Swelling experiments and NMR spectroscopy were combined to study effect of various stimuli on the behavior of hydrogels with a single- and double-network (DN) structure composed of poly( N,N′ -diethylacrylamide) and polyacrylamide (PAAm). The sensitivity to stimuli in the DN hydrogel was found to be significantly affected by the introduction of the second component and the formation of the double network. The interpenetrating structure in the DN hydrogel causes the units of the component, which is insensitive to the given stimulus in the form of the single network (SN) hydrogel, to be partially formed as globular structures in DN hydrogel. Due to the hydrophilic PAAm groups, temperature- and salt-induced changes in the deswelling of the DN hydrogel are less intensive and gradual compared to those of the SN hydrogel. The swelling ratio of the DN hydrogel shows a significant decrease in the dependence on the acetone content in acetone–water mixtures. A certain portion of the solvent molecules bound in the globular structures was established from the measurements of the 1 H NMR spin–spin relaxation times T 2 for the studied DN hydrogel. The time-dependent deswelling and reswelling kinetics showed a two-step profile, corresponding to the solvent molecules being released and absorbed during two processes with different characteristic times.",
"conclusion": "3. Conclusions The influence of various stimuli on the behavior of hydrogels with a single- and double-network structure composed of PDEAAm and PAAm was investigated. Swelling measurements and NMR spectroscopy were combined to provide information about changes in hydrogels on macroscopic and molecular scales. The sensitivity to stimuli of the DN hydrogel was found to be significantly affected by the introduction of the second component and the formation of the double network. The interpenetrating structure in the DN hydrogel causes the units of the component, which is insensitive to the given stimulus in the form of the SN hydrogel, to be partially formed as globular structures in the DN hydrogel. The SN-D hydrogel composed of PDEAAm shows distinct temperature and salt responsiveness. Due to the hydrophilic PAAm groups, temperature- and salt-induced changes in the deswelling of the DN-DA hydrogel are less intensive and gradual compared to the SN-D hydrogel. On the other hand, the swelling ratio of the DN-DA hydrogel shows a significant decrease in dependence on the acetone content in acetone–water mixtures. This is certainly caused mainly by the presence of acetone-sensitive PAAm units, but as it was shown from the NMR experiments, 50% of acetone-insensitive PDEAAm units also contribute to the collapsed structures, which may affect the deswelling extent of the DN-DA hydrogel in water–acetone mixtures. A certain portion of water (HDO) and acetone molecules bound in globular structures was established from the measurements of the 1 H NMR spin–spin relaxation times T 2 for the studied DN hydrogel. In water–acetone mixtures with a high content of acetone, the spin–spin relaxation times T 2 for bound solvent molecules are up to three orders of magnitude lower than the relaxation time of the free molecules, signifying that the collapsed polymer structures of the DN hydrogel are very compact and rigid, and the solvent molecules that are bound to/in them are similarly limited in their mobility. The time-dependent deswelling and reswelling kinetics showed a two-step profile, corresponding to the solvent molecules being released and absorbed during two processes with different characteristic times. The response to the given stimuli was found to be a key factor for the kinetics and the fastest deswelling and swelling processes were detected for water–acetone mixtures with a higher acetone content.",
"introduction": "1. Introduction Stimuli-responsive polymer hydrogels that are able to respond to the changes in temperature [ 1 ], pH [ 2 ], humidity [ 3 ], light [ 4 ], specific ions or molecules [ 5 ], electrical fields [ 6 ], solvent [ 7 ] and ionic strength [ 8 ] have been extensively studied due to potential applications in the areas of drug delivery [ 9 ], microlenses [ 10 ], sensors [ 1 ] and artificial organs [ 11 ]. Among the various stimuli, temperature is the most studied in the field of stimuli-responsive polymers because of the key role of temperature in nature [ 12 , 13 , 14 , 15 , 16 , 17 ]. The thermosensitivity of polymer hydrogels is associated with a variable balance between different types of interactions, especially hydrogen bonds and hydrophobic interactions. At temperatures below the volume phase transition temperature, hydrogels absorb water to reach the swollen state and, above the transition temperature, they release water and shrink. On a molecular level, the volume phase transition (collapse) in crosslinked hydrogels is assumed to be a macroscopic manifestation of a coil–globule transition, as was shown for poly( N -isopropylacrylamide) (PNIPAAm) in water by light scattering [ 18 ]. Subsequently, temperature as a stimuli was found for other acrylamide-based hydrogels, such as poly( N,N′ -diethylacrylamide) (PDEAAm) [ 19 ] or poly( N -isopropylmethacrylamide) [ 20 ]. The collapse of hydrogels can be induced not only by temperature but also by the composition of the solvent. The coexistence of two polymer phases differing in the conformation and concentration of chains in the swollen polymer network was predicted theoretically and experimentally proven on polyacrylamide (PAAm) networks swollen in acetone–water mixtures [ 7 , 21 , 22 ]. It was later found that the presence of charges on the PAAm chains plays a decisive role in the formation of a phase transition (collapse) of PAAm networks, leading to a jump-wise change not only in swelling, but also in mechanical properties [ 23 ]. The change in swelling properties with solvent composition was also found for PAAm hydrogels in water/alcohols mixtures [ 24 ]. Most thermosensitive hydrogels show a co-nonsolvency effect in mixtures of two good solvents, i.e., they are swollen in both pure solvents but shrink into a compact form in mixtures of these solvents. The co-nonsolvency phenomenon was studied in particular in PNIPAAm hydrogels in various water–organic solvent mixtures [ 25 , 26 ]. The conformational transitions of polymers can be induced by the presence of salts and their concentrations. A salt-induced phase transition in a hydrolyzed ionic PAAm hydrogel was detected in a water–organic solvent mixture [ 5 ]. The effect of the phase transition in the charged PNIPAAm in aqueous NaCl solutions on swelling and mechanical properties was described [ 27 ]. The influence of various salts was studied on the deswelling isotherms of thin films of photo-crosslinked PNIPAAm and PDEAAm [ 28 ]. It was reported that introducing a double network (DN) structure for various combinations of polymers is an effective approach to prepare stimuli-responsive hydrogels. Compared to conventional hydrogels or single network (SN) hydrogels, DN hydrogels are associated with improved mechanical properties as well as a high degree of swelling. DN hydrogels have an interpenetrating polymer network structure and the properties of these two networks, such as network density, rigidity, molecular weight and crosslinking density, exist in sharp contrast [ 29 ]. The enhanced mechanical properties of DN hydrogels are related to the asymmetric combination of two networks, when upon deformation, internal fractures in the first network are formed and act as additional crosslinkers [ 30 ]. DN stimuli-responsive hydrogels have been studied primarily with regard to their sensitivity to temperature and pH. DN PNIPAAm/PNIPAAm hydrogels containing inorganic polysiloxane nanoparticles [ 31 ] or comprising an ionized first network with electrostatic co-monomer [ 32 ] were investigated with regard to the influence of the hydrogel composition on the volume phase transition, morphology, equilibrium swelling, deswelling–reswelling kinetics and mechanical properties. A pH- and temperature-responsive DN hydrogel based on PNIPAAm and polyacrylic acid using graphene oxide as an additive were synthesized and the influence of additive and acid contents on various physical properties was studied [ 33 , 34 ]. Highly stretchable and tough thermo-responsive DN hydrogels composed of poly(vinyl alcohol)-borax and poly (AAM-co-NIPAAM) were characterized using FTIR spectroscopy and studied for their mechanical properties, thermal behavior, swelling property and thermo-responsive behavior [ 35 ]. Recently, we investigated the temperature response of DN hydrogels based on thermoresponsive PNIPAAm and PDEAAm [ 36 , 37 ]. This study showed that the temperature response of the studied DN hydrogels is tunable by the network crosslinking density. In present work, we investigate SN and DN hydrogels composed of PDEAAm and PAAm with regard to sensitivity to various stimuli. Using swelling characteristics and NMR spectroscopy, the response to temperature, the presence and concentration of NaCl salt, and the composition of the water–acetone mixtures was studied. Macroscopic detection using swelling experiments was combined with NMR spectroscopy, which, especially in the case of two-component hydrogels, allows us to observe the behavior of each component separately on the molecular scale. Using NMR relaxation experiments, the different dynamic state of solvent molecules was detected. The swelling and deswelling kinetic experiments of DN hydrogel in various solutions were performed and the corresponding kinetic time parameters were determined.",
"discussion": "2. Results and Discussion 2.1. Hydrogels Synthesis The details of the preparation of SN and DN hydrogels were reported previously [ 37 ]. Briefly, first, the SN-D hydrogel was prepared by the redox polymerization of aqueous solutions containing a monomer, N,N′-diethylacrylamide (DEAAm) (c DEAAm = 127.2 g∙L −1 ), a crosslinking agent, N,N′-methylenebisacrylamide (MBAAm) (c MBAAm = 1.5 g∙L −1 ), an initiator, ammonium persulfate (APS) (c APS = 1 g∙L −1 ), and a catalyst, N,N,N′,N′-tetramethylenediamine (TEMED) (c TEMED = 15 g∙L −1 ). Afterwards, the DN-DA hydrogel was prepared from the specimen cut from SN-D hydrogels swollen to equilibrium in a large volume of aqueous solutions containing a second monomer, acrylamide (AAm) (c AAm = 142.2 g∙L −1 ), a crosslinking agent (MBAAm) (c MBAAm = 0.15 g∙L −1 ), a photoinitiator, and 2- oxoglutaric acid (OGA) (c OGA = 0.15 g∙L −1 ) by UV irradiation between two glassy plates separated by a spacer of silicone rubber. After the synthesis, the hydrogel samples were thoroughly washed with a large amount of distilled water to remove residual unreacted reagents. A schematic illustration of the preparation of the DN hydrogel PDEAAm/PAAm and the chemical structures of PDEAAm and PAAm is presented in Scheme 1 . 2.2. Effect of the Temperature on the Hydrogels’ Behavior The temperature dependences of the swelling ratios of SN-D and DN-DA hydrogels are shown in Figure 1 . Compared to the SN-D hydrogel, the transition region of DN-DA hydrogel is less steep, slightly shifted to higher temperatures, and the DN-DA hydrogel deswells less during temperature increases. The DN-DA hydrogel is thus less sensitive to temperature, which is caused by the hydrophilic chains of PAAm component. As we have shown in [ 37 ], a significant amount of water molecules interacting with PAAm units were detected in the collapsed DN-DA hydrogel structures at elevated temperatures, resulting in the high swelling ratio of DN PDEAAm/PAAm hydrogels. Figure 2 shows the high-resolution 1 H NMR spectra for the SN-D and DN-DA hydrogels detected under the same instrumental conditions at two temperatures (25 and 66 °C). The assignment of resonances to various proton types is shown directly in the spectra measured at 25 °C and it is the following: water signal (peak A), CH 2 group of PDEAAm (peak B), backbone chain groups CH of PDEAAm and PAAm (peak C and C′, respectively), the backbone chain groups CH 2 of PDEAAm and PAAm (peak D) and CH 3 group of PDEAAm (peak E). The chemical structures of PDEAAm and PAAm with the assigned 1 H NMR signals are shown in Scheme 1 . At lower temperatures, the hydrogels are swollen in water and polymer chains are flexible and the NMR signals of all polymer units are clearly resolved. As it is seen from the 1 H NMR spectra measured at a higher temperature, the signals B, C and E of the PDEAAm component are markedly reduced in their integrated intensities. Evidently, at elevated temperatures, the mobility of the PDEAAm units in rather compact globular structures is reduced and the corresponding NMR lines become too broad and undetectable in high-resolution NMR spectra [ 38 , 39 ]. On the other hand, the PAAm signals C′ and D in the spectra of the DN hydrogel ( Figure 2 b) change less with temperature. Equation (3) was used to calculate the collapsed p -fraction of units with significantly reduced mobility. For I 0 , we used the values of the integrated intensities obtained at 25 °C and the correction for the fundamental decrease in the integrated intensity with increasing temperature as 1/ T was included [ 37 , 39 ]. Figure 3 shows the temperature dependences of the p -fraction as obtained for the methylene CH 2 signals in the PDEAAm units (signal B) and for the CH signals in the PAAm units (signal C′) in the SN-D and DN-DA hydrogels. It is seen that the maximum value of p -fraction p max as detected for the polymer units of the temperature-sensitive PDEAAm in the SN-D and DN-DA hydrogels are equal to 1, which means that all PDEAAm units are immobilized and involved in collapsed globular structures. In comparison with the swelling experiments, the NMR-determined transition is much more steeper as these methods detect other processes during the temperature-induced transition in hydrogels. The time-consuming release of solvent molecules as detected by the swelling experiments results in a broad transition interval. Contrarily, NMR spectroscopy follows the relatively fast aggregation of polymer units, which leads to a sharp change in the dependence of the p -fraction on temperature [ 37 ]. The relatively high value of p max = 1 as detected for the PDEAAm units of the DN-DA hydrogel could be a consequence of formation of heterogeneous structure in DN hydrogels as we have already reported for PNIPAm/PAAm and PDEAAm/PAAm hydrogels [ 36 , 37 ]. It has been shown that the agglomerates are formed during the formation of the DN structure, and the PDEAAm units in these structures are limited in their mobility and therefore do not contribute to high-resolution NMR spectra even at temperatures below the transition. The PDEAAm units, which remain mobile at lower temperatures, collapse upon subsequent heating and the integrated intensity corresponding to their NMR signals is thus reduced. Since NMR spectroscopy follows the change in the hydrated state mainly of these PDEAAm units, a relatively intense transition with the high maximum values of p -fraction is detected in contrast to a small temperature change in the swelling ratio ( Figure 1 ). Interestingly enough, it is evident from Figure 3 that p -fraction as detected for the CH signal of the PAAm units in the DN-DA hydrogel increases with temperature to the maximum value p max ≅ 0.3. This means that approximately 30% of temperature-insensitive PAAm units are restricted in their mobility at higher temperatures. The increase in the p -fraction in temperature for the PAAm units is more gradual than for the PDEAAm units and it is probably connected with the process of water release, when thermodynamic and interaction conditions for conformational change occur for some PAAm units of the dehydrated hydrogel and these PAAm units gradually collapse into compact globular-like structures. Previously, we did not observe such behavior in IPN networks consisting of PDEAAm and PAAm, where the network density of the first and second components was very low and the NMR signals of PAAm practically did not change with temperature, which implies that virtually all AAm units showed high mobility even at elevated temperatures [ 40 ]. It is therefore possible that the densely crosslinked and temperature-sensitive PDEAAm component causes a part of the hydrophilic units of PAAm to pack into dehydrated structures, which leads to a limitation in their mobility. 2.3. Effect of Solvent Composition on the Hydrogels’ Behavior In Figure 4 , the swelling ratio of the SN-D and DN-DA hydrogels is shown as a function of the acetone content in water–acetone mixtures for the two temperatures of 25 and 45 °C. At room temperature, the swelling characteristics of the SN-D hydrogel composed of PDEAAm is not practically affected by the presence of acetone in mixture and the swelling ratio decreases slightly with increasing acetone content. At the higher temperature of 45 °C, a visible increase in the swelling ratio is observed ( Figure 4 a). For pure water and 20 vol.% acetone content, obviously the temperature effect will cause the collapse of the hydrogel and the swelling ratio thus shows low values. The increasing value of the swelling ratio with increasing acetone content is obviously related to the co-nonsolvency behavior of the PDEAAm hydrogels, where the addition of acetone to water will cause higher swelling ratios. According to our findings, the co-nonsolvency behavior of the PDEAAm hydrogels in a mixed water–acetone solvent has not been investigated, but the phase transition of PNIPAm polymer in water–acetone solutions was studied in [ 41 ]. Assuming that the PDEAAm polymer has a similar con-onsolvency behavior as the PNIPAm polymer, then apparently at higher temperatures and at the acetone content higher than 20 vol.%, the hydrogel enters a swollen phase. In comparison with the SN-D hydrogel, the DN-DA hydrogel formed by the double network PDEAAm/PAAm shows a completely different dependence of the swelling ratio on the acetone content. Swelling ratios measured at both temperatures show a significant decrease in the region of 30–60 vol.% of acetone ( Figure 4 b). It is clear that the swelling behavior of the DN-DA hydrogel is mainly determined by the PAAm component, which is known to be sensitive to the water–acetone solvent composition [ 7 , 21 ]. As shown in Figure 4 b, the DN-DA hydrogel has high values of equilibrium swelling in pure water. This behavior suggests that the attractive interactions between the polymer chain and the water molecules dominate over the attractive interactions between the polymer chains. In water–acetone mixtures, molecules of these solvents have an attractive interaction, leading to an increase in the free energy for polymer–polymer contact and induce the collapse of the polymer network [ 24 ]. Using Equation (3), we calculated the p -fraction for the PDEAAm and PAAm NMR signals dependent on the acetone content. For I 0 , we took values based on integrated intensities as obtained for the hydrogels in D 2 O solution at 25 °C. As it is shown in Figure 5 , the SN-D spectra of the PDEAAm hydrogel did not change with increasing the content of acetone, leading to a p -value of 0–0.1 regardless of the water–acetone content. On the other hand, the PAAm signals in the DN-DA hydrogel show a significant increase in the p -fraction dependent on the acetone content; the p -fraction value varies from 0 for pure water to 1 for pure acetone. Both findings are consistent with the results of the swelling experiments ( Figure 4 ). What is interesting and noteworthy is the third dependence in Figure 5 for the PDEAAm component signals in the hydrogel DN-DA. PDEAAm is insensitive to the composition of the water–acetone content when in the SN-D hydrogel (empty blue squares), but in the DN-DA hydrogel, the p -fraction increases from p max = 0 for pure water to p max ≅ 0.5 for pure acetone (full blue squares). The increase in p -fraction for the PDEAAm component is delayed when compared to the PAAm component, starting with up to 60 vol.% acetone. This behavior is similar to that found for the temperature dependence; the units of the component that is insensitive to the stimulus (if it is in the single hydrogel version) partially packed into immobile structures, but need a stronger stimulus to do so. 2.4. Effect of Salt on the Hydrogels’ Behavior The equilibrium swelling ratios of the SN-D and DN-DA hydrogels as a function of NaCl salt concentrations are shown in Figure 6 . It has been noticed that the swelling ratio of the SN-D hydrogel at both temperatures in different NaCl solutions decreased strongly with the increase in salt concentration from 0.01 to 0.05 M and reaches very low values ≅0.2 for salt concentration 3 M and higher ( Figure 6 a). This behavior is consistent with the previously established effect of NaCl on the deswelling of the PDEAAm hydrogels [ 28 ]. At a higher temperature, the swelling ratio is more suppressed due to the combination of the two stimuli to which the PDEAAm hydrogel is sensitive. As previously found, nonionic PAAm hydrogels induce a far weaker salt-induced swelling change compared to the ionic PAAm hydrogels [ 5 ]. As it is seen in Figure 6 b, the introduction of PAAm as the second component in the DN-DA hydrogel leads to a very gradual decrease in swelling to a value ≅5. At the lower temperature of 25 °C and a low salt content, there is a somewhat sharper decrease in swelling and it is probably mainly the PDEAAm units that are affected by salt. For salt concentrations higher than 1.5 M, the swelling drop is slower and the PAAm units are probably gradually affected, while no effect of the temperature is observed. Similar to the dependence of the NMR spectra on the temperature and acetone content, we determined the p -fraction for different molar concentrations of NaCl using Equation (3) and, for I 0 , we took values based on integrated intensities as obtained for the hydrogels in D 2 O solution at 25 °C. Figure 7 shows the dependence of the p -fraction on the NaCl concentration as determined for the SN-D and DN-DA hydrogels at 25 °C. For the SN-D hydrogel, the p -fraction grows very fast in the NaCl concentration region 0–2 M and it reaches a maximum value p max = 1, which means that all PDEAAm units are packed in collapsed structures. The PDEAAm component in the hydrogel DN-DA shows a similar increase in the p -fraction and reaches a maximum value for the NaCl concentration of 3 M. Both findings fully correspond to the results obtained from the swelling experiments ( Figure 5 ). The increase in the p -factor, and thus the decrease in the intensity of the high-resolution NMR signals of the PAAm groups in the DN-DA hydrogel, are much slower in the dependence of the NaCl concentration and the maximum value of the p -fraction only reaches a value of 0.5. This means that roughly 50% of the PAAm units in the double network are unaffected by NaCl and prefer to interact with water molecules. Therefore, the swelling ratio of the DN-DA hydrogel for the maximum concentration of NaCl reaches a relatively high value ( Figure 6 b). 2.5. Effect of All Stimuli on the Hydrogels’ Behavior Figure 8 displays the swelling ratios that were detected for the SN-D and DN-DA hydrogels in different environments, i.e., in pure water and acetone, and for a maximum concentration of NaCl = 6 M at two different temperatures. The influence of various stimuli on the swelling behavior of hydrogels is thus clearly demonstrated. The SN-D hydrogel composed of PDEAAm shows temperature and salt responsiveness. Compared to the hydrogel swollen in pure water at room temperature, the swelling ratio in the salt solution decreases almost 10 times, and this change is much more pronounced than when the temperature is increased. This may be related to the earlier finding that the conformation of PDEAAm is more compact in the presence of NaCl than that in the presence of a salt-free solution [ 42 ]. The sensitivity to stimuli in the DN-DA hydrogel is significantly affected by the introduction of the second component and the formation of the double network. Due to the hydrophilic PAAm groups, the extent of deswelling after temperature increase is reduced. On the other hand, the DN-DA hydrogel has a 15 times smaller swelling ratio in pure acetone compared to water. This is certainly caused mainly by the presence of acetone-sensitive PAAm units, but due to the interpenetrating structure of the DN network, 50% of acetone-insensitive PDEAAm units also contribute to the collapsed structures ( Figure 5 ), which may affect the deswelling extent of the DN-DA hydrogel in acetone. 2.6. NMR Relaxation of the Solvent Molecules NMR relaxation experiments on the nuclei of solvents should generally provide information on the mobility of the solvent molecules, and consequently on polymer–solvent interactions. The dynamical behavior of the solvent molecules was studied using the measurements of the 1 H NMR spin–spin relaxation time T 2 on water (HDO) and acetone signals and the T 2 values as obtained for the DN-DA hydrogel at various solutions are summarized in Table 1 . Single-exponential relaxation decay characterized by single relaxation T 2 was detected for the DN-DA hydrogel at the temperature of 25 °C ( Figure S1a ). Heating at 45 °C leads to bi-exponential relaxation decay ( Figure S1b ) and the T 2 components of bi-exponential dependences differ significantly from each other and they can be marked as relaxation times of free ( T 2 = 5.0 s) and bound ( T 2 = 1.0 s) water. The main reason for these differences is that the motion of bound water in collapsed structures is spatially restricted and anisotropic [ 39 ], while free water molecules are either contained in less swollen polymer structures or released from the interior of the hydrogel. The occurrence of bound water molecules with slow motion was also previously detected in collapsed poly(vinyl methyl ether) and PNIPAm hydrogels [ 43 , 44 , 45 , 46 , 47 ] and interpenetrating PNIPAm-based hydrogels [ 36 ]. Single- and bi-exponential relaxation decays were also detected for water and acetone signals in water–acetone mixtures with the DN-DA hydrogel ( Figure S1c–f ). For the 20 vol.% acetone solution, the solvent molecules show a single, relatively high relaxation time T 2 value, which corresponds to their high mobility in the swollen structures of the DN-DA hydrogel. In 80 vol.% acetone solution, the DN-DA hydrogel is collapsed with a very-low swelling ratio ( Figure 4 b) and the solvent molecules are either in a free or bound state and, at the same time, the value of the relaxation time of bound molecules is up to three orders of magnitude lower than the relaxation time of the free molecules. This signifies that the collapsed polymer structures of the DN-DA hydrogel are very compact and rigid, and the solvent molecules that are bound to/in them are similarly limited in their mobility. Furthermore, it is clear from Table 1 that both water and acetone molecules show a similar behavior. Both solvent molecules occur either in the free state, without any restriction in its mobility, or in the bound state, where they are bound in relatively immobilized globular structures, thus indicating that the decisive factor in this behavior is, in both cases, a polar character of these molecules and hydrogen bonding. A similar behavior as that described above for water and acetone molecules from 1 H relaxation measurements was previously found for water and ethanol molecules in poly(vinyl methyl ether)/D 2 O/ethanol solutions [ 48 ]. 2.7. Deswelling and Swelling Kinetics Time-dependent deswelling and swelling kinetics are shown in Figure 9 and Figure 10 , respectively. To compare the effect of various stimuli, the DN-DA hydrogel samples swollen at equilibrium in water of 25 °C were immersed in pure acetone or 6 M NaCl solution or water of 45 °C. The deswelling process was monitored gravimetrically as a function of the time of deswelling ( Figure 9 a). After attaining the equilibrium collapsed state, the hydrogels were again immersed in water of 25 °C and the reswelling behavior was monitored until the new equilibrium state was obtained ( Figure 10 a). Similar procedures were performed for the water–acetone mixtures with various acetone contents and the relevant deswelling and reswelling curves are shown in Figure 9 b and Figure 10 b, respectively. Swelling/deswelling processes in hydrogels are assumed to follow first-order kinetics [ 49 , 50 ] and the time dependence of the swelling ratio SR ( t ) after stimuli change could be described by the equation [ 51 ]\n (1) S R t = S R ∞ + A 1 exp − t τ 1 + A 2 exp − t τ 2 \nwhere SR ∞ is equilibrium swelling ratio, A 1 and A 2 are pre-exponential factors, and τ 1 and τ 2 are characteristic time parameters. The measured time dependences in Figure 9 and Figure 10 showed a two-step character and it was necessary to fit them with the function with two characteristic times. Table 2 contains the fitting characteristic time parameters τ 1D , τ 2D for the deswelling curves in Figure 9 and τ 1S , τ 2SD for the swelling curves in Figure 10 . The swelling and deswelling curves for 20 vol.% acetone solution could not be fitted with sufficient accuracy to Equation (1) and the parameters are not included in Table 2 . From Table 2 , it follows that the time-dependent processes occur in two steps: a rapid process with a characteristic time τ 1 = 5–15 min is followed by a slow deswelling/reswelling step, which is characterized by a time τ 2 = 70–130 min. The dependences for deswelling in 80 vol.% and 100 vol.% acetone solutions showed only one characteristic time, apparently because the first process was so fast that it could not be detected. The two-step behavior of the deswelling curve is probably caused by the formation of two-phase structure, where at the beginning of the deswelling process, the outer part of the hydrogel sample deswells relatively quickly as solvent molecules have short diffusion distances for transport outside of the hydrogel. The surface part of the hydrogel thus shrinks and forms a hydrophobic barrier for the transport of solvent molecules from the inner part of the hydrogel, causing a very slow release of water and thus the longer characteristic time represented in the second step. Two-step behavior of deswelling curve was previously found for semi-interpenetrating hydrogels based on PNIPAAm [ 52 ]. Similarly, the two-step reswelling profile of the hydrogel can be explained by the formation of a two-phase structure. At the beginning of reswelling, the sample surface is in contact with water molecules, so the shrunk chains on the surface of the hydrogel begin to relax and water is easily absorbed. As the hydrogel swells, a two-structure is formed: the surface part with solvated network chains and the inner part with unswollen network chains. As the surface area of the inner part of the hydrogel decreases with increasing swelling ratio, the swelling process slows down after the initial rapid swelling. Similar two-step reswelling profile was also found and described for the PAAm hydrogels [ 53 ]. If we compare the characteristic times for the deswelling process in water–acetone mixtures, it is clear from Table 2 that, with increasing acetone content, both characteristic times decrease and the deswelling process becomes faster. In 100 vol.% acetone, the characteristic time τ 2D is practically one order of magnitude shorter compared to the deswelling time in the mixture with 30 vol.% acetone. The deswelling rate is thus obviously correlated with the change in the swelling ratio after immersion a sample from pure water into the water–acetone mixture. The DN-DA hydrogel shows the greatest change in the swelling ratio in 80 vol.% and 100 vol.% acetone ( Figure 4 b) and, at the same time, it has the fastest deswelling process in these mixtures compared to other water–acetone mixtures. Analogously, the same behavior can be observed when comparing the effect of various stimuli on the deswelling rate in the DN-DA hydrogel. As the DN-DA hydrogel has a relatively high swelling ratio in the 6 M NaCl solution or in pure water at 45 °C, the hydrogel deswells slowly under the influence of temperature or salt. Interestingly, the value of the time τ 1D for deswelling in 45 °C water is very low, indicating a rapid initial process of deswelling. This could be related to the faster diffusion of water molecules at a higher temperature and their ability to reach the outer part of the hydrogel quickly. The values of characteristic times for reswelling processes in Table 2 indicate that reswelling is somewhat slower compared to deswelling. The difference is more significant for the processes that had a fast deswelling in water–acetone mixtures with 60–100 vol.% of acetone. A slower swelling in hydrogels was observed earlier, e.g., in polyacrylamide hydrogels in aqueous NaCl solutions [ 54 ] or in temperature-sensitive PNIPAAm hydrogels [ 50 ]. The dissolution of the collapsed structures can be slowed down by the existing entanglements [ 55 ]. The solvent must thus disrupt the compact globular structures, and this is a slower process than when the solvent is expelled out of the expanded hydrogel chains, which are quickly dehydrated. Similar to deswelling, the reswelling processes are described by two characteristic times, τ 1S and τ 2S . The short time τ 1S corresponding to initial stage of swelling is practically the same for all solutions in which the hydrogel was immersed, with the exception of water at temperature 45 °C, when the time τ 1S is very short due to the faster diffusion of water molecules. For water–acetone mixtures, the time τ 2S decreases with increasing acetone content, but its value is higher compared to the time τ 2D for the deswelling process."
} | 8,516 |
30509471 | null | s2 | 8,205 | {
"abstract": "The chapter focuses on the methods involved in producing and characterizing two key nickel-iron-sulfur enzymes in the Wood-Ljungdahl pathway (WLP) of anaerobic conversion of carbon dioxide fixation into acetyl-CoA: carbon monoxide dehydrogenase (CODH) and acetyl-CoA synthase (ACS). The WLP is used for biosynthesis of cell material and energy conservation by anaerobic bacteria and archaea, and it is central to several industrial biotechnology processes aimed at using syngas and waste gases for the production of fuels and chemicals. The pathway can run in reverse to allow organisms, e. g., methanogens and sulfate reducers, to grow on acetate. The CODH and ACS intertwine to form a tenacious CODH/ACS complex that converts CO"
} | 182 |
38611700 | PMC11013183 | pmc | 8,207 | {
"abstract": "Polyurethane elastomers are among the most versatile classes of industrial polymers—typically achieved through a two-step synthesis of segmented block copolymers, comprising very long and soft segments that provide elasticity and significantly long and hard segments that provide strength. The present research focused on the design of a single-step synthesis of a new segmented polyurethane consisting of very short soft and hard segments, crosslinked by preferentially side-reacted hierarchical tertiary oligo-uret network structures, thus exhibiting significant strength, elasticity, and toughness. Despite the theoretically linear structure, both FTIR and solid-state 13 C NMR spectroscopy analyses indicated the quasi-equal presence of urethane groups and tertiary oligo-uret structures in the resulting polymer, indicating a preferential consecutive side reaction mechanism. Thermal analysis indicated the significant crystallization of soft segments consisting of only four ethylene oxide units, which was, hereby, demonstrated to occur via an extended chain mechanism. Tensile mechanical properties included significant strength, elasticity, and toughness. Increasing the soft segment length led to a decreased tertiary oligo-uret secondary crosslinking efficacy. The preferential hierarchical side reaction mechanism was, hereby, further confirmed through the synthesis of a completely new type of hyper-branched polymer via diisocyanate and a mono-hydroxy-terminated reagent. The structure–property relations and reaction mechanisms demonstrated in the present research can facilitate the design of new polyurethanes of enhanced performance and processing efficacy for a variety of novel applications.",
"conclusion": "4. Conclusions Polyurethane elastomers are typically achieved through a two-step synthesis finally resulting in segmented block copolymers comprising very long soft segments that provide elasticity and significantly long hard segments that provide strength. The present research focused on the design of a highly efficient single-step synthesis of a new segmented polyurethane consisting of very short soft and hard segments, crosslinked by preferentially side-reacted hierarchical tertiary oligo-uret network structures, thus exhibiting significant strength, elasticity, and toughness. A segmented poly(ether urethane) was synthesized by reacting PEG200 with HDI at a 1:1 molar ratio, the theoretically linear structure of which consisted of a very short soft segment of only four ethylene oxide units and a hard segment of only two urethane groups. Despite the theoretically linear structure of the newly synthesized polymer, both FTIR and solid-state 13 C NMR spectroscopy analyses indicated a highly non-linear structure, exhibiting a quasi-equal presence of urethane groups and tertiary oligo-uret network structures in the resulting polymer. This is highly consistent with a mechanism of the preferential consecutive side-reaction of isocyanate groups with secondary nitrogens in a hierarchical consecutive pattern, rather than creating new allophanate structures with the remaining non-side-reacted urethane groups in the polymer and even rather than creating new urethane groups. Although PEG200 itself is an amorphous liquid which cannot crystallize under any conditions, the DSC analysis of the resulting polyurethane exhibited a significant endotherm at a temperature characteristic of the melting of the crystallized PEG. This was further also confirmed in the FTIR spectrum of the polymer by the strong presence of an additional C–O–C stretching absorbance in a well-known location that is characteristic of crystalline PEG. Thus, the data obtained suggest the possible presence of crystalline domains mainly consisting of PEG chains. It was hereby deduced that the mechanism of the PEG200 soft segment crystallization in this polymer was the parallel extended-chain mechanism, through the orientation of the PEG segments anchored in place by the crosslinking structures at their extremities—and not via a chain-folding crystallization mechanism. The suggested parallel extended-chain mechanism was also highly consistent with, and confirmed by, the tensile stress–strain curve profile and the relatively low elongation at break (as compared to average elastomers), which was highly consistent with the state of the extended-chain pre-alignment of the polymer soft segments. The optical microscopy photographs of the polymer indicated very significant chain orientation, also exhibiting occasional regions of fibrillation. Tensile mechanical testing exhibited a combination of significant strength, elasticity, and toughness. In view of the very short soft and hard segments of the present polymer, it is hereby deduced that the flexibility of the complex hierarchical tertiary oligo-uret crosslinking network structures strongly contributed to both the elasticity and strength of the resulting segmented polyurethane, which is in significant agreement with the purpose and objectives of the present research. Increasing (doubling) the length of the polymer soft segment at constant hard segment length led to a decreased efficacy of tertiary oligo-uret formation—and mainly a decreased efficacy of the secondary crosslinking of the formed tertiary oligo-uret structures, due to the increased distance between the crosslinking sites—which resulted in very poor properties of the resulting polymer. The preferential occurrence mechanism of the above-described side reactions and of the resulting hierarchical tertiary oligo-uret networks formation were hereby further confirmed and elaborated through the design and synthesis of a completely new type of hyper-branched polymer, by reacting the same diisocyanate (HDI) with a mono-hydroxy-terminated PEG, instead of the diol-terminated PEG used in the synthesis described above. Both FTIR and solid-state 13 C NMR spectroscopy analyses indicated a very strong presence of tertiary oligo-uret structures, along with a very significant amount of remaining non-side-reacted urethane groups. This unambiguously indicated that the N=C=O groups in the system preferentially further consecutively side-reacted with the significantly smaller amount of already formed allophanate groups, instead of forming additional new allophanates with the remaining non-side-reacted urethane groups. In order to further demonstrate that the occurrence of two or more distinct carbonyl-stretching FTIR absorbances are most probably due to the occurrence of side reactions and not due to a combined presence of hydrogen bonding and non-bonding, as commonly suggested for polyurethanes, an additional synthesis was hereby performed, leading to NH- and C=O-containing structures, with abundant hydrogen bond formation ability but without the possibility of the occurrence of any side reactions. Accordingly, nylon 6,6 polyamide was hereby synthesized via interphase polymerization at room temperature. A single extremely thin and sharp carbonyl-stretching absorbance was obtained in the spectrum, strongly indicating that the combined presence of hydrogen-bonded and non-bonded structures (which are most probably present in this polymer, as also in polyurethanes) may of course cause a shift of the absorbance peak but most probably does not cause any peak-splitting of the FTIR carbonyl-stretching absorbance, as was previously often attributed to polyurethanes. The structure–property relations and reaction mechanisms demonstrated in the present research can facilitate the design of new polyurethanes of enhanced performance and processing efficacy for a variety of novel applications, ranging from various medical devices to ballistic impact-resistant devices and as composite matrices. The novelty of the present research pertains to advantageously and purposefully using the recently discovered tertiary oligo-uret-forming side reactions in the design of new non-linear polyurethane elastomers, in order to obtain the desired properties of polyurethane elastomers, through the following: (1) shorter single-step syntheses; (2) using very short hard segments in combination with tertiary oligo-uret networks and consequently using much smaller amounts of diisocyanate; and (3) using non-aromatic diisocyanates, which will be highly beneficial in terms of both safety and environmental considerations and also significantly reduce industrial manufacturing costs. The present research also demonstrated that the optimal efficacy of tertiary oligo-uret development and its efficient secondary crosslinking occurs when using very short soft segments; this efficacy decreases with increasing soft segment length. Thus, the importance and efficacy of using very short soft and hard segments in these polymers was hereby demonstrated. A completely new type of hyper-branched polymer was hereby synthesized for the first time by advantageously using the tertiary oligo-uret-forming side reactions, which were also used to conclusively demonstrate the occurrence of these side reactions, without which this synthesis would not have been possible. In addition, the present research presents a new approach in the diagnostic interpretation of the FTIR carbonyl-stretching absorbances in polyurethanes, as related to the occurrence of side reactions.",
"introduction": "1. Introduction Polyurethanes are among the most important and highly versatile classes of industrial polymers, exhibiting a vast variety of applications ranging from very common consumer products, such as shoe soles, foams, textile fibers, coatings, and parts for the automotive industry, to biomedical applications, such as medical devices and artificial implants. Polyurethanes are typically synthesized through the reaction between di-functional isocyanates and di-functional primary alcohols, leading to the formation of urethane-containing repeating polymer units of thermoplastic polyurethanes [ 1 ]. The use of aliphatic or aromatic diisocyanates in combination with very short diols results in the formation of highly rigid and strong polymers, which are highly suitable for applications that require such properties [ 1 , 2 , 3 ]. Nevertheless, all applications of polyurethanes require elasticity to some extent, with some demanding large deformations. Elastic polymers are commonly termed elastomers. Polyurethane elastomers are typically synthesized so as to consist of alternating hard and soft segments, preferably exhibiting microphase segregation due to purposefully designed significant differences in polarity [ 1 , 2 , 3 ]. The most commonly used soft segments are diol-terminated soft polymers of molecular weights in the range of thousands of g/mol (e.g., 2000–6000 g/mol), mostly diol-terminated polyethers or polyesters. The synthesis of a segmented polyurethane typically requires a two-step process, consisting of the preliminary synthesis of a macro-diisocyanate prepolymer (through the reaction of a diisocyanate with each of the two terminal hydroxyls), which is subsequently further reacted with additional di-functional reagents such as diols, diamines, or dicarboxylic acids, commonly termed as chain extenders, resulting in additional urethane, urea, or amide groups, respectively [ 1 , 2 , 3 , 4 , 5 , 6 , 7 , 8 , 9 , 10 , 11 , 12 , 13 , 14 ]. As the urethane groups, urea groups, and amide groups are relatively planar and rigid and have the ability of hydrogen bonding, they constitute the polymer’s hard segments and are the main source of the polymer’s strength. The less ordered soft segments stretch and align in the direction of an applied tensile stress and relax when the stress is released, thus creating the elastic properties of the polymer [ 1 , 2 , 3 , 4 , 5 , 6 , 7 , 8 , 9 , 10 , 11 , 12 , 13 , 14 , 15 , 16 , 17 , 18 , 19 ]. Although the reaction between a diol-terminated soft segment and a diisocyanate is sufficient for the formation of a polyurethane polymer [ 1 , 2 ], the above-described macro-diisocyanate and chain extension synthesis pathway is usually performed with the purpose of obtaining a longer hard segment, thus obtaining significantly strong elastomers, especially when using relatively long and soft segments [ 3 , 4 , 5 , 6 , 7 , 8 , 9 , 10 , 11 , 12 , 13 , 14 , 15 , 16 , 17 , 18 , 19 ]. A well-known example of this is Lycra ® , which is a poly(ether urethane urea) segmented elastomer mostly applied as a textile fiber and also in some biomedical uses [ 20 , 21 , 22 , 23 , 24 , 25 ]. In polyurethane synthesis, the only theoretical chemical change occurring is the formation of the urethane groups and either urea or amide when chain-extended with a diamine or a dicarboxylic acid, respectively. Each of these groups contains one carbonyl, the presence and spectral location of which is highly analytically diagnostic for the monitoring and confirmation of the polymerization reaction occurrence and for the identification of the types of groups which are formed. FTIR spectroscopy analysis is among the most widely used analytical methods for the assessment of the polyurethane synthesis process and of the final product chemical structure, both in research and in industrial processes. Nevertheless, most industrial polyurethanes contain significantly long and soft segments, thus resulting in a very high intensity of the analytical signals stemming from the very abundant ether or ester groups of said soft segments and, thus, a very low intensity of the analytical signals related to the carbonyls of the hard segment structures, due to their much lower concentration in the polymer. As a consequence, the accurate diagnostic identification of the different carbonyl types occurring in polyurethane hard segments is a challenging task in these materials. The ability to perform accurate diagnostic carbonyl type identification is also highly important for the correct assessment and analysis of side reactions that may occur in polyurethane synthesis, which may highly affect the final polymer properties and possible applications. Polyurethane synthesis is accompanied by side reactions occurring predominantly between isocyanate groups and the secondary nitrogen groups (i.e., NH groups) of already formed urethane groups, due to the inherently very high reactivity of the isocyanate groups. This side reaction is long-known to result in the formation of allophanate structures [ 1 , 2 , 3 ]. Despite the very long-known tendency of diisocyanates to side-react with the NH groups of urethanes and ureas to form allophanates and biurets, respectively [ 1 , 2 , 3 ], very numerous polyurethane synthesis-related research studies over many decades have attributed the occurrence of two, and even multiple, distinct carbonyl-stretching absorbances in the FTIR spectra of polyurethanes to a combined presence and absence of hydrogen bonding between some portions of the urethane groups—without even considering the very possible occurrence of side reactions and, consequently, the possible formation of chemical structures containing additional carbonyl types, thus exhibiting different adjacent carbonyl-stretching absorbance locations. In a very early book entitled Polyurethanes — Chemistry , Technology and Properties , published in 1964 [ 26 ], the authors state as follows: “There is no doubt that isocyanates react with urethanes, but surprisingly few accounts occur in the literature of the isolation and characterization of a product. If the isocyanate reacts ‘normally’ with a urethane, the product will be a substituted allophanic ester” [ 26 ] (p. 108). It is even more surprising that, to the present day, most of the reported polyurethane synthesis-related research works using diisocyanates do not even mention the very high possibility of side reactions occurring—and mostly attribute the occurrence of two, or even multiple, distinct adjacent carbonyl-stretching FTIR absorbances to the effects of hydrogen bonding and the non-bonding of the urethane carbonyls [ 27 , 28 , 29 , 30 , 31 , 32 , 33 , 34 ]. It was recently demonstrated, though, that once allophanate structures are formed, they preferentially further consecutively side-react with additional isocyanate groups to form crosslinking hierarchical tertiary oligo-uret network structures [ 35 , 36 ]. The preferential formation of tertiary oligo-uret structures was further demonstrated to also occur in diisocyanate-derived polyurea synthesis [ 37 , 38 ] and was additionally found to be highly dependent on the polyurea synthesis pathway [ 38 ], the relatively mild and slow water–diisocyanate pathway predominantly and preferentially produces tertiary oligo-uret structures [ 38 ], while the extremely quick and highly exothermal diamine–diisocyanate synthesis pathway of the exact same polymer almost exclusively leads to the formation of biuret structures [ 37 ]. These tertiary oligo-uret structures, as also described in the referenced research works [ 35 , 36 , 37 , 38 ], may schematically be represented by the following: -N(R)C=ON(R)C=ON(R)C=ON(R)C=ON(R)… and so on (with R representing the diisocyanate R group, such as, for example, hexamethylene). The length of the tertiary oligo-uret structure depends on the number of hierarchical consecutive side reactions that occurred. All the nitrogens in the structure, except the last one, are tertiary nitrogens. Thus, the so-called linear polyurethanes are actually non-linear, due to the concomitant occurrence of these side reactions [ 1 , 2 , 3 , 26 , 35 , 36 , 37 , 38 ]. Side reactions are, by definition, commonly regarded as unwanted—and are, thus, named as side reactions, occurring beside the wanted reaction—and when occurring in significant proportion during polymer synthesis, they may significantly affect the final polymer properties and possible applications. This effect on the final polymer properties is especially accentuated in polyurethanes where side reactions may lead to significant polymer crosslinking [ 1 , 2 , 3 , 26 , 35 , 36 , 37 , 38 ]. As the above-described side reactions concomitantly occur during polyurethane synthesis, it would be highly beneficial to advantageously use these side reactions in the design and synthesis of new types of polyurethanes, with possibly new structures, properties, and applications. The presence of these highly complex hierarchical crosslinking networks—concomitantly inducing both strength and flexibility—may offer the possibility of significantly reducing the hard segment length while still obtaining the desired mechanical properties that are commonly obtained with very large, hard segments, which commonly also contain aromatic R groups. Since long, hard segments are typically obtained via a two-step prepolymer and chain extension process, designing a single-step polyurethane synthesis also has the distinct advantage of requiring the use of much reduced amounts of diisocyanate, which would be highly beneficial in terms of both safety and environmental considerations. Also, the compensating presence of the hierarchical tertiary oligo-uret crosslinking networks may offer the possibility of using aliphatic diisocyanates—instead of aromatic—which is also highly beneficial in the same context. The present research focused on the design of a single-step synthesis and characterization of a new segmented polyurethane consisting of very short hard and soft segments, combining high flexibility and enhanced mechanical properties. The concept of the present research stems from the initial hypothesis that the above-described expected preferential formation of crosslinking tertiary oligo-uret network structures may increase the efficacy of a very short hard segment. Concomitantly, this formation may also enable the generation of significant elastic properties using a very short soft segment, due to a certain degree of flexibility in the long hierarchical tertiary oligo-uret crosslinking network structures.",
"discussion": "2. Results and Discussion The synthesis of segmented polyurethane designed in the present research was performed using hexamethylene diisocyanate (HDI) and PEG200 at a molar ratio of 1:1. The theoretical linear structure of the resulting polymer is presented in Scheme 1 . As observed in Scheme 1 , the very short, hard segment consists of only two urethane groups in the polymer repeating unit, separated by the hexamethylene group of the original HDI. The soft segment of the polymer, i.e., PEG200, is very short when compared to industrial polyurethanes such as the well-known Lycra ® , in which the soft segment consists of polytetramethylene glycol (PTMG) 2000 and the hard segment consists of two urethane groups and two urea groups in the polymer repeating unit [ 20 , 21 , 22 , 23 , 24 , 25 ]. The synthesis resulted in a solid, strong, and highly flexible polymer, the weight of which was close to the stoichiometrically calculated weight. Figure 1 exhibits the FTIR spectrum of the segmented polyurethane, synthesized with hexamethylene diisocyanate (HDI) and PEG200 at a molar ratio of 1:1. Although as seen in Scheme 1 , the theoretical polymer repeating unit contains only one type of carbonyl, i.e., the urethane carbonyl, the FTIR spectrum of the synthesized polymer clearly exhibits two adjacent strong and sharp carbonyl-stretching absorbances at around 1710 cm −1 and at 1687 cm −1 of almost equal intensity. As was also previously demonstrated, the absorbance at around 1710 cm −1 belongs to the carbonyl-stretching vibrations of the urethane group [ 26 , 27 ], though slightly shifted to the right as compared with the urethane carbonyl-stretching absorbance previously observed at 1717 cm −1 [ 35 , 36 ]; this shift most probably occurred due to hydrogen bonding. This absorbance ( Figure 1 ), though, also appears to exhibit a slight shoulder or thickening to the left at around 1717 cm −1 , which may be due to non-hydrogen-bonded urethane groups. The absorbance at 1688 cm −1 belongs to the carbonyl-stretching vibrations of tertiary oligo-uret structures, as also previously demonstrated [ 35 , 36 , 37 , 38 ]. As demonstrated previously [ 35 , 36 , 37 , 38 ] and as described in the above Introduction section, the tertiary oligo-uret structures are the result of the further side reaction of already formed secondary nitrogens during synthesis with an isocyanate, thus creating a tertiary nitrogen and a newly formed NH group—which, in turn, can similarly further side-react with an additional isocyanate, again creating a tertiary nitrogen and a new NH group, and so on. Each additional consecutive side reaction increases the array of electron-withdrawing groups preceding the terminal NH in the structure—increasing the polarity of the H atom on the said NH group, thereby also increasing its reactivity toward an isocyanate group. Hence, the reaction exhibits the preferential predominant formation of the tertiary oligo-uret network structures [ 35 , 36 , 37 , 38 ]. The urethane carbonyl-stretching absorbance is located at a higher wavenumber than that of the tertiary oligo-uret structures, which is consistent with its close proximity to the more electronegative oxygen atom in the urethane group. A small shoulder can be observed to the right of this peak, at around 1640 cm −1 . An absorbance at this location was previously demonstrated in diisocyanate-derived polyureas, as the stretching absorbance of carbonyls in biuret structures [ 37 , 38 ]. As two-sided biurets are actually short tertiary oligo-uret structures, in the present polymer, this absorbance most probably belongs to carbonyls in short tertiary oligo-uret structures that constitute short branches of larger tertiary oligo-uret structures and/or branches of secondary tertiary oligo-uret crosslinkers of the larger tertiary oligo-uret network structures, which, in turn, crosslink the main polymer chains. The very strong absorbance at 1103 cm −1 belongs to the C–O–C stretching vibrations of the PEG ether groups. It is important to also take note of the strong and sharp adjacent absorbance at 1050 cm −1 ( Figure 1 ). This absorbance (as a shoulder to the right of the main C–O–C stretching absorbance) is highly characteristic of PEG in crystalline form (and is totally absent in amorphous PEG) [ 39 ]. This strong indication that the PEG segments in the polymer are in a crystalline state is highly surprising, since PEG200 itself is an amorphous liquid that does not crystallize under any conditions (PEG crystallization at room temperature occurs at and above MW = 1000). Additional absorbances in the spectrum ( Figure 1 ) are as follows: the strong and sharp absorbance at 1538 cm −1 due to the C–N–H deformation; the strong N–H stretching absorbance at 3337 cm −1 ; the symmetric and antisymmetric N–C–N stretching vibrations at 1460 cm −1 and 1261 cm −1 ; and the characteristic methylene group absorbances, i.e., the symmetric and antisymmetric CH stretching vibrations at 2851 cm −1 and 2953 cm −1 ; the CH bending vibrations at 1450 cm −1 ; and the CH rocking vibrations at 616 cm −1 . It is also interesting to note that the spectrum exhibits a very much increased intensity of the asymmetric CH stretching absorbance at 2953 cm −1 , as compared to the symmetric CH stretching absorbance at 2851 cm −1 , which is highly consistent with a very significant chain alignment in this polymer, with a most probable consequent parallel extended-chain interaction, thus preferentially enabling the asymmetric CH-stretching vibration, and less so enabling the symmetric CH-stretching vibration—being relatively hindered by the said parallel inter-chain interaction. It is also important to take note of the almost complete absence of the N=C=O absorbance at 2270 cm −1 , indicating that all the isocyanate groups in the system have reacted. Another important parameter that may be observed in the FTIR spectrum of the polymer ( Figure 1 ) is the very pronounced width of the N–H stretching absorbance at 3337 cm −1 . It was recently demonstrated for polyurea synthesized via the water–diisocyanate synthesis pathway [ 38 ] that the much enhanced width of this absorbance is not due to effects of hydrogen bonding, but it occurs as a consequence of this peak consisting of a relatively wide distribution of N–H stretching absorbance locations, stemming from the wide length distribution of the tertiary oligo-uret hierarchical network structures in the polymer. With an increasing length of the tertiary oligo-uret structure and an increasing number of tertiary nitrogens and carbonyls which are situated between the two terminal NH groups, the electron-withdrawing activity of the environment is increased, thus lowering the NH groups’ electron density ( Scheme 2 ). This, in turn, leads to a wide distribution of the FTIR NH-stretching vibration locations, as a function of the various tertiary oligo-uret lengths in the polymer and of the additional location for the NH groups in the non-side-reacted urethane structures [ 38 ]. This is in strong contrast with the extremely thin and sharp N–H stretching absorbance peak in a polymer exclusively containing biuret structures [ 37 ]. This is due to the fact that all biuret structures are exactly the same and exhibit a symmetrical structure. This was further confirmed through the synthesis of a new polymer, inherently containing oligo-uret structures in the polymer repeating unit, leading to a very high content of tertiary oligo-uret structures following further side reactions. Consequently, several separate sharp N–H stretching absorbances with a common, very wide base were present in the FTIR spectrum of this polymer, which confirmed the above-described interpretation [ 38 ]. Figure 2 exhibits the differential scanning calorimetry (DSC) analysis of the resulting polymer. A very significant endotherm appears at 48.25 °C (ΔH = 19.31 J/g), which is characteristic of PEG melting [ 39 ]. This may indicate the possibility that the PEG200 soft segments in the polymer have crystallized, which is in strong agreement with the observed strong and sharp FTIR absorbance (at 1050 cm −1 ) to the right of the main C–O–C stretching absorbance, which is highly characteristic of PEG in crystalline form [ 39 ] ( Figure 1 ). In the crystalline state, the FTIR spectrum of PEG is, in some aspects, different from that in the molten state, which may be explained by the fact that in the crystalline state, the chain conformation consists of helical turns, with internal turns around the O–CH 2 , CH 2 –CH 2 , and CH 2 –O bonds of trans, gauche, and trans, respectively [ 39 ]. In the molten state, the conformation becomes disordered [ 39 ]. The theoretical melting enthalpy for 100% crystalline PEG is ΔHm 0 = 196.8 J/g [ 40 ]. Thus, the degree of crystallinity of the PEG200 soft segments in the present polymer may be calculated by relating the melting endotherm enthalpy hereby obtained (ΔH = 19.31 J/g) to the ΔHm 0 of PEG and normalizing it to the fraction of the PEG segments in the polymer, which accounts for a PEG soft segment degree of crystallinity of 18% (i.e., 18% of the soft segments in the polymer have crystallized). Again, this is highly surprising, since the PEG of MW = 200 in its pure unreacted form is an amorphous liquid that does not crystallize under any conditions; nevertheless, the data obtained suggest the possible presence of crystalline domains mainly consisting of PEG chains. The very wide and shallow endotherm at around 90 °C is due to the evaporation of the hygroscopic water molecules, which were adsorbed from the atmosphere by the relatively hydrophilic polymer. Polymers typically crystallize via two possible crystallization mechanisms: 1. The extended-chain mechanism, in which the polymer chains are stretched and oriented in parallel during the crystallization process; 2. The chain-folding mechanism, in which the polymer chains repeatedly fold during the crystallization process to form lamellar crystalline structures [ 41 , 42 , 43 , 44 , 45 , 46 , 47 ]. The crystallization of polymers via the extended-chain mechanism can only occur when a very significant stretching force is applied on the polymer during the crystallization process—as commonly applied during the industrial production of polymeric fibers. Otherwise, polymers naturally crystallize via the chain-folding mechanism [ 41 , 42 , 43 , 44 , 45 , 46 , 47 ]. The molecular weight of PEG200 is much lower than the crystallization threshold for PEG, and, also, the very short chain length at this molecular weight does not offer the possibility of a chain-folding mechanism. Still, the above-described results indicate a significant crystallization of the PEG200 soft segments in the polymer. The only possible mechanism of PEG200 soft segment crystallization that most probably occurred here is the parallel extended-chain orientation of the PEG segments, anchored in place by the crosslinking structures at their extremities, in close-enough parallel proximity to enable crystallization. The parallel orientation of the chains most probably occurred both before and during the crosslinking process, due to the significant shear forces induced through the mechanical stirring of the reactor content during synthesis. The co-crystallization of urethane groups along with the PEG200 segments may be ruled out due to the highly pronounced differences between these two chemical structures, as reflected in the following characteristics: pronounced different polarities favoring microphase separation; the significant planar structure and rigidity of urethane groups versus the very pronounced flexibility of the PEG ether groups; and the fact that a large portion of the initially formed urethane groups are side-reacted and held within crosslinking network structures. Although hydrogen bonding most probably does occur in at least some of the non-side-reacted urethanes, these very short hard segments consisting of only two urethanes, separated by six methylenes, also most probably do not separately crystallize. Furthermore, a well-known phenomenon occurring in segmented polymers (i.e., in block copolymers) is inter-segmental interference, in which the crystallization of a segment of more dominant physical properties (i.e., of a hard segment) diminishes and even prevents the crystallization of the adjacent soft segments covalently bonded thereto [ 48 , 49 ]. This was also demonstrated in very recent research [ 27 ] showing that the crystallization of the hard segments in the synthesized polyurethanes prevented the crystallization of the polycaprolactone (PCL) soft segments of these polymers. Thus, any occurrence of urethane hard segment crystallization in the present polymer would have definitely prevented the crystallization of the hereby very short PEG200 soft segments. Scheme 2 exhibits a possible schematic representation of the molecular structure of the polymer. Each N–H group in the structure can potentially further side-react with an isocyanate group, creating a tertiary nitrogen and a new secondary nitrogen, which may in turn react with an isocyanate, and so on, creating increasingly complex hierarchical crosslinking networks. Crosslinking may occur not only between main polymer chains but also within and between the crosslinking network structures. It may be observed that although the tertiary oligo-uret structures are highly complex, they offer the possibility of the formation of the close and parallel alignment of groups of polymer chains in a crosslinked state—which is in strong agreement with the extended-chain crystallization mechanism described above. Figure 3 exhibits the tensile mechanical properties of a representative tested sample of the segmented polyurethane synthesized with hexamethylene diisocyanate (HDI) and PEG200 at a molar ratio of 1:1. The polymer exhibits a significantly high tensile strength of close to 28 MPa (the average tensile strength of the seven samples tested in the present research was 28.8 MPa, with a standard deviation of 3.6—as detailed in Table 1 ). By comparison, the tensile strengths of previously reported segmented polyurethanes, synthesized using the same diisocyanate (HDI) but with much longer hard segments (consisting of two urethanes and two amides and, additionally, containing a C=C double bond) and a PTMG2000 soft segment [ 11 ] or a polybutadiene2000 soft segment [ 12 ], were 50 MPa and 25 MPa, respectively. Additional examples of polyurethane elastomer tensile properties reported in the literature are as follows: polyurethane elastomers consisting of PTMG soft segments, MDI, and a butanediol (BD) chain extender exhibited tensile strengths of between 20 and 45 MPa and exhibiting elongations of between 650 and 1100% [ 50 , 51 , 52 ]; poly (urethane urea) elastomers containing various soft segments consisting of polyethers, or polyesters such as polycaprolactone, exhibited tensile strengths of between 10 and 38 MPa and exhibiting elongations of between 100 and 360% [ 53 , 54 , 55 , 56 , 57 ]. The presently designed polyurethane is within the range between these tensile strength values of polyurethane elastomers, despite the very short hard segment of the present polymer. Nevertheless, as an extremely wide range of chemical structures can be designed and exist in the vastly large and versatile family of polyurethanes, the Young’s moduli of polyurethanes range from 1 to 1000 MPa [ 57 ], and typical elongations at break are typically higher than 200% [ 50 , 51 , 52 , 53 , 54 , 55 , 56 , 57 , 58 ]. The relatively low elongation at break obtained in the present poly(ether urethane) is due to and highly consistent with the above-discussed pre-aligned state of the short polymer soft segments, the stretching of which being the main source of the significantly high elongations of polyurethane elastomers. It is, therefore, hereby suggested that the main source of the elongation in the present polymer is most probably the flexibility of the significantly large crosslinking hierarchical tertiary oligo-uret network structures, resulting from the above-discussed side reactions. The industrial synthesis of the polymer in appropriately shaped molds may provide further optimization, which will most probably lead to even higher tensile strength. A well-known and commonly used industrial processing method for crosslinked polyurethanes and poly(urethane urea)s is reaction injection molding (RIM) [ 59 , 60 ]. Nevertheless, the occurrence of some flaws, imperfections, and internal stresses in the final product polymer is practically inevitable, even in highly optimized industrial injection molding processes [ 61 ]. It is also important to take note of the occurrence of a controlled failure mechanism (occurring gradually) as opposed to catastrophic failure (which occurs abruptly/instantly). This is highly advantageous in terms of mechanical properties, as it results in a relatively high breaking energy (as reflected in the relatively large area beneath the curve) and, thus, a relatively high polymer toughness. A relatively high modulus of 58.2 MPa ( Figure 3 ) (average of seven measurements 110.2 MPa— Table 1 ) was obtained, along with a significantly steep ascending first linear part of the stress–strain curve. This is in contrast with the typically encountered elastomer tensile stress–strain curve, which commonly exhibits a very shallow initial ascent (and a very low modulus), which is then followed by a very steep ascent as the soft segments reach a high degree of parallel alignment in the direction of the applied tensile stress [ 11 , 12 , 50 , 51 , 52 , 53 , 54 , 55 , 56 , 57 , 58 ]. The fact that the present stress–strain curve ( Figure 3 ) immediately starts with a steep linear ascent and lacks the usually encountered initial shallow ascent is due to the fact that in the present polymer, the soft segments are already aligned and extended—as discussed above. Figure 4 exhibits optical microscopy photographs of the surface of the poly(ether urethane) synthesized with HDI and PEG200 at a molar ratio of 1:1. The optical microscopy images are indicative of the very significant chain orientation occurring, which occasionally even results in regions of fibrillation and occasionally occurring inter-fibrillar gaps within these regions, as commonly observed in polymeric fibers. This is highly consistent with and confirms the above-discussed soft segments orientation and their possible extended-chain crystallization mechanism. Figure 5 exhibits the solid-state 13 C NMR spectrum of the segmented polyurethane, synthesized with hexamethylene diisocyanate (HDI) and PEG200 at a molar ratio of 1:1. Two very close resonances at 25 ppm and at 28 ppm belong to the carbons of the four inner methylene groups of the HDI hexamethylene group (the two central methylenes resonate at 25 ppm and the following two methylenes at 28 ppm). The resonance at around 40 ppm is attributed to the outermost two methylenes of the HDI hexamethylene group, bonded to the nitrogen atoms. The very strong and sharp resonance at 69 ppm belongs to the two PEG methylenes, bonded to the ether oxygen atoms. The very high intensity of this resonance stems from the high abundance of these groups originating from the PEG200 soft segments. This resonance exhibits a relatively low-intensity shoulder at around 64 ppm, belonging to the two outermost methylenes of the PEG segments which are bonded to the urethane (or urethyl-in allophanate and tertiary oligo-uret structures) ether oxygen atom. A strong carbonyl carbon resonance appears at 155 ppm, exhibiting another very significant shoulder resonance to the right at 148 ppm. The strong resonance at 155 ppm significantly widens to the left, indicating a strong concomitant adjacent presence of the characteristic resonance of carbonyls in tertiary oligo-uret structures at 158 ppm [ 37 , 38 ]. This indicates that at least four types of carbonyls are present in the polymer structure. The tertiary oligo-uret carbonyl was previously identified in polyureas to resonate at 158 ppm [ 37 , 38 ]. The urethane carbonyl was previously demonstrated to resonate at 155 ppm [ 36 ]. These two resonances in the spectrum ( Figure 5 ) are of almost equal intensity (and strongly overlapping due to their close proximity), which is in strong agreement with the almost equal intensities of the carbonyl-stretching absorbances of these two structure types in the FTIR spectrum of the same polymer ( Figure 1 ). Nevertheless, the very wide common base of these carbonyl resonances inevitably indicates that these very wide resonance peaks are composed of a very wide distribution of carbonyl types, each resonating at a slightly different ppm, due to slightly different electron densities stemming from the different electron-withdrawing environments induced by the vast variety of possible structural configurations in the crosslinked polymer hierarchical network (as schematically represented in Scheme 2 ). Thus, for example, the tertiary oligo-uret structures may be of a variety of different lengths, exhibiting an increasing electron-withdrawing environment for each carbonyl with the increasing length of a specific structure. As also seen in Scheme 2 , the tertiary oligo-uret structures can include PEG soft segments in the structure or can alternatively form without including the PEG segments in different regions of the crosslinking networks. The adjacent presence or absence of the ether oxygens of the PEG soft segments may also influence the electron-withdrawing environment and, thus, the electron densities of some of the carbonyls in a specific structure. Also, relatively short tertiary oligo-uret structures may form as secondary crosslinkers of larger tertiary oligo-uret network structures that, in turn, crosslink the main polymer chains (as seen in Scheme 2 ). The resonance shoulder at 148 ppm may be attributed to the carbonyls of these short tertiary oligo-uret secondary crosslinking structures, exhibiting a less electron-withdrawing environment and, thus, a higher electron density of the carbonyls within these structures. These, and many more possible structural network variations, most probably contributed to the occurrence of the very large width of the carbonyl carbon resonances in the present NMR spectrum. An additional poly(ether urethane) was hereby synthesized, using a higher molecular weight of PEG (i.e., PEG400) under the same conditions and the same molar ratio with HDI. The synthesis resulted in a wax-like material, without any significant properties. This is most probably due to the fact that the molecular weight of the PEG soft segment is hereby significantly increased (doubled), while the very short, hard segment length (of only two urethane groups) remains constant, and the distances between them concomitantly increase, i.e., they occur at a significantly decreased frequency along the polymer chain. This also increases the distance between side reaction sites, which also most probably decreases the efficacy of secondary crosslinking between the more distanced tertiary oligo-uret network components. Figure 6 exhibits the FTIR spectrum of the segmented polyurethane, synthesized with hexamethylene diisocyanate (HDI) and PEG400 at a molar ratio of 1:1. A very strong and sharp carbonyl-stretching absorbance is observed at 1716 cm −1 , belonging to the urethane groups. An additional sharp but significantly lower-intensity carbonyl-stretching absorbance appears at 1686 cm −1 , belonging to the formed tertiary oligo-uret network structures. The significantly lower intensity of this absorbance as compared to the urethane carbonyl absorbance and as compared to the same absorbance in the polyurethane synthesized with PEG200 ( Figure 1 ) indicates a much lower efficacy of tertiary oligo-uret network structures formation in the polyurethane synthesized with PEG400, which is consistent with the very poor properties of the resulting polymer. It is also important to notice the sharp carbonyl-stretching absorbance at 1643 cm −1 , which is now of significant intensity (as compared to the small shoulder around this region in the spectrum of the polyurethane synthesized with PEG200, in Figure 1 ). This is most probably due to a significantly increased amount of relatively short tertiary oligo-uret branching—which did not result in final fully developed secondary crosslinking networks, due to the much greater distance between side reaction sites resulting from the hereby much longer PEG400 soft segments. Two additional small carbonyl-stretching shoulders at 1807 cm −1 and at 1771 cm −1 are due to the presence of a small amount of non-further-reacted allophanate groups, again indicating a less complete tertiary oligo-uret network structure formation in this polymer, as compared with the polyurethane synthesized with PEG200, where residual allophanate groups are not observed ( Figure 1 ). Figure 7 exhibits the solid-state 13 C NMR spectrum of the poly(ether urethane), synthesized with HDI and PEG400 at a molar ratio of 1:1. Two resonances at 25 ppm and at 28 ppm belong to the carbons of the four inner methylene groups of the HDI hexamethylene group (the two central methylenes resonate at 25 ppm and the following two methylenes at 28 ppm). The split resonances at 39 ppm and 41 ppm are attributed to the outermost two methylenes of the HDI hexamethylene group, bonded to the nitrogen atoms (which are split since they are bonded to either a secondary nitrogen, or a side-reacted tertiary nitrogen). The very strong and sharp resonance at 69 ppm belongs to the two PEG400 methylenes, bonded to the ether oxygen atoms. The very high intensity of this resonance stems from the very high abundance of these groups originating from the PEG400 soft segments. The very small resonance at around 62 ppm belongs to the two outermost methylenes of the PEG400 segments which are bonded to the urethane (or urethyl—in allophanate and tertiary oligo-uret structures) ether oxygen atom. Two sharp carbonyl carbon resonances are seen at 158 ppm and at 155 ppm, due to the tertiary oligo-uret carbonyls and the urethane carbonyls, respectively. The very small resonance shoulder at 162 ppm belongs to the allophanate urethyl carbonyl carbon, indicating the presence of a small amount of allophanate groups in the polymer that did not further side-react to form tertiary oligo-uret structures. This is in agreement with the two small carbonyl-stretching shoulders at 1807 cm −1 and at 1771 cm −1 observed in the FTIR spectrum of the same polymer, belonging to a small amount of non-further side-reacted allophanate groups ( Figure 6 ). It is interesting to note the total absence of carbonyl carbon resonance at 148 ppm, belonging to the shorter secondary tertiary oligo-uret crosslinkers of larger tertiary oligo-uret network structures that, in turn, crosslink the main polymer chains which strongly appeared in the solid-state 13 C NMR spectrum of the poly(ether urethane) synthesized with HDI and PEG200 ( Figure 5 ). This is also in agreement with the observed less efficient tertiary oligo-uret formation in the FTIR spectrum of the same polymer and the strong appearance of the carbonyl stretching absorbance at 1641 cm −1 , most probably belonging to the short non-crosslinking tertiary oligo-uret branches ( Figure 6 ), which is also highly consistent with the very poor physical properties of this polymer. Figure 8 exhibits the DSC analysis thermogram of the poly(ether urethane) synthesized with HDI and PEG400 at a molar ratio of 1:1. The polymer is essentially amorphous, exhibiting no significant melting endotherm (by looking closely, though, a very small, almost imperceptible endotherm may be observed at around 53 °C, indicating that maybe a very small portion of the chains may have crystallized). This is in strong contrast with the above polymer synthesized with HDI and PEG200, which exhibited a very significant melting endotherm at a temperature characteristic of PEG melting [ 39 ] ( Figure 2 ). This amorphous state is most probably due to the above-described less efficient formation of the tertiary oligo-uret networks and their less efficient secondary crosslinking—thus most probably not efficiently stabilizing the soft segments in an aligned configuration. Figure 9 exhibits optical microscopy photograph of the surface of the poly(ether urethane) synthesized with HDI and PEG400 at a molar ratio of 1:1. The polymer images are not indicative of any significant chain orientation in the polymer, which is also in agreement with the amorphous state of the polymer. This is in strong contrast with the highly oriented and even fibrillar structure image of the polymer synthesized with HDI and PEG200 under the same magnification. It is also important to take note of the occasional appearance of micro-cracks in the polymer, which are also most probably the source of the very poor mechanical properties of this polymer. Due to the poor mechanical properties of this polymer and the inherently present micro-cracks, the polymer broke either during sample preparation or during the fastening of the machine grips on the samples—thus, the measurement of the mechanical properties of this polymer was not possible. In order to further confirm and elaborate the preferential occurrence of the above-described side reactions and of the resulting tertiary oligo-uret network formation, a completely new type of synthesis was hereby designed and performed, this time reacting the diisocyanate (HDI) with a mono-hydroxy-terminated PEG, namely with PEG350 monomethyl ether (PEG350M) (instead of the diol-terminated PEG used in the synthesis described above). The synthesis was performed under exactly the same conditions as the above-described synthesis, and at a HDI–PEG350M molar ratio of 1:1. Consequently, there are two N=C=O groups for each OH group in the reaction system. Thus, one of the following possibilities may hypothetically occur as a result of the synthesis: In the case that no (or very few) side reactions will occur, only trimers will be formed, and only urethane groups will be present in the final product, along with a large excess amount of unreacted isocyanate groups. In the case that a significant amount of urethane groups remain non-side-reacted—and the remaining excess of free N=C=O groups in the system preferentially further side-react with the significantly smaller amount of already formed allophanate secondary uret groups to form increasingly longer tertiary oligo-uret structures, instead of forming additional new allophanates with the remaining urethane groups—this will conclusively support the above-suggested energetically favorable mechanism for the preferential occurrence of the consecutive side reactions. In this case, the resulting polymer will be a hyper-branched structure, consisting of relatively long PEG350 branches one-sidedly connected to tertiary oligo-uret hierarchical network structures. The synthesis resulted in a soft amorphous polymer. Figure 10 exhibits the FTIR spectrum of the resulting polymer, hereby synthesized with PEG350M and HDI at a 1:1 molar ratio. The almost complete absence of an absorbance peak at 2270 cm −1 indicates that practically all the isocyanate groups in the system have reacted. Two very strong and sharp carbonyl-stretching absorbances are present in the spectrum—the absorbance at 1717 cm −1 belongs to the urethane carbonyl; and the higher-intensity absorbance at 1687 cm −1 belongs to the carbonyls in tertiary oligo-uret structures. A shoulder to the right of this peak, at around 1640 cm −1 , may be clearly observed. As also described above, an absorbance at this location was previously demonstrated in diisocyanate-derived polyureas, as the stretching absorbance of carbonyls in biuret structures [ 37 , 38 ]. As two-sided biurets are actually short tertiary oligo-uret structures, in the present polymer, this absorbance most probably belongs to carbonyls in relatively short tertiary oligo-uret branching—which did not result in final fully developed secondary crosslinking networks. This absorbance shoulder is of a higher intensity than the shoulder at around the same location in the FTIR spectrum of the polymer synthesized with PEG200 ( Figure 1 ). This most probably stems from some steric interference of the relatively long PEG350 hyper-branching with the efficacy of the tertiary oligo-uret secondary crosslinking structures development. Nevertheless, this absorbance is of a much lower intensity than in the FTIR spectrum of the polymer synthesized with PEG400 ( Figure 6 ), as also explained above. The absorbance at 1537 cm −1 belongs to the C–N–H deformation vibrations. It is interesting to note that the intensity of this absorbance is much lower in the present spectrum ( Figure 5 ) than the same absorbance peak in the spectrum of the polymer synthesized with PEG200 ( Figure 1 ). This strongly indicates a predominantly higher abundance of tertiary nitrogens in the present polymer than in the polymer synthesized with PEG200. The absorbance at 3350 cm −1 belongs to the N–H stretching vibrations of the urethane groups, as well as of the two terminal N–H groups at the extremities of each tertiary oligo-uret structure ( Scheme 2 ). The extremely great width of this peak strongly indicates the presence of a wide distribution of lengths and hierarchical configurations of the tertiary oligo-uret structures, as recently demonstrated [ 29 ] and as described above. This absorbance (at 3350 cm −1 ), however, is of a much lower intensity in the present spectrum ( Figure 5 ) than in the spectrum of the polymer synthesized with PEG200 ( Figure 1 ). This, again, strongly indicates a predominantly higher abundance of tertiary nitrogens in the present polymer than in the polymer synthesized with PEG200. As the amount of methylene groups is significantly increased in the present hyper-branched polymer as compared to the polymer synthesized with PEG200 due to the significantly higher molecular weight of the PEG350M, the CH-related absorbances could not be used as internal standards for intensity comparison. Thus, the intensities of the above-discussed NH-related absorbances were qualitatively compared to the intensities of the carbonyl stretching absorbances—the significantly lower intensities of the NH-related absorbances in the hereby hyper-branched polymer, than in the polymer synthesized with PEG200, as related to the intensities of the carbonyl-stretching absorbances of the same polymers, respectively, clearly indicates a significantly higher content of tertiary nitrogens (and a significantly lower NH content) and, thus, a significantly higher content of tertiary oligo-uret structures in the PEG350M-containing polymer. The strong absorbance at 1104 cm −1 belongs to the C–O–C stretching vibrations of the PEG350M ether groups. It is important to also note the complete absence of the adjacent peak at 1050 cm −1 in the present spectrum ( Figure 10 )—this peak was strongly present in the spectrum of the polymer synthesized with PEG200 ( Figure 1 ) which is characteristic of PEG in crystalline form [ 39 ]. The complete absence of this peak in the FTIR spectrum ( Figure 8 ) is in strong agreement with the amorphous nature of the present polymer, synthesized with PEG350M. This is also highly consistent with the fact that the present polymer is a hyper-branched polymer, the significantly long PEG350 branches of which sterically inhibit any close intermolecular or intersegmental interactions. Also, in the present polymer, the PEG branches are not oriented (extended) since they are only chemically anchored at one end—the other end being free—and are, thus, most probably in a relatively disordered configuration. This is also reflected in the location of the urethane carbonyl-stretching absorbance in the FTIR spectrum of the present polymer, at 1717 cm −1 ( Figure 10 ), as compared to its location in the spectrum of the polymer synthesized with PEG200, at 1710 cm −1 ( Figure 1 ). The slight shift to the right (i.e., to a slightly lower wavenumber) of the urethane carbonyl ( Figure 1 ) most probably stems from some hydrogen bonding of these groups, which may be sterically possible in the non-side-reacted urethanes that are chemically bonded to the extended and parallel-oriented crystallized PEG200 soft segments ( Scheme 2 ). In the present hyper-branched polymer, the occurrence of such hydrogen bonds is not possible due to the steric hindrance of the long PEG350 branches. Figure 11 exhibits the solid-state 13 C NMR spectrum of the polymer hereby synthesized with PEG350M and HDI at a 1:1 molar ratio. The two very close, partially overlapping resonances at 25 ppm and at 29 ppm belong to the carbons of the four inner methylene groups of the HDI hexamethylene group (the two central methylenes resonate at 25 ppm and the following two methylenes at 29 ppm). The resonance at around 42 ppm is attributed to the outermost two methylenes of the HDI hexamethylene group, bonded to the nitrogen atoms. This resonance exhibits a shoulder resonance to the right at around 39 ppm. This is a strong indication that these carbons are bonded to two different types of nitrogen atoms, i.e., the tertiary nitrogen and the secondary nitrogen attributed to the 42 ppm and the 39 ppm resonances, respectively. The fact that the resonance at 42 ppm is of a much higher intensity than the resonance at 39 ppm indicates a much higher content of tertiary nitrogens in the polymer than secondary nitrogens. The very small, low-intensity resonance at 57 ppm belongs to the methyl–ether end-group carbon of the PEG350M branches which, as end-groups, are in very low concentration in the polymer. The very strong and sharp resonance at 69 ppm belongs to the two PEG methylenes, bonded to the ether oxygen atoms of the PEG350M branches. The very high intensity of this resonance stems from the high abundance of these groups originating from the PEG350M branches in the polymer. Four types of carbonyl carbon resonances are clearly present in the spectrum ( Figure 11 ). Among these, the resonance exhibiting the strongest intensity and located at 158 ppm is attributed to the carbonyls in the tertiary oligo-uret structures. This, again, indicates the very high content of carbonyls belonging to tertiary oligo-uret structures in the hyper-branched polymer. The adjacent resonance at 155 ppm belongs to the carbonyls in the non-further-reacted urethane groups of the polymer. This is highly consistent with the above-described assumption that the very wide carbonyl carbon resonance in the NMR spectrum in Figure 5 included both the adjacent resonances, at 155 ppm and at 158 ppm, overlapping and in quasi-equal intensities. The adjacent resonance to the right at 148 ppm, as discussed above, belongs to the relatively short tertiary oligo-uret secondary crosslinkers of larger tertiary oligo-uret network structures (as seen in Scheme 2 ). The carbonyls of these short tertiary oligo-uret secondary crosslinking structures comprise a less electron-withdrawing environment and, consequently, a higher electron density of the carbonyls within these structures, thus resonating at a lower ppm. A very small resonance shoulder at 162 ppm belongs to the allophanate urethyl carbonyl carbon, indicating the presence of a small amount of allophanate groups in the polymer that did not further side-react to form tertiary oligo-uret structures. The very high content of tertiary oligo-uret structures in this new polymer, along with the concomitant presence of a significant amount of remaining non-further-reacted urethane groups, indicate and confirm the above-described preferential occurrence of the consecutive side-reactions to form the hierarchical tertiary oligo-uret network structures. These results are also in strong agreement with the results previously obtained by reacting HDI with a mono-amine-terminated molecule (butylamine) at the same molar ratio of 1:1, which also exhibited a predominant and preferential formation of tertiary oligo-uret structures along with a significant amount of remaining non-further-reacted urea groups [ 28 ]. Nevertheless, the polymer hereby synthesized with PEG350M resulted in a relatively soft material, due to the very soft and relatively long PEG350M hyper-branching—whereas the polymer previously synthesized by reacting HDI with the mono-amine-terminated butylamine resulted in a highly rigid, rock-hard, hyper-branched polymer, due to the very short butyl branches, and, thus, most probably a much lower free volume and a much tighter, higher degree of crosslinking. Scheme 3 exhibits a possible schematic representation of the molecular structure of the hyper-branched polymer, synthesized with HDI and PEG350M at a molar ratio of 1:1. Each N–H group in the structure can potentially further side-react with an isocyanate group, creating a tertiary nitrogen and a new secondary nitrogen, which may in turn react with an isocyanate, and so on, creating increasingly complex crosslinked hierarchical hyper-branched networks with multiple PEG350M branches. The fact that, as demonstrated above, a hyper-branched crosslinked polymer network was obtained by reacting a diisocyanate with a mono-hydroxy-terminated molecule at a molar ratio of 1:1 (thus constituting a 2:1 -N=C=O:OH molar ratio), provides a conclusive proof of the preferential and predominant occurrence of the side-reactions leading to the tertiary oligo-uret network structures formation—and without which, this polymer would have been impossible to obtain. This result is also in strong agreement with a similar hyper-branched crosslinked polymer network previously obtained by reacting a diisocyanate with a mono-amine-terminated reagent at the same molar ratio [ 37 ]. Figure 12 exhibits the DSC analysis thermogram of the hyper-branched polymer, synthesized with HDI and PEG350M at a molar ratio of 1:1. As can be clearly observed in the thermogram, the polymer is totally amorphous—which is in complete agreement with the fact that the polymer is highly branched, with significantly long PEG350M branches, thus inhibiting crystallization of this polymer. Nevertheless, the significantly crosslinked tertiary oligo-uret network structure renders the polymer a soft rubbery consistency. Figure 13 exhibits a photograph taken of a test tube containing the hyper-branched polymer synthesized with HDI and PEG350M at a molar ratio of 1:1. The still-hot soft polymer was inserted in test tubes immediately after synthesis, and the sealed test tube was incubated at room temperature and in the dark for at least three months. As observed, the polymer is highly clear and transparent with an apparently very uniform consistency. Nevertheless, some horizontal separation regions are observed along the test tube contents, which are most probably due to the polymer contraction during the incubation period, as the polymer rigidity increased within the incubation period at room temperature. An amount of the still-hot soft polymer was also spread in a large glass Petri dish to form a film; in parallel, an amount of the still-hot soft polymer was spread and pressed between two Kapton sheets—for subsequent mechanical testing. It is interesting to note that these gaps did not occur in the films incubated in the Petri dish or between the Kapton sheets. Nevertheless, the polymer exhibited an extremely strong adherence to both the glass surface and to the Kapton sheets; thus, it could not be non-destructively separated from neither the glass nor the Kapton surface—making mechanical testing impossible at this stage. It is interesting to note, though, that the rubbery film obtained between the two Kapton sheets exhibited very pronounced flexibility. After repeatedly bending the film into a U-shape and releasing the hold, the film repeatedly jumped back to its original flat shape. The observed very strong adherence of the hyper-branched polymer to significantly different types of surfaces may be explained by the very high polarity of the tertiary oligo-uret networks, combined with the well-known surfactant property of the multiple PEG branches. In addition, by proving the feasibility and mechanism of occurrence of these diisocyanate-derived mono-amine- and mono-hydroxy-mediated synthesis pathways, a completely new family of hyper-branched crosslinked polymers may, hence, be further developed, with a wide variety of potential biomedical and pharmaceutical applications. As also indicated in the Introduction section, despite the very long-known tendency of diisocyanates to side-react with NH groups of urethanes and ureas to form allophanates and biurets, respectively [ 1 , 2 , 3 , 26 ], very numerous polyurethane synthesis-related research publications over many decades attributed the occurrence of two, and even multiple, distinct carbonyl-stretching absorbances in the FTIR spectra of polyurethanes to a combined presence and absence of hydrogen bonding between some portions of the urethane groups—without even considering the very possible occurrence of side reactions and, consequently, the possible formation of chemical structures containing additional carbonyl types, thus exhibiting different adjacent carbonyl-stretching absorbance locations [ 27 , 28 , 29 , 30 , 31 , 32 , 33 , 34 ]. As also described here above, four recent research works [ 35 , 36 , 37 , 38 ] have conclusively demonstrated the highly preferential occurrence of further consecutive side reactions of allophanates and biurets with additional isocyanate groups, resulting in highly complex hierarchical crosslinking networks of tertiary oligo-uret structures. The characteristic FTIR carbonyl-stretching absorbance of this structure was repeatedly demonstrated to accurately and consistently occur at 1687 cm −1 in all these recent research works [ 35 , 36 , 37 , 38 ] and further in the present research. The individual carbonyl-stretching FTIR absorbances of urethane [ 36 ], allophanate [ 35 ], urea [ 36 , 37 , 38 ], and biuret [ 37 , 38 ] were determined by using different diisocyanate molar ratios (of between 1:1 and 1:6) [ 35 , 36 , 37 , 38 ] or by sterically inhibiting side reactions and/or further tertiary oligo-uret formation [ 36 , 37 , 38 ] and, also, via the deliberate synthesis of a new polymer inherently exhibiting oligo-uret structures in the polymer repeating unit [ 38 ]. It was, thus, also demonstrated that the isolated individual carbonyl-stretching FTIR absorbances of urethane groups (at 1717 cm −1 ), urea groups (at 1621 cm −1 ), and biuret groups (at 1637 cm −1 ) each consisted of a single sharp absorbance peak—all of which were not split [ 36 , 37 , 38 ]. The carbonyl-stretching FTIR absorbance of the allophanate structures, though, exhibited two distinct carbonyl-stretching absorbance peaks, representing the two types of carbonyls in this structure—i.e., the urethyl carbonyl and the uret carbonyl of the allophanate structure (at 1805 cm −1 and at 1770 cm −1 respectively) [ 35 ]. A shifting effect of hydrogen bonding on the location of the urethane carbonyl-stretching FTIR absorbance is commonly suggested to occur in the range of 1705 cm −1 –1725 cm −1 , but as two distinct absorbance peaks—one due to the non-hydrogen-bonded urethane carbonyls and the other due to hydrogen-bonded carbonyls [ 62 ]. The inevitable question that arises here is why a hydrogen bonding-/non-bonding-related peak-splitting effect is mainly reported in relation to diisocyanate-derived polymers, such as polyurethanes and poly(urethane urea)s, and not in other polymers which also exhibit a very high ability of hydrogen bonding, such as polyamides—in which there are no side reactions that can alter the final polymer chemistry. In a very recent publication entitled Infrared Spectroscopy of Polymers XIII: Polyurethanes [ 62 ], the author states the following: “The spectra of polyurethanes are similar to polyamides since both polymer types contain C=O and N-H bonds. Therefore, we will begin with a review of the spectra of polyamides. The structure and infrared spectra of polyurethanes and polymeric amides or polyamides are similar, hence a review of polyamides is in order” [ 62 ]. Accordingly, in order to further examine the often suggested paradigm of combined hydrogen bonding and non-bonding as being the source of two or multiple distinct FTIR carbonyl-stretching absorbances in polyurethanes, an additional synthesis was hereby performed, leading to NH- and C=O-containing structures, but without the possibility of occurrence of any side reactions. Accordingly, nylon 6,6 polyamide was hereby synthesized via interphase polymerization at room temperature (with adipoyl chloride and hexamethylenediamine). Absolutely no side reactions occur via the interphase polymerization of this polymer—hence, only one type of carbonyl exclusively exists in the polymer. Also, highly abundant hydrogen bonding is likely to occur in this polymer, due to the very high content of amide groups along the polymer chain. Nevertheless, as is well-known in all linear (non-crosslinked) polymers, at least a certain degree of chain entanglements is almost inevitably expected to occur. This, along with chain-folding regions, is among the main reasons that crystallizable polymers (nylon 6,6 included) are always semicrystalline—and can never reach full crystallinity (except in some polymeric fibers) [ 31 , 32 , 33 , 34 , 35 , 36 , 37 ]. Thus, and for the same reasons, it is highly reasonable and expected that not all the amide groups of the presently synthesized nylon 6,6 would be able to create hydrogen bonds, considering the very short distance range of effective hydrogen bonding (which is only slightly longer than a covalent bond). Accordingly, if the hydrogen bonding/non-bonding carbonyl-stretching FTIR peak-splitting paradigm is valid, as often suggested for polyurethanes, the FTIR spectrum of the nylon 6,6 should exhibit at least two distinct adjacent carbonyl-stretching absorbances—although only one type of carbonyl exists in the polymer. Figure 14 exhibits the FTIR spectrum of the nylon 6,6 hereby synthesized via interphase polymerization. A single extremely thin and sharp carbonyl-stretching absorbance appears in the spectrum at 1637 cm −1 —which strongly confirms that only a single type of carbonyl is present in the polymer, i.e., the amide carbonyl. Also, a single extremely thin and sharp NH-stretching absorbance at 3304 cm −1 and a very thin and sharp CNH deformation absorbance at 1537 cm −1 are clearly observed in the spectrum of the polymer—again indicating that only one type of NH is present in the polymer, i.e., the amide NH. This strongly confirms that when only one type of carbonyl is present in the polymer—in this case, the amide carbonyl—a single carbonyl-stretching absorbance peak is present in the spectrum. Hydrogen bonding between amide groups—which is most probably present in nylon 6,6—most probably does cause a shift in the carbonyl-stretching absorbance, but it certainly does not cause any splitting of the peak. As described above, this was also recently demonstrated for diisocyanate-derived polymers containing only urethane [ 36 ], only biuret [ 37 ], or only urea [ 37 , 38 ] groups, each exhibiting a single sharp carbonyl-stretching FTIR absorbance. As also observed in the present research, the hydrogen bonding of the non-side-reacted portion of urethane groups did cause a shift in the urethane carbonyl-stretching absorbance (from the usual 1717 cm −1 location to 1710 cm −1 , also exhibiting a thickening or slight shoulder of the peak at around 1717 cm −1 )—but not the splitting of the peak. The tertiary oligo-uret carbonyl-stretching absorbance at 1687 cm −1 did not shift (as it never does) since the structure does not have a polar hydrogen, in view of the fact that all nitrogens in the structure are tertiary nitrogens and, therefore, do not create hydrogen bonds. Figure 15 exhibits a summary comparison of the FTIR spectra of the hereby synthesized poly(ether urethane) types, namely the following: the poly(ether urethane), synthesized with HDI and PEG200 at a molar ratio of 1:1 (a); the poly(ether urethane), synthesized with HDI and PEG400 at a molar ratio of 1:1 (b); the hyper-branched polymer, synthesized with HDI and PEG350M at a molar ratio of 1:1—thus exhibiting a –N=C=O:OH molar ratio of 2:1 (c). The tertiary oligo-uret carbonyl-stretching absorbances accurately and consistently appear at 1688 cm −1 in all the spectra and are perfectly aligned with the connecting red dashed line, annotated in Figure 15 —these resonances are strongly consistent, with the exact same location of the tertiary oligo-uret FTIR carbonyl-stretching absorbance previously reported in polyurethanes, polyureas, hyper-branched polyurea [ 35 , 36 , 37 , 38 ], and in poly(hexamethylene oligo-uret) [ 38 ]. When viewing the adjacent urethane carbonyl-stretching absorbance, a shift from 1717 cm −1 ( Figure 15 b,c) to around 1710 cm −1 (with a slight shoulder/thickening at around 1717 cm −1 ) is observed, which most probably occurs due to hydrogen bonding, which is consistent with and within the very recently reported range for the hydrogen bonding-related shift of the urethane carbonyl-stretching absorbance [ 60 ]. The urethane carbonyl-stretching absorbance in the spectrum is observed to shift within this range ( Figure 15 a)—but no splitting is observed ( Figure 15 a–c). The strong and sharp absorbance at 1050 cm −1 to the right of the main ether absorbance is easily observed in the FTIR spectrum of the poly(ether urethane), synthesized with HDI and PEG200 at a molar ratio of 1:1 ( Figure 15 a), characteristic of PEG in the crystalline state, as also discussed above. Also, the spectrum of the same polymer exhibits a very much increased intensity of the asymmetric CH-stretching absorbance at 2950 cm −1 , as compared to the symmetric CH-stretching absorbance at 2850 cm −1 , which is highly consistent with the very significant chain alignment in this polymer and the consequent parallel extended-chain interaction, predominantly enabling the asymmetric CH-stretching vibration, as also discussed here above. The unique structure and properties combination of the presently developed segmented polyurethane, alongside the significantly advantageous single-step synthesis pathway hereby performed, may lead to the further development of segmented polyurethanes of enhanced performance and to improved industrial processing efficacy for a variety of possible applications, ranging from various medical devices to ballistic impact-resistant devices and composite matrices. The mechanisms of the side reactions that occur are demonstrated in the present research, and their effect on the properties of segmented polyurethanes can facilitate the optimization of industrial processes and potentially open a new door to the development of segmented polyurethanes exhibiting a wide range of novel properties and applications."
} | 18,801 |
37891652 | PMC10612212 | pmc | 8,208 | {
"abstract": "Background Whole-cell biocatalysis has been exploited to convert a variety of substrates into high-value bulk or chiral fine chemicals. However, the traditional whole-cell biocatalysis typically utilizes the heterotrophic microbes as the biocatalyst, which requires carbohydrates to power the cofactor (ATP, NAD (P)H) regeneration. Results In this study, we sought to harness purple non-sulfur photosynthetic bacterium (PNSB) as the biocatalyst to achieve light-driven cofactor regeneration for cascade biocatalysis. We substantially improved the performance of Rhodopseudomonas palustris -based biocatalysis using a highly active and conditional expression system, blocking the side-reactions, controlling the feeding strategy, and attenuating the light shading effect. Under light-anaerobic conditions, we found that 50 mM ferulic acid could be completely converted to vanillyl alcohol using the recombinant strain with 100% efficiency, and > 99.9% conversion of 50 mM p -coumaric acid to p -hydroxybenzyl alcohol was similarly achieved. Moreover, we examined the isoprenol utilization pathway for pinene synthesis and 92% conversion of 30 mM isoprenol to pinene was obtained. Conclusions Taken together, these results suggested that R. palustris could be a promising host for light-powered biotransformation, which offers an efficient approach for synthesizing value-added chemicals in a green and sustainable manner. Graphical Abstract \n Supplementary Information The online version contains supplementary material available at 10.1186/s13068-023-02410-3.",
"conclusion": "Conclusion In summary, we have developed PNSB-based biocatalysts for light-driven cofactor regeneration to power the whole-cell catalysis. Using a highly active expression system, blocking the side-reactions, controlling the feeding strategy and attenuating the light shading effect, both lignocellulose derivatives of FA and p CA were, respectively, transformed into 7.7 g/L VA and 6.21 g/L p HBA with ~ 100% conversion under light-anaerobic conditions. Besides, 1.88 g/L pinene with 92% conversion was achieved from isoprenol using IUP. Compared to traditional whole-cell biocatalysis, the autotrophic R. palustris as a biocatalyst with cofactor regeneration powered by light instead of carbohydrates, could significantly reduce the cost for future industrial development.",
"discussion": "Discussion Biocatalysis is considered to be a sustainable and environment-friendly alternative to chemical synthesis. Although cyanobacterium affords encouraging source of reducing power of NADPH [ 31 , 32 ], it is not ideal for NADH regeneration. Given its adaptable metabolism and ability to catabolize a wide variety of feedstocks such as hexose, pentose, volatile organic acid, aromatics, and etc., R. palustris has been studied and applied in biodegradation of aromatic compounds, environmental remediation, biofuel production, agricultural biostimulation, and bioelectricity production [ 16 ]. In this study, we have developed R. palustris as the whole-cell biocatalyst, and harnessed the light to power cofactor regeneration (ATP and NADH) for biotransformation applications. The performance of PNSB-based biocatalysis was substantially improved using a highly active and conditional expression system, blocking the side-reactions, controlling the feeding strategy and attenuating the light shading effect. When compared to the traditional heterotrophic microorganisms that require additional carbohydrates for cofactor regeneration, photoautotrophic PNSB-based biocatalyst might avoid competition with food industry and significantly reduce the operating cost. We demonstrated light-powered biotransformation for VA and p HBA productions using the endogenous CoA-dependent non-β-oxidation route of R. palustris together with NADH-dependent ADH2 from S. cerevisiae . Both FA and p CA were efficiently converted to the corresponding “C n-2 ” alcohols, with ~ 100% conversion under light-anaerobic conditions. The main ways to VA synthesis are via direct extraction from host plants, and chemosynthesis from vanillin (e.g. Pd/C, Pt/C, and Au/C) [ 33 ]. Recently, whole-cell bioconversion of 1 g/L 3,4-dihydroxybenzyl alcohol to 499.36 mg/L VA with 45.1% conversion was obtained by E. coli [ 34 ], and 3.89 g/L VA was produced from glucose and glycerol by E. coli in fed-batch fermentation [ 35 ]. Traditional p HBA biosynthesis is either extracted from Gastrodia elata or obtained as the intermediate metabolite [ 36 ]. We previously achieved the biotransformation of 5 mM L -tyrosine to 581 mg/L p HBA with 93.6% conversion using E.coli whole-cell catalysis [ 37 ]. When compared to previous studies, we have reached better titers of VA (7.7 g/L) and p HBA (6.21 g/L), suggesting that R. palustris is an appealing and feasible host for light-powered lignocellulose valorization. In the future, it will be possible to engineer R. palustris to synthesize diverse aromatic compounds such as vanillin, vanillylamine [ 38 ], protocatechuic acid and gallic acid [ 12 ] in a similar manner. In addition, we also developed a biocatalytic system for pinene synthesis from isoprenol utilizing IUP. It is reported that an engineered whole-cell catalytic system of E. coli improved pinene titer to 166.5 mg/L using sucrose as the carbon source and the substrate, and 0.97 g/L pinene was produced from glucose by an engineered E. coli under fed-batch fermentation conditions [ 39 ]. In this study, 1.88 g/L pinene with 92% conversion was achieved, which was much higher than the titers obtained by traditional biocatalysis and metabolic engineering efforts [ 39 , 40 ]. For the future production of other terpenes such as sesquiterpenes, it will require a proper balance of DMAPP and IPP levels by adjusting Idi expression or simply use a fixed ratio of prenol and isoprenol without introducing Idi-mediated isomerization to achieve the theoretical maximum. To further improve light-powered biotransformation, we also engineered the carotenoid biosynthetic pathway of R. palustris to weaken the light shading effect and improve the transmission of light through the cell culture. Our results confirmed that the engineered strains with lower light absorption could accelerate the conversion process. It was reported that disruption of puc operon [ 41 ] or overexpression of pufQ , a regulatory gene to puf operon [ 42 ], could decrease the pigment contents with low light absorption, which might be conducive to ATP and NADH synthesis. Due to photophosphorylation coupling with electron transport during photosynthesis, slight overexpression of cycA encoding for cytochrome c 2 , an important electron carrier in electron transport chain, was found to improve ATP and NADH synthesis [ 43 ]. In addition, transposon mutagenesis screening has identified several mutants with reduced pigments [ 44 , 45 ]. These engineering strategies might be similarly implemented in our light-powered biotransformation to further improve the strain performance in the future work."
} | 1,759 |
22882210 | null | s2 | 8,210 | {
"abstract": "Biofilm formation in Bacillus subtilis requires the differentiation of a subpopulation of cells responsible for the production of the extracellular matrix that structures the biofilm. Differentiation of matrix-producing cells depends, among other factors, on the FloT and YqfA proteins. These proteins are present exclusively in functional membrane microdomains of B. subtilis and are homologous to the eukaryotic lipid raft-specific flotillin proteins. In the absence of FloT and YqfA, diverse proteins normally localized to the membrane microdomains of B. subtilis are not functional. Here we show that the absence of FloT and YqfA reduces the level of the septal-localized protease FtsH. The flotillin homologues FloT and YqfA are occasionally present at the midcell in exponentially growing cells and the absence of FloT and YqfA negatively affects FtsH concentration. Biochemical experiments indicate a direct interaction between FloT/YqfA and FtsH. Moreover, FtsH is essential for the differentiation of matrix producers and hence, biofilm formation. This molecular trigger of biofilm formation may therefore be used as a target for the design of new biofilm inhibitors. Accordingly, we show that the small protein SpoVM, known to bind to and inhibit FtsH activity, inhibits biofilm formation in B. subtilis and other distantly related bacteria."
} | 337 |
38830860 | PMC11148140 | pmc | 8,211 | {
"abstract": "Plastic waste is an environmental challenge, but also presents a biotechnological opportunity as a unique carbon substrate. With modern biotechnological tools, it is possible to enable both recycling and upcycling. To realize a plastics bioeconomy, significant intrinsic barriers must be overcome using a combination of enzyme, strain, and process engineering. This article highlights advances, challenges, and opportunities for a variety of common plastics.",
"introduction": "Introduction Plastic has become an ever-present and ubiquitous fixture in our daily lives. Due to their varied characteristics, plastics have a wide range of applications from medical implants to food packaging. Owing to a low cost and rapid production, plastics quickly became single-use items and these same features perpetuate their overuse. Since the commercial introduction of plastics in the 1950s, yearly plastic production has increased over one hundred-fold with cumulative plastic production estimated to be in the tens of thousands of million metric tons by 2050 1 . At current estimates, nearly 400 million tons of plastic will be produced this year—a number equivalent to the collective mass of every human on the planet. Across all this plastic, only about 14% is ultimately recycled 2 . Unrecovered plastics that end in landfills or oceans can persist, leading to an accumulation of plastic in the natural environment and ultimately causing severe environmental consequences 3 . As an alternative, this plastic waste can serve as a potent feedstock for both bio-enabled recycling and upcycling."
} | 391 |
23719585 | PMC4070969 | pmc | 8,212 | {
"abstract": "Arbuscular mycorrhizal fungi (AMF) in an agricultural ecosystem are necessary for proper management of beneficial symbiosis. Here we explored how the patterns of the AMF community in rice roots were affected by rice cultivation systems (the system of rice intensification [SRI] and the conventional rice cultivation system [CS]), and by compost application during growth stages. Rice plants harvested from SRI-managed plots exhibited considerably higher total biomass, root dry weight, and seed fill than those obtained from conventionally managed plots. Our findings revealed that all AMF sequences observed from CS plots belonged (only) to the genus Glomus , colonizing in rice roots grown under this type of cultivation, while rice roots sown in SRI showed sequences belonging to both Glomus and Acaulospora . The AMF community was compared between the different cultivation types (CS and SRI) and compost applications by principle component analysis. In all rice growth stages, AMF assemblages of CS management were not separated from those of SRI management. The distribution of AMF community composition based on T-RFLP data showed that the AMF community structure was different among four cultivation systems, and there was a gradual increase of Shannon-Weaver indices of diversity ( H ′) of the AMF community under SRI during growth stages. The results of this research indicated that rice grown in SRI-managed plots had more diverse AMF communities than those grown in CS plots.",
"conclusion": "Conclusions Different cropping systems, alternating flooding and draining cycles of water management, affected AMF community structure in paddy fields. The results revealed that rice plants grown under SRI had a more diverse AMF community than those grown under CS conditions.",
"discussion": "Discussion Although wetland rice has previously been considered non-mycorrhizal, a positive response to AMF inoculation has been observed ( 29 ). Our results presented here ( Table 1 ) showed that AMF associated in rice roots could colonize under waterlogged conditions (anaerobic conditions), and there was an increase of the AMF colonization rate over the sampling times. Solaiman and Hirata ( 30 ) also reported that 28% colonization by Glomus spp. at six weeks of growth (after growing in a dry nursery) persisted in a wetland. Although AMF is an obligate aerobic in nature, it can probably survive in association with rice roots under anaerobic conditions because it obtains O 2 from the atmosphere through rice aerenchymatous tissue. It was found in this study that AMF diversity differed among cultivation treatments. The changes in AMF diversity in response to the different cultivation systems and the application of compost might be due to several factors, such as pH, nutrient content, total soil C and N, and temperature, which are known to influence AMF distribution ( 11 ). In addition, organic matter addition was able to increase AMF hyphae growth ( 17 , 14 ). Sooksa-nguan and colleagues ( 32 ) also revealed changes in microbial communities in which differences in rice cultivation systems affected the structures of Bacteria and Archaea communities. This suggest that the SRI practice, where flooding was alternated with periods of draining, allowed for higher O 2 availability in the soil with higher AMF establishment than conventionally managed practice, while the addition of compost might favor nutrients as well as being a growth condition that could induce diversity of AMF. The high level of AMF diversity in SRI practice was correlated with a clear increase of the yield components, such as total biomass, root dry weight, and seed fill of rice grown under SRI management ( Table 3 ). Similar observations of improved performance of rice yield with wider spacing under SRI have also been reported previously ( 18 , 24 ). A previous report indicated that maintenance of a high CO 2 assimilation rate via a delay in leaf senescence is an important factor that can increase the crop yield of SRI ( 6 ). Younger seedlings used with SRI perform better in terms of various root characteristics (root length density and root weight density) than older seedlings ( 25 ). SRI water management practices also help in improving root systems ( 4 ), while continuous flooding can cause degeneration of as much as three-fourths of roots by the flowering stage ( 15 ). In addition, the high level of AMF colonization ( Table 1 ) was most likely correlated with high P uptake in rice plants grown under SRI management ( Table 2 ); however, some treatments did not show a correlation between AMF colonization and high P uptake. For example, at 30 d, AMF colonization under CS control treatment was lower than that in SRI-control, but P uptake in CS-control was higher than in SRI-control. It is possible that rather than AMF colonization, other factors were more effective in causing differences in P uptake among the treatments, such as the rice growth stage and compost treatment. Thus, diverse AMF colonization under SRI management most likely played a role in enhancing the P concentration and uptake, and thereby improving rice growth, showing that the enhancement of P uptake in response to AMF infection may have been the result of the increased absorption of nutrients due to the greater surface area of AMF hyphae extending from the roots. The outcome observed here, that the highest P uptake was detected at 90 d ( Table 2 ), suggested that rice plants at the end of the panicle initiation stage may be undergoing full growth and demand large amounts of nutrients. Although the colonization of AMF was effective on P uptake and promoted rice yield components, the grain yields were not higher in SRI-managed rice than in CS plots. It is possible that the AMF species in these communities and their effectiveness may significantly affect nutrient availability and plant growth. Moreover, AMF may be more effective in the vegetative stage, rather than for reproductive growth. Our finding was similar to some upland rice-growing regions where rice yield was reduced in SRI-managed plots ( 26 ). Thus, AMF in SRI management may have a role in nutrient uptake into rice before plants enter the reproductive stage. The majority of sequences detected in rice roots belonged to the genus Glomus . Other reports on the molecular diversity of AMF also showed that Glomus was the dominant genus in AMF communities in different ecosystems ( 7 , 10 , 40 ). The sequences clustering in the genera Acaulospora were found only in the plots managed under SRI, and were not detected under CS conditions. The occurrence of Glomus groups in rice roots grown in CS plots suggests that these fungi might be more tolerant of anaerobic conditions. As in previously reported studies carried out for other eco-systems such as tropical forests ( 11 ), agricultural sites ( 10 ), wetland soils ( 40 ), gypsum soils ( 1 ), or polluted soils ( 37 ), they also found an AMF community dominated by Glomus species. Nonetheless, it remains unknown why Glomus species are better adapted to disturbed environments. Further analyses are needed to confirm whether differences in water management may favor particular genera or species of AMF over others. It is also necessary to investigate whether any changes in species composition may be significantly affected by nutrient availability and plant growth since the AMF community differed not only by cultivation system, but also by the application of compost. Although two major AMF in the genera Glomus and Acaulospora were detected from rice roots with or without compost application ( Fig. 1 ), the species diversity of these two genera may be different as PCA, which was derived from different RFLP patterns of AMF sequences obtained from the clone library exhibited a difference in AMF community composition between control and compost applications ( Fig. 3 ). It was also found that Glomus (JF906743) and Glomus (JF906749) were only detected in S(b) treatments ( Fig. 1 ). Moreover, the diversity of the AMF community was also the highest in SRI with compost treatment ( Fig. 4 ). This result supported the hypothesis that the nutrient content or total soil C and N may influence AMF distribution, especially when rice was grown under SRI management that had more O 2 available than in the conventional system. In addition, Glomus sp. tends to be easily detected more than other AMF genera, even in other environments. Thus, it is possible to detect many species of the genus Glomus compared with other genera. Here, we questioned whether differences in diversity between communities can be meaningfully portrayed by common diversity indices applied to T-RFLP profiles derived from AMF communities. Our results demonstrated diversity estimates presented by TRF richness (number of TRFs in profile), H ′ and evenness (frequency distribution of TRFs) of AMF communities derived from T-RFLP data. Interestingly, a high percentage of AMF colonization was observed for the high diversity of the AMF community in roots growing in SRI plots. SRI plots sampled at 90 and 120 d revealed the highest level of colonization and showed higher AMF diversity (based on H ′ derived from SSU) than the other plots at 90 d ( Table 1 , Fig. 4 and also see Table S1 ). High AMF diversity was reported in the dry Afromontane forest ( H ′=2.58 based on ITS) and in a tropical forest ( H ′=2.33 based on SSU) ( 11 ). There was higher AMF diversity in the SRI plot, potentially because higher AMF diversity may reflect the high colonization of AMF in aerobic rice field plots. T-RFLP can be applied as a quantitative method to estimate the relative abundance based on peak heights or peak areas ( 20 ). Analyses by multidimensional scaling ordination ( Fig. 3 ) showed that PCA exhibited a difference in AMF community composition among cultivation managements and compost applications, and marked shifts in composition were detected by multidimensional scaling ordination based on abundance data. Moreover, in the results of the rarefaction curve, the expected number of AMF species in rice roots was higher in the plots of CS-control than in CS-compost ( Fig. 2A ), which showed the same trend as described in the phylogenic tree ( Fig. 1 ). However, as the analysis described the T-RFLP peak number, it was implied that the diversity and species number of AMF was higher in CS-compost than in CS-control ( Fig. 4A ). Disagreement of the results suggests that the marked shifts in community compositions might have been biased by the PCR approach. The limitations of T-RFLP for quantitative characterization of communities have been well argued. These problems become more evident when working with complex ecosystems in which diversity of the microorganisms is high, with differences in the ability to extract DNA from different organisms, genome size, and G+C content ( 33 , 22 ). Almost every step (PCR amplification or digestion or TRF run) of the T-RFLP technique can introduce biases or errors to the analysis ( 38 ). PCR amplification of a species is known to be directly influenced by the presence of other species in the PCR mixture, affecting the apparent abundance of a species in the PCR reaction. As a result of these biases, representation of the real microbial community and diversity in ecosystems will probably not be possible; however, the T-RFLP facilitates the separation of mixtures of PCR-amplified gene fragments based on terminal restriction fragments (T-RFs) and allows large numbers of samples to be analyzed simultaneously ( 35 ). Thus, this technique is still suitable for monitoring the changes in microbial communities in response to changes in environmental factors."
} | 2,930 |
39006488 | null | s2 | 8,215 | {
"abstract": "Under current trends, 60% of India's population (>10% of people on Earth) will experience severe food deficiencies by 2050. Increased production is urgently needed, but high costs and volatile prices are driving farmers into debt. Zero budget natural farming (ZBNF) is a grassroots movement that aims to improve farm viability by reducing costs. In Andhra Pradesh alone, 523,000 farmers have converted 13% of productive agricultural area to ZBNF. However, sustainability of ZBNF is questioned because external nutrient inputs are limited, which could cause a crash in food production. Here, we show that ZBNF is likely to reduce soil degradation and could provide yield benefits for low-input farmers. Nitrogen fixation, either by free-living nitrogen fixers in soil or symbiotic nitrogen fixers in legumes, is likely to provide the major portion of nitrogen available to crops. However, even with maximum potential nitrogen fixation and release, only 52-80% of the national average nitrogen applied as fertilizer is expected to be supplied. Therefore, in higher-input systems, yield penalties are likely. Since biological fixation from the atmosphere is possible only with nitrogen, ZBNF could limit the supply of other nutrients. Further research is needed in higher-input systems to ensure that mass conversion to ZBNF does not limit India's capacity to feed itself."
} | 342 |
38350021 | PMC10910447 | pmc | 8,216 | {
"abstract": "Liquid-like surfaces\nfeaturing slippery, omniphobic, covalently\nattached liquids (SOCALs) reduce unwanted adhesion by providing a\nmolecularly smooth and slippery surface arising from the high mobility\nof the liquid chains. Such SOCALs are commonly prepared on hard substrates,\nsuch as glass, wafers, or metal oxides, despite the importance of\nnonpolar elastomeric substrates, such as polydimethylsiloxane ( PDMS ) in anti-fouling or nonstick applications. Compared to\npolar elastomers, hydrophobic PDMS elastomer activation\nand covalent functionalization are significantly more challenging,\nas PDMS tends to display fast hydrophobic recovery upon\nactivation as well as superficial cracking. Through the extraction\nof excess PDMS oligomers and fine-tuning of plasma activation\nparameters, homogeneously functionalized PDMS with fluorinated\npolysiloxane brushes could be obtained while at the same time reducing\ncrack formation. Polymer brush mobility was increased through the\naddition of a smaller molecular silane linker to exhibit enhanced\ndewetting properties and reduced substrate swelling compared to functionalizations\nfeaturing hydrocarbon functionalities. Linear polymer brushes were\nverified by thermogravimetric analysis. The optical properties of PDMS remained unaffected by the activation in high-frequency\nplasma but were impacted by low-frequency plasma. Drastic decreases\nin solid adhesion of not just complex contaminants but even ice could\nbe shown in horizontal push tests, demonstrating the potential of\nSOCAL-functionalized PDMS surfaces for improved nonstick\napplications.",
"conclusion": "3 Conclusions We were able to overcome the\nmajor challenges in PDMS activation of fast hydrophobic\nrecovery and glassy-layer formation\nby substrate extraction and optimized plasma activation. This resulted\nin chemically homogeneous, slippery surfaces and preserved the optical\nproperties of the specimen. High transparency and self-cleaning behavior\npose benefits, for instance, in energy applications, such as solar\ncell coatings. 51 Superficial cracking was\nminimized as silica-like layer formation was suppressed, and mechanical\nsurface homogeneity and uniform stability were preserved, which could\nallow the deposition of homogeneous films for flexible sensors. Our\nreport focused on preparing liquid-like surfaces comprising siloxane-based\npolymer brushes on elastomeric PDMS by applying the grafting-from\nconcept. As we introduced superficial polymers postcuring, the set\nmechanical properties of the established formulations were not compromised.\nDifferent decomposition mechanisms allowed the confirmation of surface\npolymer brushes even when the chemical composition was comparable.\nTheir slippery character was enhanced by the addition of a molecular\nlinker in our Mix coating. Proof-of-concept testing showed\nexcellent repellency of solid contaminants and revealed the potential\nfor anti-icing applications as ultralow ice adhesion strengths were\nobtained, which allow for the passive removal of accumulating ice.\nSuch qualities also prove valuable in reducing scaling, 17 improving flow in PDMS -based devices, 65 and minimizing biofouling, 66 the latter of which especially poses a significant challenge\non PDMS surfaces. 11 , 67 We anticipate that\nminimized superficial cracking and simple surface modification to\nobtain slippery interfaces can alleviate the current constraints in PDMS applications.",
"introduction": "1 Introduction High\nflexibility and stretchability, strong insulating properties,\nand chemical robustness render elastomeric materials interesting for\na wide spectrum of applications ranging from molding applications, 1 to flow 2 , 3 and biomedical devices. 4 Among these materials, polydimethylsiloxane ( PDMS ) stands out for its heat resistance, exceptional optical\ntransparency, and rapid fabrication. 5 Surface-exposed\nmethyl groups on the siloxane backbone cause PDMS to\nbe inherently hydrophobic and nonpolar. This is crucial for certain\nuses as it increases biocompatibility in medical devices 6 and contributes to PDMS ′\noverall chemical inertness. 7 However, this\ncomes with the drawback of increased solubility for organic solvents\nor small lipophilic molecules 8 and swelling\nupon contact. 9 Also, the flow properties\nof aqueous solutions in microfluidics are impeded, 10 and surfaces experience a higher chance of biofouling. 11 There is therefore a need to customize PDMS ′ surface properties depending on its intended\napplication. Recent research suggests that integrating nanoscale\nliquid behavior\nto solid surfaces is essential to enhance liquid sliding, 12 , 13 minimize adhesion of contaminants, 14 − 16 reduce biofouling, 17 or improve intermembrane transport. 18 Conventionally, achieving a liquid interface\non PDMS has been done by swelling of the PDMS network with lubricants, 19 thus creating\nSLIPS ( Figure 1 a).\nWhile SLIPS can reestablish their surface and heal once disturbed\nas the fluid flows back after displacement 14 , 20 and are virtually defect-free, they are also prone to lubricant\ndepletion and thus have limited durability, as their lubricant retention\nrelies on physical rather than covalent interactions. 21 , 22 Another approach is to incorporate hemitelechelic polymers, which\nhave a reactive end group, into the bulk polymer matrix to endow lubricating\nproperties, but this requires laborious coating or gel engineering\nof formulations to obtain the desired mechanical properties. 23 Further, this results in the modification of\nthe entire bulk of the material, rather than in engineering the material’s\nsurface properties. An ideal approach would be to develop a method\nsuitable for equipping prefabricated elastomers with liquid interfaces\nto expand their potential range of applications. Figure 1 Different strategies\nto achieve a liquid interface on a solid material.\n(a) Lubricant wetting of a porous substrate to achieve slippery liquid-infused\nporous surfaces (SLIPS) and (b) covalent surface functionalization\nfor interfacial liquid behavior. One route for this would be covalent attachment of lubricating\nmolecules such as linear PDMS to the substrate surface 24 ( Figure 1 b). Compared to carbon-based polymers, the siloxane backbone\noffers great rotational flexibility 18 , 25 as well as\nlow glass transition temperatures, thus tending toward liquid-like\nbehavior at room temperature. 25 , 26 Current research has\nfocused predominantly on hard and polar substrates, 27 such as silicon wafers, 12 , 13 , 16 , 22 , 26 glass, 13 , 16 , 22 aluminum, 22 or stainless steel substrates, 18 as they offer uniform and smooth surfaces, avoiding liquid\npinning problems. Their already oxygen-rich surfaces additionally\nfacilitate chemical functionalization. Two key methods are usually\nemployed to achieve liquid-like surfaces (LLS) on nonporous hard substrates:\nthe first route involves the grafting-to of polymers that bear one\nreactive chain end group, typically monofunctionally terminated silicone\noil 28 , 29 or even unreactive silicone oil. 25 However, the chain lengths of the grafted polymers\ninherently limit the resulting liquid film thickness, and their steric\nhindrance can obscure potential grafting sites, thereby reducing surface\ncoverage. 30 , 31 In the second route, a grafting-from case,\nsmaller molecular precursors, usually various silane precursors, are\nused to grow longer chains on the substrate, 29 allowing for higher grafting densities by avoiding diffusion obstruction. 13 , 16 , 29 , 30 A salient consideration is the type of silane precursor used—silanization\nof PDMS typically involves silanes bearing either one or three hydrolyzable\ngroups. This results in the grafting of a silane monolayer or the\ngeneration of a cross-linked, immobile siloxane multilayer, 32 , 33 both of which limit LLS formation. To avoid cross-linking, achieve\nlinear polymerization, and maximize subsequent interfacial slip, silanes\nwith two hydrolyzable groups are instead required. 32 , 34 The remaining two organic functional groups can then bear different\nmoieties, so tunable functionalities can be integrated into the polymer\nbrushes based on the precursor used, which we demonstrate in this\nwork. However, adapting these concepts to soft substrates such\nas PDMS requires careful consideration. Hydrophobic PDMS requires activation prior to chemical functionalization. 35 Plasma or UV/ozone introduces the necessary\nfunctional groups, such as hydroxyl groups, accessible for chemical\nbinding, but the oxidizing environment frequently leads to the formation\nof a hard silica-like layer on soft PDMS substrates. 36 Activated PDMS is susceptible to\nsurface cracking, especially during mechanical deformation due to\nthe mismatch of mechanical properties between soft PDMS relative to the generated silica-like surface. 37 This leads to a prominent challenge known as “hydrophobic\nrecovery” as PDMS reverts its surface chemistry\ndue to the diffusion of low molecular weight (LMW) species from the\nbulk through the cracks to the surface. 38 Additionally, cracks in the silica-like layer increase inhomogeneity\nof the surface, and optical properties may suffer from the presence\nof cracks as light transmission is altered between the two layers\ncausing light scattering. 37 These prominent\nissues can be addressed by shorter plasma activation times to minimize\nthe formation of silica-like layers but comes at the cost of either\nincomplete surface activation or faster hydrophobic recovery, as high-mobility PDMS chains can quickly reorient or redistribute to minimize\nsurface energy 39 , 40 ( Figure 2 ). Figure 2 Challenges in PDMS plasma activation. This reported work aims to explore the adaptation\nof LLS from hard\nsubstrates to elastomeric PDMS substrates. We target\na reduction in the formation of the silica layer and hydrophobic recovery\nthrough a preceding extraction step of the LMW species combined with\ncareful control over the PDMS plasma activation parameters.\nThese steps enable straightforward covalent grafting of liquid-like\nlinear polysiloxane brushes via the grafting-from method by dip-coating.\nDifferent moieties of the brushes are provided by the employed silanes.\nThis approach equips commercially available silicone elastomer substrates\nwith increased droplet mobility, anti-adhesion, and anti-icing properties.",
"discussion": "2 Results and Discussion 2.1 Substrate Preparation and\nFunctionalization PDMS samples were prepared\n( Figure 3 a) by using\na commercially available PDMS prepolymer mixture (Sylgard\n184) through a hydrosilylation\nreaction with a platinum catalyst. Besides the main framework components,\ncommercial mixtures also include additives, such as fillers, cross-linking\ninhibitors, and solvents, 5 which contribute\nto the presence of mobile LMW species and uncross-linked oligomers\nwithin the final elastomer. 9 , 41 LMW species were extracted\nby immersion in toluene from all silicone elastomer substrates for\n24 h and are subsequently denoted as “ ePDMS ”\n( Section 4.2.1 , Table S2 , Figure S2 ). Figure 3 (a) Scheme of the sample preparation process. Step 1 depicts the\npreparation of PDMS elastomer sheets and the subsequent\nextraction process in toluene. In step 2, samples are activated with\nair plasma and dip-coated to graft polymer brushes to the surface.\n(b) Silanes and abbreviations utilized in PDMS functionalization. To investigate the effect of plasma activation\nparameters on the\nformation of silica-like layers, the elastomer samples were activated\nin air (0.14 mbar) for an exposure time of 60 s at either 75, 150,\nor 225 W for 13.56 MHz (HF) plasma and 50, 100, or 150 W for 40 kHz\n(LF) plasma. After plasma treatment, the PDMS and ePDMS surfaces became hydrophilic, allowing for complete wetting\nby water. Removal of LMW fragments by extraction extended the longevity\nof PDMS activation by retarding hydrophobic recovery, 42 as water droplets on the surface still possessed\na low contact angle 20 h after plasma exposure ( Figure S3 ). After activation, the ePDMS was functionalized, adapting\na published procedure 16 by initially dip-coating\nan acidified silane in isopropanol solution and polymerizing of silanes\nupon concentration during drying. The silanes utilized were fluoroalkylsilane\n1 H ,1 H ,2 H ,2 H -perfluorooctylmethyldimethoxysilane ( FAS )\nand dimethoxydimethylsilane ( DMS ). Additionally, a coating\nmixture, in which the silane portion of the solution comprising 50\nwt % FAS and 50 wt % DMS to serve as a smaller\nlinker molecule in between bulkier FAS , potentially allowing\nlonger siloxane chain formation, was investigated. This functionalization\nis denoted as Mix ( Figure 3 b). Correspondingly, the functionalized ePDMS substrates are denoted with the employed silane as subscript: ePDMS FAS , ePDMS DMS , and ePDMS Mix . The dipped specimens were air-dried for 20 min, followed by rinsing\nto remove unreacted silane ( Section 4.2.2 ). The solvent readily successfully dewetted\nfunctionalized surfaces during washing, unlike for unfunctionalized ePDMS . As a reference, regular (unextracted) PDMS also underwent functionalization with FAS and is referred\nto as PDMS FAS . 2.2 Wetting Properties and Surface Tension of\nFunctionalized Substrates As previously demonstrated in slippery\nomniphobic covalently attached liquid coatings on glass substrates, 16 bifunctional silanes should yield linear siloxane\nchain growth 34 and form LLSs with interfacial\nslip and low friction. Contact angle hysteresis (Δθ =\nθ a – θ r ) as the difference\nbetween advancing (θ a ) and receding contact angle\n(θ r ) has recently been suggested as the prime indicator\nfor such behavior. 43 For water, θ a showed little variation between substrates, while significant\ndifferences in θ r and thus Δθ were found\n( Table 1 ). For the\ninvestigated substrates, θ r on ePDMS were the lowest, as droplet withdrawal was obstructed by surface\nfeatures, leading to dewetting at lower angles. θ r increased for ePDMS DMS as the\nchain mobility arising from the grafted PDMS chains aided droplet\ndepinning. θ r also increased for ePDMS FAS as low surface energy groups decreased\nthe work of adhesion between the test liquid and the surface. For ePDMS Mix , we observed further increased\nθ r , which we attributed to a synergistic effect of\nlow surface energy groups and increased chain mobility, as partial FAS replacement with less sterically demanding DMS increased interfacial slip of the resulting LLS ( Figure S4 ). Dynamic contact angles for diiodomethane can be\ninterpreted in a similar manner, although the variation in θ a was more pronounced due to fluorination of ePDMS FAS and ePDMS Mix as well as the higher surface roughness of ePDMS . For hexadecane, no dynamic contact angles could be obtained for ePDMS or ePDMS DMS due\nto the immediate swelling of the substrates by the probing liquid.\nSwelling of the substrates was prevented in ePDMS FAS and ePDMS Mix due to fluorination of the polymer brushes ( Figure S5 ). The lower degree of fluorination in the Mix functionalization resulted in a lower θ a , while the θ r remained similar. Surface tensions\n(γ sv ) were calculated from advancing contact angles\n(θ a ), as they are sensitive to the low surface energy\npart of a surface. 44 The Owens, Wendt,\nRabel, and Kaelble (OWRK) method 45 was\nemployed for water (γ lv = 72.8 mN/m of which γ lv P = 51.0 mN/m and γ lv D = 21.8 mN/m) and diiodomethane (γ lv = 50.8 mN/m\nof which γ lv D = 49.0 mN/m and γ lv P = 1.8 mN/m 46 ) as\ntest liquids ( Table 1 ). For ePDMS γ sv = 17.3 mN/m is obtained,\nwhich is lower compared to the commonly reported literature values\nfor PDMS of ∼21 mN/m. 5 , 47 PDMS extraction\nled to increased surface roughness already indicated by increased\nθ a and Δθ and consequently lower calculated\nγ sv . When comparing functionalized ePDMS substrates, calculated γ sv correlates with the\nincreasing degree of fluorination of ePDMS DMS , ePDMS Mix ,\nand ePDMS FAS , respectively. For ePDMS FAS , γ sv = 15.0\nmN/m is in good agreement with the γ sv of fluorosilicones\n(14–15 mN/m 48 ). Table 1 Advancing Contact Angles (θ a ), Receding Contact\nAngles (θ r ), Contact\nAngle Hysteresis (Δθ) for Water, Diiodomethane, and Hexadecane\n(Hex) and Surface Tension (γ sv ) Calculated According\nto OWRK Method a Sample θ a,H 2 O θ r,H 2 O Δθ H 2 O θ a,CH 2 I 2 θ r,CH 2 I 2 θ CH 2 I 2 θ a,Hex θ r,Hex Δθ Hex γ sv [mN/m] ePDMS DMS 108.0 ± 1.8° 80.0 ± 1.0° 28.0° 66.9 ± 2.0° 50.3 ± 1.1° 16.6° — — — 25.1 ePDMS FAS 114.6 ± 1.4° 87.0 ± 1.0° 27.9° 85.1 ± 1.6° 47.8 ± 1.5° 37.3° 77.8 ± 1.3° 34.8 ± 1.6° 43.7° 15.0 ePDMS Mix 112.2 ± 1.0° 93.6 ± 2.0° 18.5° 73.4 ± 1.3° 62.6 ± 1.3° 10.8° 57.1 ± 1.4° 33.6 ± 1.8° 23.5° 21.5 ePDMS 113.2 ± 0.6° 75.7 ± 1.6° 37.5° 93.4 ± 0.5° 53.2 ± 0.4° 40.2° — — — 17.3 a Dynamic contact\nangles of hexadecane\ncould not be measured on ePDMS and ePDMS DMS due to substrate swelling, indicated\nby “—” Δθ further serves as a measure for surface heterogeneity;\nΔθ typically increases for heterogeneous surfaces due\nto pinning and depinning of the contact line on contrasting surface\nchemistries. 49 Dynamic contact angle measurements\nwere carried out to compare the surface homogeneities of PDMS FAS and ePDMS FAS . For PDMS FAS , no θ r was measurable due to droplet contact line pinning during\nprobing liquid aspiration, as indicated in Figure 4 a. Impaired dewetting on functionalized PDMS is caused by uneven functionalization 44 as unextracted and more mobile LMW chains contribute to\nhydrophobic recovery but also react with and deplete coating reagents.\nFor ePDMS FAS , a stable θ r value was observed for more than 30 s during dewetting ( Figure 4 b), indicating that\nthe extraction step produced a homogeneously functionalized surface. Figure 4 Results\nof contact angle measurements. (a) Contact angle (θ)\nand droplet diameter during dynamic water contact angle measurements\non PDMS FAS , θ continuously\ndecreases in the circled region; (b) upon extraction, a stable (plateaued)\nθ r is apparent on ePDMS FAS . For the different substrates,\nsliding angles (α) follow the\nsame trend as Δθ, apart from that of hexadecane ( Table 2 ). α for hexadecane\non ePDMS DMS was lowered compared\nto fluorinated brush counterparts, as both the PDMS substrate\nand polymerized brushes of DMS were readily swelled by\nhexadecane, leading to enhanced lubrication and dynamic dewetting\nbehavior. 24 , 26 In contrast, the swelling of grafted fluorinated\nbrushes was inhibited and resulted in a higher sliding angle for ePDMS FAS and ePDMS Mix . As swelling of the brushes was impeded, this\ndesignates an opposing trend in solvent effects usually observed in\nLLS, where organic solvents swell the grafted brushes to create a\n“blended liquid–liquid interface”. 24 On hard and polar substrates where polymer brushes\nare present as a thin layer followed by a dissimilar chemical composition,\nthis leads to greatly improved droplet mobility. However, this constitutes\nan undesirable process for PDMS substrates, as the probing\nliquid would persistently cause swelling of the elastomer and ingress\ninto the substrate. Therefore, blending low-surface-energy groups\ninto the polymer brushes allowed for good droplet mobility without\ncompromising the underlying silicone substrate. Table 2 Sliding Angles (α) for Diiodomethane,\nHexadecane (Hex), and PEG-200 Sample α CH 2 I 2 α Hex α PEG-200 ePDMS DMS 9.8 ± 1.2° 12.8 ± 2.9° >30° ePDMS FAS 11.8 ± 0.7° 21.8 ± 1.1° 23 ± 1.2° ePDMS Mix 5.8 ± 1.1° 17.8 ± 1.7° 19 ± 0.7° ePDMS 14.8 ± 1.0° 15.3 ± 0.8° >30° Contact angle goniometry showed increased static contact\nangles\n(θ) of the testing liquids after functionalization of ePDMS with fluorinated silanes due to unfavorable interactions\nwith low surface tension fluorinated groups. 50 For comparison, silane coatings were also applied to glass substrates.\nEncouragingly, the observed trend of increasing θ, lowered Δθ,\nand calculated γ sv was similar for both silanized ePDMS and silanized glass substrates, indicating the applicability\nof this approach to both hard and soft substrates ( Tables S3–S5 ). To address the stability of the\nfunctionalization over time, we\nundertook a daily rinsing cycle of ePDMS Mix with water and isopropanol (representing an aqueous and an\norganic solvent) and monitored the dynamic contact angles for water\nand diiodomethane over 7 days ( Figure 5 ). The liquid-like properties of the functionalization\nremained stable, as no significant change in Δθ was found. Figure 5 Dynamic\ncontact angles of water and diiodomethane on ePDMS Mix over 7 daily cycles of rinsing with water\nand isopropanol. 2.3 Effect\nof Activation and Functionalization\non Optical Properties of Functionalized Substrates Coatings\nand materials transparency are important for any applications requiring\nlight transmission, such as for optoelectronic displays, 51 , 52 solar harvesters, or transparent release films in manufacturing. PDMS is transparent in the UV–vis region above 280\nnm, 3 , 53 with partial transmittance in the region\nbetween 240 and 280 nm. Plasma activation led to a slight decrease\nin transmittance in ePDMS . For all functionalizations\n( DMS , FAS , and Mix ) on ePDMS , transmittance reduction by approximately 6% for LF-plasma-activated ePDMS ( ePDMS LF ) and 3%\nfor HF-plasma-activated ePDMS ( ePDMS HF ) occurred mainly in the range from 280 to 400\nnm ( Figure 6 b). However,\nthe impact of LF versus HF plasma can be more clearly observed in\nthe 240–280 nm region, as all ePDMS LF samples showed a transmittance lower than that of ePDMS HF samples. Above 400 nm, transparency\ndifferences for both activation frequencies are negligible, and samples\nappeared to visually retain their transparency after surface activation\nand functionalization ( Figure 6 a). Figure 6 (a) Bending of samples reveals thin-film iridescence on ePDMS LF Mix . Samples remain\ntransparent after activation and coating. (b) Light transmittance\nof ePDMS substrates. (c) Optical microscopy of ePDMS LF Mix reveals superficial cracking at all power settings, highlighted\nby ink infiltration. (d) Optical microscopy of ePDMS HF Mix shows decreased\nand finer cracking for 25 and 75% power compared to that in (c). No\ncracking was found on samples activated at 50% power. The ePDMS LF additionally\nexhibited\nsuperficial iridescence ( Figure 6 a) when viewed during mechanical deformation. This\niridescence was indicative of a glassy, silica-like layer introduced\nby plasma activation, with a thickness in the range of the wavelength\nof visible light, which is in good accordance with previously reported\nthicknesses of the silica-like surface layer after plasma treatment. 39 , 54 Interestingly, for ePDMS HF ,\nthe iridescence was distinctly decreased. The mismatch in elastic\nmoduli between the glassy layer and the\nelastomer bulk can result in spontaneous cracking of the former. 37 , 54 This phenomenon was indeed observed in some of our plasma-activated\nsamples. It should be noted that, unlike for previous studies, 37 , 38 , 41 the activated specimens were\npurposely subjected to severe mechanical deformation, e.g., PDMS folding on itself by 180°, to deliberately cause\nsurface cracking of the silica-like layer. Optical microscopy showed\nprominent cracking for all ePDMS LF ( Figure 6 c). In contrast,\nfor ePDMS HF , significantly decreased\ncracking and reduced crack prevalence was observed at all tested power\nsettings, with no cracking observable for ePDMS HF Mix activated at 50% power\n( Figure 6 d). To aid\nin crack visualization, the sample surfaces were marked with a waterproof\nink. Subsequent rinsing with acetone revealed residual ink visible\nin the surface cracks on samples activated with LF plasma, while no\nink retention could be observed on samples activated with HF plasma\n( Figure S6 ). 2.4 Effect\nof Extraction and Functionalization\non Mechanical Properties As the extraction, activation, and\nfunctionalization processes might impact the mechanical properties\nof PDMS , we conducted tensile tests according to an adapted\nnorm for mechanical testing of elastomers (ASTM D412). Dogbone-shaped\ntest specimens were stamped out from PDMS sheets and strained until\nfailure at 500 mm/min. We found similar stress–strain profiles\nfor all test series ( Figure 7 ) and obtained the ultimate tensile strength for PDMS , ePDMS , and ePDMS Mix at 7.2 ± 1.0 MPa, 7.0 ± 1.1 MPa, and 7.9 ±\n0.3 MPa, respectively. Given the standard deviation, this indicates\nno loss of tensile strength throughout the PDMS modification\nand functionalization process. Additionally, the elastic moduli were\nobtained, yielding 1.60 ± 0.28 MPa for PDMS , 1.70\n± 0.28 MPa for ePDMS , and 1.85 ± 0.12 MPa for ePDMS Mix and showing a marginal increase\nin sample stiffness. Figure 7 Overlaid stress–strain curves for PDMS (blue), ePDMS (pink), and ePDMS Mix (green). 2.5 Surface Composition of Functionalized Substrates We analyzed the surface elemental composition of substrates via\nX-ray photoelectron spectroscopy (XPS) and conducted Ar-ion etching\nwith serial etching steps to gain depth information. Since silanes\nfor functionalization and bulk PDMS all contain Si, C,\nand O, we selected F as a marker for successful formation of ePDMS Mix and ePDMS FAS substrates. XPS survey spectra indeed confirmed\nthe presence of fluorine in ePDMS HF FAS , ePDMS LF FAS , ePDMS HF Mix , and ePDMS LF Mix ( Figures 8 a and S7 ). Figure 8 (a) Elemental\nsurvey of ePDMS HF Mix . (b) Etching profile for ePDMS activated with\nLF plasma shows higher oxygen content and less carbon\noverall. (c) Etching profile for ePDMS activated with\nHF plasma shows a lower disparity between oxygen and carbon at %.\n(d) Thermal decomposition curves for unaltered, extracted, and extracted\nand functionalized PDMS under a nitrogen atmosphere. When only the effect of plasma activation was investigated,\nelevated\noxygen content was observed for ePDMS LF over the entire etching experiment (255 s total etch time),\nas evident from the increase in atomic percent (at %) of oxygen at\nthe expense of carbon ( Figure 8 b), which is indicative of the formation of a silica-like\nlayer. 42 , 55 In contrast, ePDMS HF exhibited lower oxygen incorporation and a residual\ncarbon content of about 20 at % or higher ( Figure 8 c). The silicon content was not affected\nby the activation plasma frequency. Additionally, deconvolution of\nelemental scans for silicon showed organic silicon in the range of\n101.7 eV and SiO 2 -species at approximately 103.0 eV. 56 ePDMS HF exhibited an elevated organic silicon content in comparison with ePDMS LF ( Figure S11 ). Carbon elemental scans feature multiple carbon\nspecies with binding\nenergies ranging from 284.8–294 eV, which were assigned to\nC–C bonds at 284.8 eV, adventitious carbon species C–O–C\nat 286 eV, O–C=O at 288.5 eV, CF 2 -groups\nat 292 eV, and CF 3 -groups at 294 eV. 57 Since FAS carries two CH 2 - and\nfive CF 2 -groups and one terminal CF 3 -functionality\nin the side chain, an area ratio of 1:5 for the signals at 294 eV\nfor CF 3 -groups and 292 eV for CF 2 -groups would\nbe expected for functionalization containing FAS . Area\nratios of 1:4.9 for ePDMS LF Mix , 1:6.3 for ePDMS LF FAS , 1:10.4 ePDMS HF Mix , and 1:10.9 ePDMS HF FAS were calculated from XPS peak deconvolution ( Figures S9 and S10 ). As the C 1s\nsignal at 284.8 eV was more prevalent in HF than in LF-plasma activated ePDMS , the peak ratio for CF 2 - and CF 3 -groups was distorted by this stronger carbon peak, which explained\nthe ratio deviation from expectation. 2.6 Thermal\nDecomposition of Liquid-like Brushes Thermogravimetric analysis\nwas carried out to confirm the extraction\nprocess and successful functionalization of ePDMS ( Figure 8 d). The thermogram\nof PDMS shows an initial mass loss at 225 °C due\nto the decomposition of LMW species. In the case of ePDMS , the initial mass loss occurred later at 400 °C, which confirmed\nthe successful removal of LMW components. As LLS were prepared from ePDMS , thermal analysis showed that no volatile species were\nlost at lower temperatures but a lower residual mass was found for\nall functionalized samples, regardless of the activation conditions.\nThis is consistent with the presence of non-cross-linked, linear chains\ntethered to the surface, since minimally or non-cross-linked PDMS fully decomposes during thermal analysis as volatile\ncyclic oligomers are formed and no residual mass is observed. 58 The lack of cross-linking in superficial linear\npolymer brushes allows molecular motion and siloxane chain cyclization,\nas well as the subsequent elimination of volatile products. Furthermore,\nthe polymer brushes grafted to ePDMS are on the external\nsurface, thereby having an increased tendency of volatilization instead\nof entrapment and conversion to residue within the bulk PDMS . The lower residual mass for the functionalized samples supports\nthe observation of superficial functionalized layers, in line with\nXPS/ion-etching experiments. 2.7 Solid Adhesion in Horizontal\nPush Tests Besides contact angle measurements, which probe\nliquid–solid\ninteractions, we evaluated the adhesion between the functionalized PDMS substrates and solid contaminations. Solid adhesion was\ndetermined via a modified tensile testing system to allow for a shear-based\n(Mode II) horizontal push test configuration ( Figure 9 a). We selected materials that started off\nas fluids and subsequently solidified to maximize molecular contact\nat the testing interface. Gypsum was chosen as a representative sample\nof an inorganic material, and beeswax was chosen as a representation\nof an organic, biologically derived material. Detailed information\non the test configuration can be found in the Supporting Information . Generally, the force curves on ePDMS describe an elastic region followed by plastic deformation\nand finally adhesive failure, resulting in interfacial debonding for\ngypsum plaster and solidified beeswax. This relationship is usually\nobserved when examining lateral friction and adhesion of solids 59 as well as liquids 60 on solid surfaces. On ePDMS Mix , adhesion force curves of gypsum plaster showed comparable characteristics\nwith plain ePDMS , albeit with reduced forces ( Figure S14 ). However, the force profiles for\nbeeswax on ePDMS Mix did not show\ndistinct debonding peaks but rather suggest that the initial force\nrequired to debond beeswax roughly corresponds to the kinetic friction\nforce across the surface ( Figure 9 c). 60 , 61 The ePDMS Mix substrate exhibited reduced adhesive strength irrespective\nof plasma activation frequency in the case of beeswax, whereas for\ngypsum plaster, the adhesive strength was lowered for ePDMS LF Mix versus ePDMS HF Mix , which we attribute to a superficial smoothing effect due to the\nmore prominent glassy layer ( Figure 9 b). Interestingly, in the case of beeswax, improved\nflow of melted beeswax across the ePDMS Mix surfaces could be qualitatively observed. Beeswax that contacted\nunfunctionalized ePDMS showed clear circles from pouring\nit in liquid form during sample preparation ( Figure 9 d). We hypothesize that these concentric\nrings originate from the impeded flow of liquid wax when it contacts\nunfunctionalized ePDMS , and upon further pouring, new,\nlarger rings formed. The patterning of the beeswax is still visible\nfor ePDMS LF Mix , albeit to a lesser extent, while only minimal ring patterning\nis noticeable for ePDMS HF Mix . This is indicative of a qualitatively enhanced\nflow behavior of viscous liquids on the prepared surfaces. Additionally,\nthis could point toward an altered heat conduction mechanism on functionalized\nsamples due to superficial polymer brushes. Figure 9 (a) Schematic illustration\nof the adhesion test configuration.\n(b) Solid adhesion on ePDMS Mix and bare ePDMS . Data was subjected to one-way ANOVA, p = 0.05, * = statistically significant difference, n.s.\n= not significant. (c) Force profiles for beeswax on ePDMS and ePDMS HF Mix . (d) Beeswax interfaces after separation in adhesion testing\non ePDMS (1), ePDMS LF Mix (2), and ePDMS HF Mix (3). 2.8 Evaluation of Ice Adhesion One of\nthe most important applications of anti-adhesion coatings is the generation\nof ice-phobic surfaces, as ice accretion poses a ubiquitous problem\nin particular for aerial and marine vehicles, as well as for power,\ncommunications, and general infrastructure in terms of safety hazards\nand lowered performance efficiency. PDMS has been extensively\nstudied for ice-phobic coatings, not just because of its well-known\nhydrophobicity and low surface energies but also due to the modulus\nmismatch between stiff ice and elastomeric PDMS , which\nhas been shown to induce cavitation at the interface and aid ice-delamination. 62 We therefore extended our studies to test the\nanti-icing properties of preoptimized ePDMS HF Mix using horizontal push\ntests. Additionally, polymer brush surfaces were lubricated to investigate\nthe impact on icing. Tests were conducted at −20 and −10\n°C. At both temperatures, ice adhesion was reduced by a third\nfor ePDMS Mix compared to ePDMS . Lubrication of ePDMS Mix by soaking the functionalized elastomer in a perfluoropolyether\nlubricant (Krytox GLP105) and then wiping off the excess lubricant\n(referred to as L-ePDMS Mix ) led\nto halving of the ice adhesion strength compared to that of plain ePDMS ( Figure 10 and Table 3 ). It\nshould be noted that simply soaking ePDMS in Krytox GLP105\ndid not lead to lubrication, as lubricant dewetting occurred for the ePDMS surface. In contrast, the lubricant fully wetted ePDMS Mix ( Figure S1 ), suggesting that surface/subsurface functionalization is\nnecessary to enable infiltration of the low surface tension lubricant.\nFurther, as stick–slip-like motion of ice was observed on ePDMS as well as lubricated ePDMS ( Figure S15a,c ), we conclude that surfaces were\nnot sufficiently lubricated. In contrast, smooth sliding was facilitated\non ePDMS Mix and L-ePDMS Mix ( Figure S15e,g ). Both ePDMS Mix and L-ePDMS Mix showed ice-phobicity at −20 and\n−10 °C, as they exhibited ice adhesion strength <100\nkPa. 59 Ice adhesion on L-ePDMS Mix at −10 °C decreased by approximately\nthree times compared to that at −20 °C, with L-ePDMS Mix showing ultralow ice adhesion (<20\nkPa), which allows for passive removal of ice on moving parts or by\ngusts of wind, 63 indicating the potential\nutility of these functionalized coatings in combination with electrothermal\ndeicing systems. The decreased adhesion at higher temperatures is\nin line with the known temperature-dependent adhesion behavior of PDMS surfaces. 64 For our functionalized/functionalized\nand lubricated ePDMS specimens, we assume that the decreased\nice-adhesion occurred due to the decreased stiffness of the PDMS substrate as well as increased mobility of the liquid-like\nsurface and the lubricant at −10 °C versus −20\n°C. This was additionally affirmed by the respective force profiles,\nas sliding and stick–slip motion was observed at −10\n°C, while a distinct peak for interfacial debonding and minimal\nfriction was observed at −20 °C ( Figure S15a–d ). Figure 10 Ice adhesion on bare ePDMS and ePDMS Mix and lubricated substrates\nat −10 and\n−20 °C. Data was subjected to one-way ANOVA, p = 0.05, * = statistically significant difference, n.s. = not significant. Table 3 Ice Adhesion Strength on the Investigated\nSubstrates at −10 and −20 °C Temperature Ice adhesion strength [kPa] –20 °C ePDMS 120.2 ± 41.9 L-ePDMS 89.9 ± 20.0 ePDMS Mix 84.0 ± 23.2 L-ePDMS Mix 59.9 ± 21.3 –10 °C ePDMS 39.4 ± 10.8 L-ePDMS 33.2 ± 9.5 ePDMS Mix 28.8 ± 3.5 L-ePDMS Mix 17.5 ± 7.7"
} | 8,993 |
35072015 | PMC8762475 | pmc | 8,219 | {
"abstract": "Summary Biological visual system can efficiently handle optical information within the retina and visual cortex of the brain, which suggests an alternative approach for the upgrading of the current low-intelligence, large energy consumption, and complex circuitry of the artificial vision system for high-performance edge computing applications. In recent years, retinomorphic machine vision based on the integration of optoelectronic image sensors and processors has been regarded as a promising candidate to improve this phenomenon. This novel intelligent machine vision technology can perform information preprocessing near or even within the sensor in the front end, thereby reducing the transmission of redundant raw data and improving the efficiency of the back-end processor for high-level computing tasks. In this contribution, we try to present a comprehensive review on the recent progress achieved in this emergent field.",
"introduction": "Introduction In the era of big data and the internet of things, the unprecedented huge amount of information and complex external environment put forward more stringent requirements for developing new-generation multi-functional artificial intelligence chips ( Ham et al., 2021 ). Given that visual perception is one of the most important ways to obtain environmental information, the demand for visual information sensing, storage, and processing function devices with higher speed, greater efficiency, and lower power consumption is becoming ever more urgent. Although traditional machine vision technology has profoundly changed the lives of human beings in many fields, it has gradually become clumsy and inadequate, limited by the von Neumann bottleneck when dealing with complex tasks ( Chai, 2020 ). Therefore, the development of more intelligent machine vision technology to satisfy the new requirements of the times has become one of the most important innovation directions in the field of artificial intelligence chips in the post-Moore era ( Waldrop, 2016 ). Human visual system is capable of visual information perception and multiple target recognition in complex environments, which inspires the development of biomimetic visual systems with new optoelectronic devices for high-performance machine vision technology ( Abramoff et al., 2010 ). The main functions of the human visual system can be divided into two parts: image perception and preprocessing in the human eye and recognizing, memorizing in the visual center of the cerebral cortex. In recent years, several novel retinomorphic machine vision architectures have been developed and demonstrate strong vitality by simulating the working mechanism of the human visual system. According to different forms of functional divisions, heterogeneous and homogeneous integration architectures are the two main paradigms. Both architectures could perceive and preprocess the image information at the front end, thereby reducing redundant information and improving the overall recognition efficiency. Compared with traditional CMOS (complementary metal-oxide-semiconductor)-based machine vision systems, the novel retinomorphic optoelectronic devices exhibit obvious performance advantages. It has beendemonstrated that most energy consumption of traditional machine vision is spent on the redundant information transfer among the sensor, memory, and processor. Because the raw information can be preprocessed at the front end, the novel retinomorphic optoelectronic devices have inherent advantages in reducing energy consumption. The energy cost for writing information into memristors could also be reduced by more than 100 times. Yao and Wu et al. demonstrated that the energy consumption of electronic synapses is 1000 times smaller than the Intel Xeon Phi processor when dealing with similar face recognition tasks ( Yao et al., 2017 ). The switching time (<10 ns), endurance (10 5 ∼10 8 ), and chip scaling potential (<10 ns) are also superior to those of the traditional counterparts ( Ielmini and Wong, 2018 ; Milo et al., 2020 ). Therefore, retinomorphic optoelectronic devices may provide a new and effective approach for improving information processing efficiency and energy consumption problems in the era of big data. Here, we present an overview of the recent advances in retinomorphic machine vision technology from principle to device. Firstly, the working mechanism of the human visual system and several differences with artificial retinomorphic devices are discussed. Then, two paradigms, viz. heterogeneous and homogeneous integration architectures, will be summarized and discussed in detail. Finally, a brief discussion on the current challenges and prospects of retinomorphic machine vision is provided."
} | 1,182 |
24855656 | null | s2 | 8,222 | {
"abstract": "The ability to resist mechanical forces is necessary for the survival and division of bacteria and has traditionally been probed using specialized, low-throughput techniques such as atomic force microscopy and optical tweezers. Here we demonstrate a microfluidic technique to profile the stiffness of individual bacteria and populations of bacteria. The approach is similar to micropipette aspiration used to characterize the biomechanical performance of eukaryotic cells. However, the small size and greater stiffness of bacteria relative to eukaryotic cells prevents the use of micropipettes. Here we present devices with sub-micron features capable of applying loads to bacteria in a controlled fashion. Inside the device, individual bacteria are flowed and trapped in tapered channels. Less stiff bacteria undergo greater deformation and therefore travel further into the tapered channel. Hence, the distance traversed by bacteria into a tapered channel is inversely related to cell stiffness. We demonstrate the ability of the device to characterize hundreds of bacteria at a time, measuring stiffness at 12 different applied loads at a time. The device is shown to differentiate between two bacterial species, E. coli (less stiff) and B. subtilis (more stiff), and detect differences between E. coli submitted to antibiotic treatment from untreated cells of the same species/strain. The microfluidic device is advantageous in that it requires only minimal sample preparation, no permanent cell immobilization, no staining/labeling and maintains cell viability. Our device adds detection of biomechanical phenotypes of bacteria to the list of other bacterial phenotypes currently detectable using microchip-based methods and suggests the feasibility of separating/selecting bacteria based on differences in cell stiffness."
} | 456 |
30405656 | PMC6201211 | pmc | 8,223 | {
"abstract": "Actinorhizal plants are able to establish a symbiotic relationship with Frankia bacteria leading to the formation of root nodules. The symbiotic interaction starts with the exchange of symbiotic signals in the soil between the plant and the bacteria. This molecular dialog involves signaling molecules that are responsible for the specific recognition of the plant host and its endosymbiont. Here we studied two factors potentially involved in signaling between Frankia casuarinae and its actinorhizal host Casuarina glauca: (1) the Root Hair Deforming Factor (CgRHDF) detected using a test based on the characteristic deformation of C. glauca root hairs inoculated with F. casuarinae and (2) a NIN activating factor (CgNINA) which is able to activate the expression of CgNIN , a symbiotic gene expressed during preinfection stages of root hair development. We showed that CgRHDF and CgNINA corresponded to small thermoresistant molecules. Both factors were also hydrophilic and resistant to a chitinase digestion indicating structural differences from rhizobial Nod factors (NFs) or mycorrhizal Myc-LCOs. We also investigated the presence of CgNINA and CgRHDF in 16 Frankia strains representative of Frankia diversity. High levels of root hair deformation (RHD) and activation of Pro CgNIN were detected for Casuarina -infective strains from clade Ic and closely related strains from clade Ia unable to nodulate C. glauca . Lower levels were present for distantly related strains belonging to clade III. No CgRHDF or CgNINA could be detected for Frankia coriariae (Clade II) or for uninfective strains from clade IV.",
"introduction": "Introduction Legumes and actinorhizal plants form a N 2 -fixing root nodule symbiosis in association with rhizobia and Frankia bacteria, respectively ( Vessey et al., 2005 ). The establishment of these beneficial bacterial-plant relationships requires communication between the partners. Rhizobial symbioses have received considerable attention because several legumes are important crop species. However, actinorhizal symbioses, which play an important ecological role ( Dawson, 2008 ), have been less well studied and the molecular dialog between Frankia and their host plants is still poorly understood. One reason is that most actinorhizal plants are woody shrubs or trees for which genetic approaches are very difficult ( Wall, 2000 ; Perrine-Walker et al., 2011 ). In addition, the genetics of the bacterial partner, Frankia , is not fully developed and up to now Frankia cells remain recalcitrant to stable genetic transformation ( Kucho et al., 2009 , 2017 ). Recent progress including the sequencing of several Frankia genomes ( Normand et al., 2007 ; Tisa et al., 2016 ), transcriptomic studies ( Alloisio et al., 2010 ; Benson et al., 2011 ), proteomic studies ( Mastronunzio and Benson, 2010 ; Ktari et al., 2017 ) together with functional studies on several actinorhizal species ( Svistoonoff et al., 2014 ) have opened new avenues for identifying components involved in the initial symbiotic dialog between the two partners. The interaction of rhizobia with model legumes begins with the production and recognition of signal molecules by their respective eukaryotic and prokaryotic symbiotic partners ( Oldroyd, 2013 ). Early events leading to nodule formation involve bacterial penetration into their hosts via root hairs. Bacteria elicit the stimulation and reorientation of root hair cell wall growth. This rhizobia-induced tip growth results first in the entrapment of the bacteria within curled root hairs and then in the initiation and development of infection threads (ITs), tubular structures through which bacteria pass on their way down the root hair and into the underlying cortical cell layers ( Lhuissier et al., 2001 ). Ahead of the advancing threads, cells in the inner cortex are induced to dedifferentiate and divide, and a nodule primordium is formed. In the first part of the signal exchange, the plant roots secrete flavonoids that lead to the activation of a set of rhizobial genes (the nod genes), which are essential for infection, nodule development and the control of host specificity ( Masson-Boivin et al., 2009 ; Oldroyd, 2013 ). These genes are responsible for the synthesis of lipo-chito-oligosaccharides (LCOs) called Nod factors (NFs) that signal back to the plant ( Oldroyd, 2013 ). NF biosynthesis is dependent on nodABC genes which are present in all rhizobia able to synthetize NFs and strain-specific combinations of other nodulation genes responsible for the addition of various decorations to the core structure. ( Masson-Boivin et al., 2009 ). In model legumes, NFs perception elicits a range of responses including ion fluxes, calcium oscillations, changes in gene expression patterns, and extensive deformation of roots hairs, which has been used as a bioassay to identify the chemical nature of NFs ( Lerouge et al., 1990 ; Oldroyd, 2013 ). Much less is known about signaling molecules involved in the actinorhizal symbioses. Canonical nodABC genes are not found in the sequenced genomes of 36 Frankia strains including Frankia alni and Frankia casuarinae ( Tisa et al., 2016 ) confirming a previous report showing that F. alni DNA will not complement rhizobial nod mutants ( Cérémonie et al., 1998 ). Only distant homologs of nodB and nodC are found in F. alni genome. Unlike rhizobial nod genes, they are not organized into a cluster together with other symbiotic genes and their expression is not induced under symbiotic conditions ( Normand et al., 2007 ; Alloisio et al., 2010 ). These findings are consistent with experiments showing that chitin oligomers similar to rhizobial NFs are not be detected in F. alni culture supernatant ( Cérémonie et al., 1999 ) suggesting structural differences between the Frankia symbiotic signals and rhizobial NFs. Recently, canonical nodABC genes have been found in the genome of two uncultured Frankia strains: Candidatus Frankia datiscae Dg1 and Candidatus Frankia californicae Dg2 ( Persson et al., 2015 ; Nguyen et al., 2016 ), and in one isolated strain, Frankia sp. NRRL B-16219 ( Ktari et al., 2017 ). F. datiscae Dg1 nodABC genes are arranged in two operons which are expressed in Datisca glomerata nodules, but their involvement in symbiotic signaling is still not known ( Persson et al., 2015 ). Frankia is able to infect their host root either through intracellular (root hair) or intercellular modes. In the first case, one of the earliest visible plant response to Frankia is an extensive deformation of root hairs. This response occurs in actinorhizal plants belonging to the order Fagales ( Betulaceae, Casuarinaceae ) that display a range of relatively advanced features reminiscent of model legumes: a complex root hair infection process involving the formation of ITs and the implication of cortical cell divisions at the initial stages of infection ( Svistoonoff et al., 2014 ). Frankia culture supernatants also cause root hair deformation (RHD) and a Frankia root hair deforming factor in Alnus (AgRHDF) was identified ( Prin and Rougier, 1987 ; Ghelue et al., 1997 ; Cérémonie et al., 1999 ; Gabbarini and Wall, 2011 ). Using RHD as a bioassay, partial purification was achieved. AgRHDF is a relatively small (< 3 kDa), heat stable, hydrophilic molecule that is resistant to a chitinase treatment, but its chemical structure remains unknown ( Cérémonie et al., 1999 ). In recent years, we have developed complementary bioassays using plant genes that are specifically expressed in response to interaction with a compatible Frankia . This approach is particularly well suited for C. glauca where transgenic plants containing promoters of symbiotic genes fused to either GUS or GFP can be generated ( Svistoonoff et al., 2010a ). Expressed Sequence Tag (EST) libraries of C. glauca and Alnus glutinosa ( Hocher et al., 2006 , 2011 ) provide extensive lists of genes potentially involved in the actinorhizal symbiosis. Among the candidate genes, we identified CgNIN , the putative ortholog of legume NIN genes, which encodes a transcription factor playing a central role in rhizobial nodulation ( Schauser et al., 1999 ; Marsh et al., 2007 ; Soyano et al., 2013 , 2014 ; Yoro et al., 2014 ). In C. glauca, CgNIN also has an important role in nodulation particularly at early steps of infection ( Clavijo et al., 2015 ). After contact with either Frankia cells or cell-free Frankia supernatants, the CgNIN promoter is strongly activated at 12 to 48 h ( Clavijo et al., 2015 ). This property was used to establish a new bioassay leading to the partial purification and characterization of a NIN activating factor, called CgNINA. While rhizobial NFs are amphiphilic chitin-based molecules, CgNINA, like AgRHDF, is hydrophilic and resistant to chitinase ( Chabaud et al., 2016 ). However, it is not known to what extent CgNINA is related to factors able to deform root hairs. Further experiments concerning these Frankia symbiotic factors are reported here. We show that C. glauca was able to perceive a root hair deforming factor secreted by F. casuarinae (CgRHDF) and the properties of CgRHDF were compared to those previously identified with CgNINA. The presence of CgNINA and CgRHDF in strains representative of Frankia diversity was investigated.",
"discussion": "Discussion Presence of CgRHDF and Comparison to CgNINA In legumes, RHD is one of the earliest visible responses induced upon recognition of rhizobial NFs by the host plant, and the development of a bioassay based on RHD was crucial to identify the chemical nature of NFs ( Lerouge et al., 1990 ). The perception of NFs also provokes significant alterations of gene expression and notably the expression of symbiosis-induced genes such as MtEnod11 ( Journet et al., 2001 ) and NIN ( Schauser et al., 1999 ; Radutoiu et al., 2003 ). In actinorhizal plants infected intracellularly such as C. glauca and A. glutinosa , RHD is also one of the first visible responses to Frankia inoculation and factors able to induce RHD in Alnus (AgRHDF) have been partially purified and characterized ( Cérémonie et al., 1999 ). In C. glauca , we have used transgenic plants expressing a Pro CgNIN:GFP fusion to characterize CgNINA, a factor present in cell free F. casuarinae supernatant fluids able to activate the CgNIN promoter in C. glauca root hairs. Here we have shown that a CgRHDF is also present in cell-free F. casuarinae supernatant fluids. Furthermore, we have shown that CgRHDF and CgNINA share similar physico-chemical properties listed in Table 1 . Interestingly the experiment performed with centrifugal filters suggests that CgNINA and CgRHDF are both small molecules with slightly different sizes, CgNINA being probably smaller than CgRHDF. This observation is intriguing if CgNINA or CgRHDF are the actinorhizal analogs of rhizobial NFs because NFs are known to induce both RHD and the expression of early nodulins such as NIN. Responses obtained with centrifugal filters were however, less contrasted compared to the other experiments shown here and residual CgRHD activity was still present in the 3kDa flow through. We therefore cannot exclude that CgNINA also possess a small RHD activity and additional experiments are needed to confirm this hypothesis. If CgRHDF and CgNINA can indeed be separated, it would be interesting to know if those molecules are able to induce the high frequency nuclear Ca 2+ spiking in growing C. glauca root hairs as described previously ( Chabaud et al., 2016 ). The ability to induce Ca 2+ oscillations in response to symbiotic bacteria is a common feature of nodulating species within the nitrogen-fixing clade ( Granqvist et al., 2015 ). Alternatively we hypothesize that CgRHDF and CgNINA are the same molecule but a cofactor or a specific decoration is needed to enhance CgRHD activity without affecting its ability to activate Pro CgNIN . The 3 kDa centrifugal filter possibly eliminated decorated molecules with higher mass or cofactors and therefore strong activity was only detected with the CgNINA bioassay. Table 1 Properties of CgRHDF, CgNINA, AgRHDF, and rhizobial NFs. CgRHDF CgNINA AgRHDF Rhizobial NF Induction Inducible Inducible Inducible Inducible (root exudates) (root exudates) (root exudates) (flavonoids) Size 1–3 kDa 1–3 kDa 1.2–3 kDa 1 kDa Thermal stability Thermoresistant Thermoresistant Thermoresistant Thermoresistant Active concentration 10 -2 –10 -3 (supernatant) 10 -1 –10 -4 (supernatant) 10 -3 10 -5 (supernatant) Down to 10 -12 M Hydrophilicity Hydrophilic Hydrophilic Hydrophilic Amphiphilic Chitinase action resistant resistant resistant sensitive Comparison With AgRHDF and Rhizobial Nod Factors CgRHDF and CgNINA also share many characteristics with AgRHDF, the corresponding factor characterized using A. glutinosa / F. alni ( Table 1 ; Ghelue et al., 1997 ; Cérémonie et al., 1999 ). However, both factors appear to be structurally different from the rhizobial NFs because unlike NFs they are not found in the organic phase following a butanol extraction and are not sensitive to the endochitinase from Aeromonas hydrophila ( Cérémonie et al., 1999 ) or the exochitinase from S. griseus ( Cérémonie et al., 1999 ; Chabaud et al., 2016 ). This difference is in agreement with (1) the lack of nodA genes in the sequenced genomes of F. casuarinae and F. alni ( Normand et al., 2007 ). (2) the absence of chitin oligomers in F. alni supernatant fluids ( Cérémonie et al., 1999 ), and (3) the failure of NFs from the broad host-range rhizobia NGR234 to elicit RHD or Ca 2+ spiking in A. glutinosa or C. glauca ( Cérémonie et al., 1999 ; Granqvist et al., 2015 ; Chabaud et al., 2016 ). The possibility that actinorhizal recognition is mediated by molecules that are not hydrolyzed by tested chitinases is unexpected because downstream components of the NF signaling pathway are conserved not only between actinorhizal and rhizobial symbioses ( Svistoonoff et al., 2014 ; Griesmann et al., 2018 ), but also between rhizobial and arbuscular-mycorrhizal symbioses where chitin-derived Myc-LCOs and COs play an important role as signaling molecules ( Camps et al., 2015 ). Putative orthologs of NF receptors are present in C. glauca and A. glutinosa ( Hocher et al., 2011 ). We are currently studying whether these genes play a role in actinorhizal symbioses. Because LysM receptor kinases have been shown to recognize not only chitin-derived molecules but also peptidoglycans and exopolysaccharides ( Willmann et al., 2011 ; Kawaharada et al., 2015 ), orthologs of genes encoding NF receptors could be involved in the recognition of CgRHDF/NINA or AgRHDF, even if their chemical backbone is not chitin-based. Presence of NINA and CgRHDF in Other Frankia Strains In most legumes, NFs allow the specific recognition between the host plant and its symbiotic rhizobia ( Masson-Boivin et al., 2009 ; Oldroyd, 2013 ). Changes in specific decorations often result in host incompatibility ( Dénarié et al., 1996 ) but NFs from incompatible strains can induce symbiotic responses such as RHD and activation of symbiotic genes when applied at increased concentrations ( Roche et al., 1991 ). These results can be explained by changes of affinity between NFs and the cognate NF receptors able to recognize the chitin backbone and also the modified backbone structure. A misrecognition leads to decreased affinity but this can be compensated by increased amounts of substrate ( Dénarié et al., 1996 ). The symbiotic responses in non-host plants reported here point to a similar mechanism in C. glauca : strains from clades I and III possibly synthesize molecules sharing a common molecular backbone that is recognized by C. glauca receptors inducing RHD and the activation of Pro CgNIN promoter. Optimal recognition is achieved for compatible strains (clade Ic) and some related strains ( F. alni from clade Ia) but only the backbone would be recognized for more distant strains (clade III). This recognition is not detectable for non-infective strains (clade IV) and the distantly related strain F. coriariae suggesting that those strains do not produce sufficient amounts of this recognized backbone under the tested conditions. Comparison Between CgRHDF and AgRHDF Compared to CgRHDF and CgNINA, the distribution of AgRHDF seems less related to phylogeny. Generally, AgRHDF levels were stronger for clade III strains and several strains without any CgRHDF or CgNINA activity ( F. coriariae and two uninfective strains from clade IV). These differences suggest that the AgRHDF assay detects smaller concentrations of deforming factors or that root hairs of A. glutinosa are deformed by a wider range of molecules compared to C. glauca . This second hypothesis is in agreement with RHD detected in Alnus roots incubated with non- Frankia bacteria or fungi ( Berry and Torrey, 1983 ; Knowlton and Dawson, 1983 ; Prin and Rougier, 1987 ; Sequerra et al., 1994 ). Impact of Root Exudates We also found that the nature of RE used to incubate Frankia cultures could have an impact on RHDF and CgNINA activities. Unexpectedly lower CgRHDF and CgNINA activities were found when F. discariae was cultivated with RE from O. trinervis compared to RE from C. glauca . In legumes, specific flavonoids secreted by the host plant induce the expression of nod genes and the synthesis of NFs ( Oldroyd, 2013 ). RE secreted by the host plant probably also play a role in actinorhizae formation because the incubation with RE induces morphological changes in Frankia and accelerate the nodulation process ( Gabbarini and Wall, 2008 , 2011 ; Beauchemin et al., 2012 ). Myricaceae seed extracts also influence Frankia growth ( Bagnarol et al., 2007 ). In Alnus AgRHDF is reported to be produced either constitutively ( McEwan et al., 1992 ; Ghelue et al., 1997 ; Cérémonie et al., 1999 ) or upon induction with RE ( Prin and Rougier, 1987 ). Different plant RE have different effects on Frankia physiology. Information about CgRHDF is scarce but flavonoids isolated from Casuarina seeds have been shown to induce the production of CgRHDF by the Casuarina -infective BR strain ( Selim, 1995 ). Increased CgRHD activity in F. discariae incubated with C. glauca RE could be due to increased amounts of flavonoids in Casuarina RE compared to O. trinervis."
} | 4,629 |
28898068 | null | s2 | 8,226 | {
"abstract": "Programmable colloidal assembly enables the creation of mesoscale materials in a bottom-up manner. Although DNA oligonucleotides have been used extensively as the programmable units in this paradigm, proteins, which exhibit more diverse modes of association and function, have not been widely used to direct colloidal assembly. Here we use protein-protein interactions to drive controlled aggregation of polystyrene microparticles, either through reversible coiled-coil interactions or through intermolecular isopeptide linkages. The sizes of the resulting aggregates are tunable and can be controlled by the concentration of immobilized surface proteins. Moreover, particles coated with different protein pairs undergo orthogonal assembly. We demonstrate that aggregates formed by association of coiled-coil proteins, in contrast to those linked by isopeptide bonds, are dispersed by treatment with chemical denaturants or soluble competing proteins. Finally, we show that protein-protein interactions can be used to assemble complex core-shell aggregates. This work illustrates a versatile strategy for engineering colloidal systems for use in materials science and biotechnology."
} | 295 |
34568796 | PMC8449090 | pmc | 8,228 | {
"abstract": "Summary Organisms in nature grow with senses, nervous, and actuation systems coordinated in ingenious ways to sustain metabolism and other essential life activities. The understanding of biological structures and functions guide the construction of soft robotics with unprecedented performances. However, despite the progress in soft robotics, there still remains a big gap between man-made soft robotics and natural lives in terms of autonomy, adaptability, self-repair, durability, energy efficiency, etc. Here, the actuation and sensing strategies in the natural biological world are summarized along with their man-made counterparts applied in soft robotics. The development trends of bioinspired soft robotics toward closed loop and embodiment are proposed. Challenges for obtaining autonomous soft robotics similar to natural organisms are outlined to provide a perspective in this field.",
"introduction": "Introduction In the past, making robots like “ironman” with superior properties in terms of high-temperature resistance, superior strength, extreme environmental tolerance, fast speed, good precision, high controllability, and low cost has been the goal of robotics ( Figure 1 left) ( Kuindersma et al., 2016 ; Nelson et al., 2017 ). Although great progress has been made, their performances are still not satisfactory. For instance, motion agility and versatility are still inferior to those of a cat. Additionally, the performance in terms of energy efficiency, adaptability, self-repair, durability, and other aspects is far inferior to those of natural organisms. Figure 1 Schematic comparison of the composition and performance of natural lives, rigid robots, and soft robots Soft robotics expands the scope of research studies in robotics to serve new needs linked to adaptation and safety ( Coyle et al., 2018 ; Kim et al., 2013 ; Martinez et al., 2013 ; Pfeifer et al., 2012 ; Whitesides, 2018 ).With these goals in mind, many efforts have been devoted to adding sensation and computation skills to soft robotics without hindering their agility ( Shih et al., 2020 ; Thuruthel et al., 2019 ; Wang et al., 2018b , 2021 ). To obtain autonomy in dynamic environment, closed-loop controls are necessary to incorporate for self-correction of errors and smart human-robot interactions ( Figure 1 right) ( Kaspar et al., 2021 ). Living organisms possess the most ingenious and efficient closed-loop controls in the world. Their behaviors are not accomplished by a single organ but rather by rational and well-organized cooperation of multiple biofunctional modules ( Mykles et al., 2010 ). The muscles (actuation), skin (sensing), neural networks, and brain (computation) are seamlessly integrated, conferring them intrinsic adaptability, self-healing, robustness, versatility, and other properties that cannot be surpassed by conventional robots ( Figure 1 middle) ( Broom and Broom, 1981 ; Lehman, 2013 ). Facilitated by emerging technologies (e.g., additive manufacturing), abundant research studies have been conducted on bioinspired soft robotics ( Coyle et al., 2018 ; Ilami et al., 2021 ; Kovač, 2014 ). Despite remarkable progress of these years, a big gap between “bioinspired soft robotics” and natural organisms still exists. Living organisms depend on real-time and stored information to react, while soft robotics often suffer from nontrivial integration with actuators, sensors, and controllers. To achieve perception in soft robotics, important issues regard the development of soft actuators, stretchable sensors, and embedded electronics. Over the past few decades, several reviews dealing with the recent development of soft robotics have been reported. Such reviews have mainly focused on materials ( Appiah et al., 2019 ; Majidi, 2018 ; Miriyev et al., 2017 ; Truby and Lewis, 2016 ), manufacturing techniques ( Kuang et al., 2019 ; Rus and Tolley, 2018 ; Schmitt et al., 2018 ; Wallin et al., 2018 ), and applications ( Bao et al., 2018 ; Cianchetti et al., 2018 ; Runciman et al., 2019 ; Sitti, 2018 ). However, comprehensive reviews dealing with biologically inspired and integrated soft robotics are still lacking. In this review, an overview is provided to fill the gap. A transition from biological mechanisms to bioinspired soft actuators and sensors is first performed, and the four main integration trends of soft robotics are then provided to point out the future for integrated soft robotics."
} | 1,109 |
34783426 | PMC9299676 | pmc | 8,229 | {
"abstract": "Abstract Petrochemical based polymers, paints and coatings are cornerstones of modern industry but our future sustainable society demands greener processes and renewable feedstock materials. A challenge is to access platform monomers from biomass resources while integrating the principles of green chemistry in their chemical synthesis. We present a synthesis route starting from biomass‐derived furfural towards the commonly used monomers maleic anhydride and acrylic acid, implementing environmentally benign photooxygenation, aerobic oxidation and ethenolysis reactions. Maleic anhydride and acrylic acid, transformed into sodium acrylate, were isolated in yields of 85 % (2 steps) and 81 % (4 steps), respectively. With minimal waste and high atom efficiency, this biobased route provides a viable alternative to access key monomers.",
"conclusion": "Conclusion In conclusion, we have demonstrated a novel synthetic route towards two commonly used building blocks in the polymer, paint and coatings industry and omnipresent in materials that sustain modern society. It is shown that, starting from the biomass‐derived platform chemical furfural, by applying photooxidation, aerobic oxidation, hydrolysis and ethenolysis, the important monomer acrylic acid can be accessed in a mild and straightforward manner. The methodology presented incorporates many important features of green chemistry, by applying catalytic reactions, producing minimal waste, and providing high yields with excellent atom efficiency, proving the viability of alternative sustainable chemical routes for future chemistry.",
"introduction": "Introduction Changing the face of chemistry toward sustainable future processes and products, we are confronted with major challenges including the use of renewable feedstock and environmentally benign synthetic transformations based on the principles of green chemistry. \n [1] \n Building blocks like maleic anhydride and acrylic acid are among the cornerstones of modern materials, widely used in various polymers and coatings. \n [2] \n Current industrial production of both monomers, which globally exceeds several million metric tons per year, \n [3] \n relies on traditional petrochemical transformations; oil derived hydrocarbons are converted into building blocks via gas‐phase oxidations with multicomponent catalytic systems under forcing conditions. \n [4] \n With the growing environmental awareness and the quest to replace traditional petroleum based monomers directly by those derived from renewable resources, alternative mild and no‐waste producing chemical transformations leading to the same compounds have to be introduced (Scheme 1 ). \n [5] \n In the past decade, several routes towards maleic anhydride have been proposed ranging from biomass derived platform chemicals hydroxymethylfurfural (HMF), \n [6] \n furfural, \n [7] \n 1‐butanol \n [8] \n and levulinic acid \n [9] \n (Scheme 2 ). Similarly, advances towards biobased acrylic acid, \n [10] \n originating from lactic acid, \n [11] \n glycerol, \n [12] \n 3‐hydroxypropionic acid \n [13] \n and acrylonitrile, \n [14] \n indicate the importance of novel sustainable building blocks. Despite progress towards biobased routes towards these building blocks, applicability of these processes is still suffering from, among others, issues related to high temperatures, high pressures, low selectivities, low yields or catalyst instabilities. Very recently, Thomas et al. reported the further sustainable functionalization of acrylic acid towards various acrylate derivatives via a one‐pot catalytic transformation using biobased alcohols. \n [15] \n \n Scheme 1 General strategy for biobased acrylic acid obtained via a 4‐step green synthesis from the platform chemical furfural. Scheme 2 Previous described biobased synthesis routes towards maleic anhydride and acrylic acid starting from platform chemicals. General strategy, starting from furfural towards acrylic acid via maleic anhydride using aerobic oxidations and environmentally benign reaction conditions. Focusing on the monomer formation, we recently showed the synthesis and application of alkoxybutenolides as acrylate alternatives in the formation of biobased polymers and coatings. \n [16] \n As part of our program on clean photocatalytic oxidation, the platform chemical furfural is photooxygenated using singlet oxygen to yield hydroxybutenolide quantitatively (Schemes 2 and 3 ). Singlet oxygen is generated via triplet‐triplet annihilation of molecular oxygen upon excitation using a catalytic amount of photosensitizer and low‐energy visible light irradiation. The furan moiety of furfural undergoes a [4+2] cycloaddition with 1 O 2 , followed by transformation towards hydroxybutenolide. Scheme 3 Photooxidation of furfural to hydroxybutenolide using a rotary photoreactor setup. Rotary photoreactor scheme (bottom left), rotary photoreactor in operation (bottom right). Reaction conditions: 1 L flask, 1 M furfural in MeOH (10 mL), 10×8 W LED, 0.5 mol % methylene blue. It is worth noting that furfural is economically an attractive starting material with a price of $1 kg −1 \n \n [17] \n and has been listed by the U.S. Department of Energy as one of the most important (biobased) platform chemicals demonstrating the importance of discovering novel applications originating from furfural. \n [18] \n \n Here we report a green synthetic route to acrylic acid from biomass via furfural using sequential photochemical and catalytic oxidations with air and ethenolysis as major steps via hydroxybutenolide and maleic anhydride as key intermediates (Schemes 1 and 2 ). As hydroxybutenolide contains a hemiacetal moiety, specifically it features a single reduced anhydride, we envisioned a simple aerobic oxidation would suffice for the formation of the first building block maleic anhydride. For the synthesis of acrylic acid, one can recognize the similarities between the building block and maleic acid that is, the hydrolyzed form of maleic anhydride. It is worth noting that, although biobased routes towards maleic acid are known, \n [19] \n hydrolysis of maleic anhydride towards maleic acid is a less energy intensive process than the formation of the anhydride from maleic acid as in the latter transformation higher temperatures (50 °C vs. >130 °C) are required for the removal of water.[ \n 2 \n , \n 4 \n ] Direct ethenolysis of maleic acid would result in the production of two equivalents of acrylic acid starting from one equivalent of maleic acid (Scheme 2 ). Although ethylene is nowadays still produced on a scale exceeding hundred million tons annually from oil and gas, \n [20] \n biobased ethylene gas can be directly obtained from the dehydration of bioethanol. \n [21] \n Based on our design outlined here, we present methodology to access the key monomers maleic anhydride and acrylic acid directly from the platform chemical furfural in an environmentally benign synthesis with excellent yields, mild reaction conditions and high atom efficiencies.",
"discussion": "Results and Discussion Typically, photooxidation reactions, especially when performed on a larger scale, are limited by physical factors, such as light penetration and mass transfer of oxygen into the solution. \n [22] \n Previously, we have shown the upscaling of the photooxidation of furfural to hydroxybutenolide by developing a photo flow reactor for continuous production and a rotary photoreactor for larger batch processes. \n [16] \n The rotary photoreactor, modelled after Poliakoff and George, \n [23] \n is a versatile setup that allows for fast conversion. Optimal light penetration is achieved due to the thin film that is created by rotation of the reaction flask containing a small volume with a high concentration of furfural. Concurrently, having an oxygen atmosphere applied simply with an O 2 filled balloon results in excellent mass transfer into the solution which allows a fast conversion of 10 mmol (1 g) furfural in 20 min.[ \n 16 \n , \n 23 \n ] The photooxidation reaction was carried out under optimized conditions in methanol (MeOH) using methylene blue as photosensitizer. \n [16] \n Although a 400 W halogen lamp can be used, we opted for a more sustainable home‐built construction consisting of ten, 8 W (575 lm each) white Light Emitting Diodes (LEDs) (Scheme 3 ; Supporting Information, page 4). In the route towards the first polymer building block, maleic anhydride, we aimed to utilize molecular oxygen as green oxidant taking advantage of a report by Stahl and co‐workers on a copper‐catalyzed aerobic oxidative conversion of alcohols to esters proposed to involve hemiacetal intermediates. \n [24] \n \n We explored if a catalytic system based on copper(II), a 2,2‐bipyridine (bpy) ligand, a N ‐methylimidazole (NMI) ligand, TEMPO and oxygen, would be suitable to oxidize hydroxybutenolide to maleic anhydride. Initial experiments on this aerobic oxidation resulted in 58 % conversion, determined by 1 H NMR spectroscopy, towards the desired maleic anhydride after 8 h of stirring under an O 2 atmosphere (balloon) at room temperature (Supporting Information, Table S1). When the reaction was conducted for an extended time, in contrast to our expectations, a lower conversion was observed. Moreover, reproduction of the oxidation reaction led to inconsistent results in terms of conversion and selectivity. Literature reports indicate that maleic anhydride can readily undergo polymerization by imidazoles. \n [25] \n While NMI acts as a ligand for copper during the reaction, it is proposed that dissociation from the complex results to NMI acting as initiator for anionic polymerization of maleic anhydride. The oxidation reaction was monitored over time by 1 H NMR spectroscopy (Supporting Information, page 6, Figure S1) indicating that, while hydroxybutenolide is converted and the amount of maleic anhydride increases in the first few hours of the reaction (63 %, 3 h), it subsequently decreases over a prolonged period (27 %, 72 h), which is attributed to the anionic polymerization. In the attempt to prevent maleic anhydride from reacting, the oxidation was carried out in the absence of NMI. However, this resulted in low conversion (15 %) that could be increased only to 30 % after 16 h. Different bases were screened to enhance the conversion but resulted either in low conversion (Supporting Information, Table S1, entries 4, 7, and 8) or again resulted in conversion of maleic anhydride (Supporting Information, Table S1, entries 5, 6, and 9). Moving away from copper catalyzed aerobic oxidations, while taking into account the twelve principles of green chemistry, \n [1] \n we opted for iron catalyzed aerobic oxidation. \n [26] \n Here use is made of the combination of TEMPO and molecular oxygen, catalyzed by a transition metal, in this case iron(III), to oxidize secondary alcohols in 1,2‐dichloroethane (DCE). \n [26] \n With these specific reaction conditions, full conversion of hydroxybutenolide towards maleic anhydride was achieved, as determined by 1 H NMR spectroscopy (Table 1 , entry 1). Although promising, the use of a more environmentally benign solvent is highly warranted. Based on a list of solvents ranked regarding their environmental risk, acetonitrile was chosen as suitable candidate being considered a “green” solvent. \n [27] \n Using acetonitrile as solvent, under the same conditions with molecular oxygen, quantitative conversion towards maleic anhydride was observed after 8 h at room temperature (Table 1 , entry 2). Although the presence of sodium chloride has been shown to have a positive effect on the oxidation of allenols, \n [26] \n full conversion was still achieved in our case employing hydroxybutenolide as substrate in the absence of NaCl (Table 1 , entry 3, Figure 1 A ). Interestingly, the (relative) concentrations of both the iron catalyst and TEMPO are highly important (Table 1 , entries 4–6). Lowering the TEMPO concentration to 5 mol % drastically lowers the conversion from >99 % to 25 %. Surprisingly, when lowering the iron catalyst simultaneously to 2.5 mol % the conversion was only lowered to 48 %. Although not fully understood yet, the ratio of 1:2 between Fe(NO 3 ) 3 and TEMPO is of major importance. Decreasing both the concentrations further to 1.25 mol % and 2.5 mol %, respectively, decreases the conversion to 20 %. Finally, the oxidation was also carried out under ambient air instead of oxygen, which resulted in a slightly lower but still excellent and highly selective conversion of 98 % (Table 1 , entry 7).\n Table 1 Optimization table depicting oxidation of hydroxybutenolide to maleic anhydride with Fe(NO 3 ) 3 . Reaction conditions: hydroxybutenolide (1 mmol, 0.05 M) in solvent (5 mL), room temperature, 1 atm O 2 (balloon), 8 h. \n \n \n Entry \n \n Fe(NO 3 ) 3 \n \n [mol %] \n \n TEMPO \n [mol %] \n \n Solvent \n \n Yield \n [%] [a] \n \n \n 1 [b] \n \n \n 5 \n \n 10 \n \n DCE \n \n >99 \n \n 2 [b] \n \n \n 5 \n \n 10 \n \n CH 3 CN \n \n >99 \n \n 3 \n \n 5 \n \n 10 \n \n CH 3 CN \n \n >99 \n \n 4 \n \n 5 \n \n 5 \n \n CH 3 CN \n \n 25 \n \n 5 \n \n 2.5 \n \n 5 \n \n CH 3 CN \n \n 48 \n \n 6 \n \n 1.25 \n \n 2.5 \n \n CH 3 CN \n \n 20 \n \n 7 [c] \n \n \n 5 \n \n 10 \n \n CH 3 CN \n \n 98 \n [a] Yield determined by 1 H NMR spectroscopy in CDCl 3 using mesitylene (0.66 equiv.) as an internal standard. [b] 10 mol % NaCl added. [c] Reaction performed under air atmosphere. Wiley‐VCH GmbH Figure 1 A) Stacked 1 H NMR in CD 2 Cl 2 showing full conversion of the iron‐catalyzed aerobic oxidation of hydroxybutenolide (6.15 ppm and 7.25 ppm) into maleic anhydride (7.05 ppm) using mesitylene (mes) (0.66 equiv.) as an internal standard. B) Conversion of hydroxybutenolide into maleic anhydride over time followed by 1 H NMR spectroscopy in CDCl 3 (bottom). Reaction conditions: hydroxybutenolide (1 mmol), Fe(NO 3 ) 3 ⋅9 H 2 O (5 mol %), TEMPO (10 mol %), mesitylene (0.66 equiv.) in acetonitrile (5 mL) under 1 atm O 2 (balloon). The conversion of hydroxybutenolide to maleic anhydride was also followed over time by taking aliquots at certain timestamps, and measuring these by 1 H NMR spectroscopy, using the optimized conditions (Table 1 , entry 3). These results (Figure 1 ) complement the results of the optimization showing that 8 h of stirring at room temperature with our Fe‐based catalytic system provides full conversion towards maleic anhydride and no decomposition of the product was observed afterwards. It should be noted that several control experiments were performed to confirm the essential role of the key components (Fe‐catalyst, TEMPO) in this oxidative transformation (for experimental details and discussion, see Supporting Information, pages 8 and 9). Using the optimized conditions (Table 1 , entry 3), pure maleic anhydride was isolated by sublimation under reduced pressure as a white crystalline solid in 85 % yield. In the path towards the first biobased building block, maleic anhydride was obtained in a two‐step catalytic oxidation process (i.e. photocatalytic oxidation, Fe‐catalyzed oxidation, both using O 2 ), starting from the platform chemical furfural, with a total yield of 85 %. It should be noted that this synthetic route meets multiple requirements of the twelve principles of green chemistry (vide infra). \n [1] \n \n The next step in the journey towards biobased acrylic acid, maleic anhydride was hydrolyzed to maleic acid by heating (50 °C) in water overnight (Supporting Information, Figure S2). \n [4] \n Subsequent evaporation of the water layer resulted in quantitative amounts of the white solid maleic acid. Control experiments showed that fumaric acid was not formed confirming the selective and quantitative hydrolysis of maleic anhydride at 50 °C to maleic acid (Supporting Information, pages 10 and 11, Figures S2 and S3). With a short route to maleic acid in hand, we focused on the final step that is, metathesis of ethylene and maleic acid to produce directly two acrylic acid molecules. Realizing that high ethylene pressures might inhibit the activity of commonly used Ru‐catalyst based cross metathesis, \n [28] \n we decided to use a Schlenk technique similar to the one used in the aerobic oxidation of hydroxybutenolide to maleic anhydride, however, now equipped with an ethylene atmosphere. Starting at a high concentration of 3.3 M maleic acid, the reaction was carried out using the bench‐stable Hoveyda‐Grubbs II catalyst (5 mol %) in toluene. However, in view of the very poor solubility of maleic acid in organic solvents commonly used for metathesis reactions that is, toluene, dichloromethane (DCM), which resulted in no conversion, it was important to identify a solvent in which all the components, maleic acid, Hoveyda‐Grubbs II and ethylene, were soluble. Tetrahydrofuran (THF) qualified as such a solvent, and gratifying a yield of 76 % (NMR analysis) of acrylic acid was obtained (Table 2 , entry 3). Based on the promising results, extensive optimization of the reaction conditions was performed. Taking notice that olefin metathesis is an equilibrium reaction, \n [29] \n it is essential to know if equilibrium has been reached. Prolonging the reaction beyond 2 h did not have any effect on the conversion, indicating that the equilibrium was reached. According to Le Chatelier's principle, \n [29] \n the equilibrium can be pushed towards acrylic acid by increasing the ethylene:maleic acid ratio. In practice, this is done by simply lowering the concentration of maleic acid, which allows for an excess of ethylene in the liquid phase (for which the saturation concentration remains unchanged) with respect to maleic acid. Retaining a similar ethylene pressure, while lowering the concentration of maleic acid to 0.83 M, increased the conversion (Table 2 , entry 5) and at a concentration of 0.2 M full conversion towards acrylic acid was observed (Table 2 , entries 6 and 7).\n Table 2 Optimization table for ethenolysis of maleic acid towards acrylic acid. Reaction conditions: maleic acid, Hoveyda‐Grubbs II, 60 °C, 1 atm ethylene (balloon), 2 h. \n \n \n Entry \n \n Conc. \n [M] \n \n Cat. \n [mol %] \n \n Solvent \n \n Yield \n [%] [a] \n \n \n 1 \n \n 3.3 \n \n 5 \n \n Toluene \n \n 0 \n \n 2 \n \n 3.3 \n \n 5 \n \n DCM \n \n 0 \n \n 3 \n \n 3.3 \n \n 5 \n \n THF \n \n 76 \n \n 4 \n \n 1.7 \n \n 5 \n \n THF \n \n 75 \n \n 5 \n \n 0.8 \n \n 5 \n \n THF \n \n 91 \n \n 6 \n \n 0.4 \n \n 5 \n \n THF \n \n 97 \n \n 7 \n \n 0.2 \n \n 5 \n \n THF \n \n >99 \n \n 8 \n \n 0.2 \n \n 3 \n \n THF \n \n >99 \n \n 9 \n \n 0.2 \n \n 1 \n \n THF \n \n 92 \n \n 10 \n \n 0.1 \n \n 1 \n \n THF \n \n 89 \n \n 11 [b] \n \n \n 0.2 \n \n 3 \n \n THF \n \n 51 \n [a] Conversion determined by 1 H NMR spectroscopy in CDCl 3 using mesitylene (0.66 equiv.) as an internal standard. [b] Reaction performed at room temperature. Wiley‐VCH GmbH Studying catalyst loading, it was established that full conversion was still achieved using 3 mol %, while at 1 mol % catalyst loading the conversion was only slightly lower (92 %) (Table 2 , entries 8 and 9). Further lowering the concentration, while still using 1 mol % catalyst loading, did not result in a higher conversion (Table 2 , entry 10). Performing the reaction at room temperature resulted in a large drop in conversion, emphasizing that the temperature of 60 °C is necessary (Table 2 , entry 11). Under optimal conditions (3 mol % Ru‐catalyst, conc. 0.2 M, THF, 60 °C) the selective and high yield formation (>99 %) to acrylic acid was achieved (Figure 2 ). It should be pointed out that direct ethenolysis of maleic anhydride (which would shorten our route by one step) to yield acrylic anhydride is ineffective, presumably due to the equilibrium favoring the ring structure.\n Figure 2 Stacked 1 H NMR spectrum in CDCl 3 showing full conversion of the ethenolysis of maleic acid (top) to acrylic acid (0.2 M, bottom) using mesitylene (mes) (0.66 equiv.) as an internal standard. Isolation of acrylic acid was not performed directly as it is hard to completely remove traces of Ru‐catalyst at this stage and prevent polymerization. As the catalyst is still active after the reaction, evaporation of the solvent increased the concentration in which the equilibrium shifted towards maleic acid. Although catalyst deactivation is a known strategy, \n [30] \n we decided to isolate acrylic acid as sodium acrylate. Furthermore, acrylic acid has to be typically stored with an inhibitor to prevent radical polymerization, whereas sodium acrylate is bench stable. By addition of a 0.2 M solution of NaOH in water, acrylic acid was easily extracted from the organic layer and subsequent removal of the water led to pure sodium acrylate, which was isolated in 95 % yield (Supporting Information, Figures S10 and S11). We have shown here that acrylic acid can be prepared in a four‐step environmentally benign route from the platform chemical furfural with a total yield 81 %. The excellent atom efficiency is worth noting as no waste is produced in the oxidation, hydrolysis or ethenolysis, which above all generates two molecules of acrylic acid from a single molecule of maleic acid. Overall, this complementary biobased route towards these building blocks meets multiple requirements of the principles of green chemistry as illustrated by an analysis of this process based on key parameters as presented in Table 3 .\n Table 3 Relevant principles of green chemistry and their justification for the synthesis of maleic anhydride and acrylic acid. \n Principles of \n green chemistry \n [1] \n \n \n \n Justification \n \n Atom economy \n \n \n In the synthesis of maleic anhydride and acrylic acid, only methyl formate is produced as a side product during the photooxidation of furfural . \n \n \n \n \n \n Design for energy efficiency \n \n \n The oxidation of furfural is performed using visible‐light photocatalysis at room temperature. Scalable setups have been designed with energy‐efficient lamps (LED) . \n [16] \n \n \n \n \n \n \n \n Safer solvents and auxiliaries \n \n \n In the synthesis towards maleic anhydride and acrylic acid, methanol, acetonitrile, water and THF are used. These solvents are ranked high in terms of environmentally safe solvents . \n [27] \n \n \n \n \n \n \n \n Use of renewable feedstocks \n \n \n The platform chemical furfural, derived from the acid‐mediated dehydration of lignocellulose (H 2 O as waste), is used as the starting material for the synthesis of maleic anhydride and acrylic acid . \n [31] \n \n \n \n \n \n \n \n Catalysis \n \n \n The photooxidation of furfural, the aerobic oxidation of hydroxybutenolide and the ethenolysis of maleic acid are catalytic methods in which only molecular oxygen is used as a stoichiometric reagent. No high‐energy oxidants are required . \n \n \n \n \n \n Inherently safer chemistry for accident prevention \n \n \n All reactions have been optimized to allow safer handling (ambient or lowered temperatures, atmospheric pressures) . \n Wiley‐VCH GmbH"
} | 5,684 |
22481885 | PMC3313501 | pmc | 8,230 | {
"abstract": "Methanogenic archaeans are organisms of considerable ecological and biotechnological interest that produce methane through a restricted metabolic pathway, which culminates in the reaction catalyzed by the Methyl-coenzyme M reductase (Mcr) enzyme, and results in the release of methane. Using a metagenomic approach, the gene of the α subunit of mcr ( mcr α) was isolated from sediment sample from an anoxic zone, rich in decomposing organic material, obtained from the Tucuruí hydroelectric dam reservoir in eastern Brazilian Amazonia. The partial nucleotide sequences obtained were 83 to 95% similar to those available in databases, indicating a low diversity of archaeans in the reservoir. Two orders were identified - the Methanomicrobiales , and a unique Operational Taxonomic Unit (OTU) forming a clade with the Methanosarcinales according to low bootstrap values. Homology modeling was used to determine the three-dimensional (3D) structures, for this the partial nucleotide sequence of the mcrα were isolated and translated on their partial amino acid sequences. The 3D structures of the archaean Mcrα observed in the present study varied little, and presented approximately 70% identity in comparison with the Mcrα of Methanopyrus klanderi . The results demonstrated that the community of methanogenic archaeans of the anoxic C1 region of the Tucurui reservoir is relatively homogeneous.",
"introduction": "Introduction The organisms of the Domain Archaea ( Woese et al. , 1990 ) constitute a considerable proportion of the prokaryotes found in both aquatic and terrestrial ecosystems, where they are active in the carbon biogeochemical cycle. From a physiological viewpoint, the Domain is divided into three different groups: the haloarchaea, methanogenics, and thermophilics, the latter being dependent on sulfur ( Valentine, 2007 ). The methanogenic archaea are of special interest because of their capacity to reduce CO 2 and other compounds in the presence of H 2 ( Conrad, 1999 ), under anoxic conditions, to produce methane through a process known as methanogenesis ( Zehnder and Brock, 1979 ). Methane is one of the greenhouse gases and, with an atmospheric half-life of only 8.9 years, it inflicts an effect on global warming 24 times greater than that contributed by CO 2 ( Houghton et al. , 2001 ). As a non-fossil hydrocarbon, it also generates interest as a potential biofuel ( Hansen et al. , 2006 ; Blottnitz and Curran 2007 ; Tilche and Galatola, 2008 ). Archaean communities are distributed worldwide, and are known for their considerable heterogeneity ( Schleper et al. , 2005 ). Studies of the diversity of archaean communities have been conducted in a variety of environments, from hot ( Martinez et al. , 2006 ) and hyper-saline ( Oren, 2002 ) lakes, anoxic zones ( Lehours et al. , 2007 ) and marine sediments ( Teske and Sørensen, 2008 ) to the digestive systems of ruminants and humans ( Lange and Ahring, 2001 ). The reservoirs of hydroelectric power stations are known to emit large quantities of methane from biological sources, due to their tendency to accumulate organic material and create an environment favorable for methanogenesis ( St Louis et al. , 2000 ; Kemenes et al. , 2007 ). With the World’s most extensive system of river basins at its disposal, the Brazilian electricity network is centered on hydropower, which was responsible for 77.4% (374.015 GWh) of the electrical energy produced in the country in 2007 ( Brazilian Ministry for Mines and Energy, 2009 ). The country’s main hydropower potential is in its northern region, along the tributaries of the Amazon basin. The construction of hydroelectric dams in the Brazilian Amazon over the past three decades has been the subject of considerable controversy between government agencies and many sectors of Society. By 2010, the Brazilian Ministry for Mines and Energy plans the construction of further three major dams, which will feature increased productivity, reduced flooding, and minimal impact on local fauna, flora, and indigenous societies, in comparison with previous projects. Currently, the Tucuruí dam (UHE Tucuruí), which is located on the Tocantins River in the southeast of the Brazilian state of Pará (03°455′8″ S, 49°40′21″ W), is the country’s largest and most productive hydropower plant. The Tucuruí Reservoir is 170 km long, with a perimeter of 7700 km, a depth of 60 m, and a total of 1500 islands ( Tundisi, 2007 ). Analyses conducted by Centrais Elétricas do Norte do Brazil, the state company responsible for Tucuruí hydroelectric power plant, as part of its quality control program, has indicated the existence of specific areas, such as Region C1 ( Figure 1 ), which are favorable for methanogenesis. During most of the year, C1 has an anoxic environment, with still water and abundant organic material. Culture-independent analyses based on ssuRNA sequences have confirmed the presence of methanogenic archaea at various levels in the water column (Graças et al. , personal communication), and previous studies of methane emissions ( Fearnside, 1997 ; Lima, 2005 ) have confirmed the elevated methanogenic capacity of the reservoir. Studies of the characteristics and diversity of archaean communities have been based on the nucleotide diversity of 16S rRNA ( Keough et al. , 2003 ; Lehours et al. , 2005 ; Schleper et al. , 2005 ). Studies ( Shapiro et al. , 2007 ) have also demonstrated the importance of the secondary structure of 16S for the analysis of the diversity and phylogenetic relations of prokaryotes. Similarly, analysis of the nucleotide sequence of the Methyl-coenzyme M reductase ( mcr ) gene has proven to be an invaluable tool for the study of diversity, given that this gene is present in all methanogenic archaea ( Bapteste et al. , 2005 ). Homology modeling is a well-known analytical tool through which it is possible to modeled the three-dimensional (3D) structure of a given protein based on knowledge of its primary structure ( Martí-Renom et al. , 2000 ). This approach is only possible because the 3D structure of homologous proteins is conserved during the evolutionary process, especially in the case of the functional residuals, given that conservation of the structure is crucial to the maintenance and development of specific functions ( Höltje et al. , 2003 ). Therefore, the strategy of this approach is based on the fact that, during the evolutionary process, the structural configuration of a protein tends to be better preserved than its aminoacid sequence, and that minor changes in the sequence generally result in only slight modifications of the 3D structure (Nayeem et al. , 2003). The comparative or homology modeling of the 3D structure of the target protein is possible if at least one resolved 3D structure sequence which is homologous to the amino acid sequence is available ( Höltje et al. , 2003 ). In the present study the diversity of archaean communities derived from samples of sediment collected from C1 region of UHE Tucuruí was accessed. In order to achieve, partial nucleotide sequences of mcr α were analysed and homology modeling was carried out.",
"discussion": "Discussion The Mcr protein is made up of three subunits (α, β, and γ), which are each formed by two chains, where the α subunit, is the most popular for the inference of phylogenetic relationships ( Hales et al. , 1996 ; Luton et al. , 2002 ; Hallan et al. , 2003 ; Castro et al. , 2004 ; Springer et al. , 2005; Juottonen et al. , 2006 ). Springer et al. (1995) and Luton et al. (2002) recorded broad similarities among the topologies of the arrangements obtained for 16S rRNA and the partial sequences of the mcr α gene. Preliminary studies of the Tucuruí ecogenome (Graças et al. , personal communication) have shown, through the analysis of the 16S rRNA gene, the predominance of the order Methanomicrobiales in region C1 of UHE Tucuruí. The present analysis of mcr α corroborated these findings in 16S, and indicated an abundance of organisms closely related to the Methanomicrobiales. This order encompasses 24 recognized species in nine genera and three families, although the phylogeny of this group is extremely complex, and still not fully understood ( Cavicchiolli, 2007 ). The first rarefaction curve was based on the sequences obtained using the primers of Hales et al. (1996) . The curve tended to plateau, suggesting that the number of unique OTUs is a good representation of the diversity of the archaean community of the C1 region of the Tucuruí reservoir. However, the second rarefaction curve which was based on the addition of the OTUs obtained using the primers of Luton et al. (2002) , revealed a much higher diversity in comparison with the first curve. This indicates that, when only one pair of primers is used, the preferential amplification of some organisms may result in the underestimation of the methanogenic biodiversity of an area such as C1. The addition of the OTUs derived from the primers according to Luton et al. widened the spectrum of organisms identified, and revealed a much higher diversity. The surveys of Earl et al. (2003) and Castro et al. (2004) in the Priest Pot nature reserve in the UK and the WCA-2A conservation area in the northern Florida Everglades, (respectively) also found a predominance of methanogenic archaeans phylogenetically related to the Methanomicrobiales. These authors justified the pattern as the result of the higher temperatures and greater solar radiation during the summer months, which resulted in an increase in the microbian metabolism and anaerobic fermentation of organic material. In this case, the alterations in the biochemical composition of the water would have favored hydrogenotrophic methanogenesis, which is the preferential mode of biomethanogenesis in the Methanomicrobiales ( Cavicchiolli, 2007 ). This same hypothesis may apply to region C1 of the Tucuruí reservoir, given that the area is characterized by relatively high temperatures and levels of solar radiation, as well as abundant organic material, throughout the year. An additional factor favoring the production of hydrocarbons is the fact that the region is anoxic. Methanogenesis is carried out through the syntrophic association of micro-organisms capable of fermenting glucose into shorter-chain carbon compounds such as formate, acetate, methanol, methylamines, methylthios and CO 2 ( Thauer, 1998 ). This process can have three distinct routes: hydrogenotrophic ( Horn et al. , 2003 ), acetylclastic or methylotrophic ( Ferry, 1999 ). These pathways vary in the preferential use of different substrates for the acquisition of metabolic energy. In general, the three routes converge in a sequence of seven reactions, the last of which is induced by the enzyme Mcr, which catalyzes the synthesis of hetero-disulfide (CoM-S-S-CoB) from the substrates coenzyme M (CH 3 -S-CoM) and coenzyme B (H-S-CoB), liberating methane (CH 4 ) as a subproduct ( Thauer, 1998 ): CH 3 - S - CoM + H - S - CoB CoM - S - S - CoB + CH 4 Little is known of the 3D structure of Mcrα. Crystallographic Mcr could only be found in the PDB for three methanogenic archaeans - M. klanderi ( Grabarse et al. , 2000 ), M. barkeri ( Grabarse et al. , 2000 ), and Methanobacterium thermoautotrophicum ( Ermler et al. , 1997 ). In the present study, the Mcrα of the first two of these organisms were selected as the template for homology modeling. Despite the apparent phylogenetic distance between M. klanderi and M. barkeri , the template and target were highly homologous. For the analysis of 3D structure, the three subgroups of the gene tree were treated separately. The RMS values for the 28 structures obtained varied from 0.06 to 0.48, which indicates only a slight deviation between targets and the templates, supported by their identity values (68%–70%). No significant differences were observed in the folding of the 3D structures when target and templates of the same clade were superimposed. Given this, representative structures of each clade were selected randomly for the comparisons among the different clades, which indicated that the 3D structures of the Mcrα of archaeans of different orders tend to be well conserved ( Fong et al. , 2007 ), despite their characteristic differences in the distribution of amino acid residues on the protein surface ( Grabarse et al. , 2000 ). The Ramachandran graphs indicated that all 28 targets structures analyzed were of good stereochemical quality, given that 85 to 86% of the residuals of the amino acids modeled presented spatially viable angles, being located in favorable areas of the graph, where the angles and residual energy are compatible ( Morris et al. , 1992 ). The variety of possible random protein configurations and the respective deviation of the total structures’s energy are shown by z-score graphs ( Sippl, 1993 ). These scores varied between −5,66 and −5,92 for the 28 targets obtained from the homology modeling, which indicates that the structures generated are energetically stable. The scarcity of crystallographic models of Mcr for the archaea of the order Methanomicrobiales did not limit the effectiveness of the homology modeling, given the high identity scores (68%–70%) recorded between the template and target structures, subsequently confirmed by the generation of structures with high indices of quality. The nucleotide sequence of the alpha subunit of the mcr gene and the reduced number of unique OTUs indicates the existence of a relatively low diversity of methanogenic archaea (83%–95% identity with the NCBI data bank) in region C1 of the ecogenome of the Tucuruí reservoir. The mcr α gene has proven to be a useful tool for the study of diversity in methanogenic archaeans, especially considering its ample distribution among these organisms. In the archaeans of the C1 region of the Tucuruí reservoir low levels of polymorphism in nucleotides, predicted amino acids, and Mcrα 3D structures. The majority of the mutations observed in the nucleotide sequences did not cause any changes in the residuals, and where this did occur, the substitutions modified the 3D structure of the mcr alpha fragment only slightly. The present study is the first of its kind, to our knowledge, to investigate the ecology of the methanogenic archaea of a tropical hydroelectric reservoir, and contribute to the understanding of the principal organisms responsible for biomethanogenesis in the aquatic environments of the Amazon basin. The study also demonstrated the usefulness of the comparative modeling of proteins as a tool for the elucidation of the 3D structure of the Mcr protein alpha subunit, and that this constitutes a promising approach for the understanding of the equilibrium between methanogenesis and methanotrophy in Amazonian hydroelectric reservoirs."
} | 3,720 |
35149704 | PMC8837802 | pmc | 8,231 | {
"abstract": "Microbial composition and functions in the rhizosphere—an important microbial hotspot—are among the most fascinating yet elusive topics in microbial ecology. We used 557 pairs of published 16S rDNA amplicon sequences from the bulk soils and rhizosphere in different ecosystems around the world to generalize bacterial characteristics with respect to community diversity, composition, and functions. The rhizosphere selects microorganisms from bulk soil to function as a seed bank, reducing microbial diversity. The rhizosphere is enriched in Bacteroidetes, Proteobacteria, and other copiotrophs. Highly modular but unstable bacterial networks in the rhizosphere (common for r -strategists) reflect the interactions and adaptations of microorganisms to dynamic conditions. Dormancy strategies in the rhizosphere are dominated by toxin–antitoxin systems, while sporulation is common in bulk soils. Functional predictions showed that genes involved in organic compound conversion, nitrogen fixation, and denitrification were strongly enriched in the rhizosphere (11–182%), while genes involved in nitrification were strongly depleted.",
"introduction": "Introduction The rhizosphere is home to a rich diversity of microorganisms, many of which benefit plants by suppressing pathogenic invasions and helping to acquire nutrients from the soil 1 , 2 . Understanding the taxonomic and functional components of the rhizosphere microbiome and how they differ from those of the bulk soil microbiome (here: soil without direct root effects) is crucial to manipulate them for sustainable ecosystem functioning. Recent advances in sequencing have enabled significant progress in the elucidation of the rhizosphere microbiomes of various plants. The diversity and composition of the rhizosphere bacterial community is a function of both plant species and soil properties 3 – 5 . Although plant species or even species genotypes tend to assemble relatively distinct rhizobacterial communities 6 – 8 , these communities can be largely similar even in different environments across geographical regions 9 , 10 . Plants exert selective effects on rhizobacterial assemblages in bulk soil pool to acquire specific functional traits needed for plant fitness 8 , 11 , 12 . As a result, the rhizosphere microbiome greatly expands the functional repertoire of the plant 13 . Despite the specific nutrient requirements of plants, disease control mechanisms, and edaphic habitats, rhizosphere environments (with excess available carbon) provide broadly similar conditions for microbial life. All plants are mineral resource-limited organisms and are often affected by pathogens. To overcome these limitations, plants form rhizo-assemblages independent of their host phylogeny. Thus, all plants can exert general selective effects directed toward nutrient acquisition or pathogen suppression, regardless of their geographic origin or recent location. These general patterns in the rhizosphere and bulk soils with respect to the taxonomic and functional profiles of bacterial communities remain largely unexplored. However, this information is critical to understand and manage microbial functions in ecosystems to support future plant growth in rapidly changing environment. Recently, high-throughput sequencing of culture-independent marker genes (typically, 16S rRNA in the case of bacteria), has greatly expanded the repertoire of microorganisms living in soils 14 , and many studies have characterized root-associated microbial communities 9 . Since the raw data from most studies must be deposited in a public gene bank, this has resulted in a huge and extensive rhizosphere sequencing data set. These high-resolution nucleic acid-based molecular techniques provide excellent insights into specific microbiome members in soil habitats. Research priorities for harnessing the rhizosphere microbiome for sustainable ecosystems development include elucidating the functional mechanisms that mediate plant-microbiome interactions and defining the core of the plant microbiome 15 , 16 . The methodological advances made in these priority research areas have provided a vast amount of data for integrative analysis and subsequent synthesis. This has paved the way for investigating the general principles of rhizosphere microbiome selection from bulk soils. The rhizosphere contains numerous niches for the growth and proliferation of a phylogenetically diverse array of microorganisms, including bacteria, archaea, fungi, protists, nematodes, and viruses, but bacteria and, to a lesser extent, fungi are the most dominant forms and are fairly well-studied compared to other members of the community. Thus, we are attempting to infer the composition of bacterial communities in the rhizosphere and bulk soil, determine the rhizosphere bacteriome properties common to a wide range of plants and environmental conditions, and bridge the gap between general rhizobacterial assemblages and functions associated with community-wide dormancy capacity, heterotrophic strategies, and individual nutrient cycling processes. Here, we collected 16S rRNA amplicon-based sequencing data from all available gene banks to characterize the general bacteriome and synthetically analyzed the data using state-of-the-art bioinformatics methods. Bulk soil serves as a microbial reservoir for the rhizo-microbiome 17 and is home to considerable microbial diversity that is explicitly shaped by the environmental factors of each microhabitat 18 , 19 . We, therefore, hypothesized that (i) rhizobacterial populations are recruited primarily from the corresponding bulk soil, but are preselected by excess released root carbon, so that bacterial diversity is generally lower in the rhizosphere and bacterial networks are less stable. Compared to bulk soils, the rhizosphere can fuel its microbiota by providing abundant and readily available energy and carbon sources 20 , 21 , but microorganisms in root-free soil are always carbon limited 22 . We therefore further hypothesized that (ii) the rhizosphere is home to more abundant copiotrophic bacteria than the bulk soil. To this end, we evaluated the community weighted mean 16S rRNA gene copies (rRNA operons) because copiotrophs are assumed to have more rRNA operons than oligotrophs 23 , 24 . Since all nutrients flow from the bulk soil to the roots, we further hypothesized that (iii) the functional capacity involved in the carbon and nitrogen transformation would be greater in the rhizosphere. Plants deposit a significant proportion of their photosynthates in the rhizosphere as rhizodeposits and root debris. The rhizodeposits, including amino acids, carboxylic acids, sugars and polymeric carbohydrates such as cellulose and hemicellulose 25 , are not only critical carbon and energy sources for rhizobacteria but also key attractants for plant pathogens. We finally hypothesized that (iv) functions related to lignocellulose degradation and phytopathogens are overexpressed in the rhizosphere due to the accumulation of root litter and the attraction of pathogens to live plants. In this work, we analyzed and synthesized a very broad range of taxonomic and functional features of the bacteriome in the rhizosphere compared to bulk soil, and generalized these compositional changes to other factors such as plant species, geographic environment, and soil properties. This provided general principles for the selection of microorganisms around living roots and laid the foundation for harnessing the power of the microbiome for sustainable terrestrial ecosystem functioning.",
"discussion": "Discussion Bacterial diversity and composition in the rhizosphere and bulk soil This meta-analysis is based on pairwise data (rhizosphere vs. bulk soils) from amplicon sequencing approaches to characterize taxonomic and functional features. The meta-analysis provided fundamental insights into the plant rhizosphere microbiome on an intercontinental scale, revealing plant-driven microbial taxa and their functional properties in this unique but cohesive habitat. The design of the meta-analysis allowed us to examine the general effects of plants on the rhizosphere bacterial communities across broad range of soil properties and geographic environments. This highlights the benefit of using sequencing data to synthesize general microbiome patterns and to indicate specialized functions and life strategies of microbial taxa based on niche differences between the rhizosphere and bulk soils. The fact that rhizosphere microbiota differs from bulk soil microbiota is well documented 12 , 16 , 26 – 28 , and this is attributed to significant differences in physico-chemical properties driving niche differentiation 4 , 21 , 29 , 30 . In addition to environmental differences between niches, the general contrast between bulk soil and the rhizosphere was a very important cause of differences in microbiota composition 12 , 31 . Bacterial observed species richness, Shannon’s diversity index, and Faith’s phylogenetic diversity in rhizosphere were generally lower than in bulk soil under all environments. Thus, bacterial diversity decreases as substrate availability increases, a condition that is common in the rhizosphere. The general view is that the rhizosphere microbiota, a subset of the community in bulk soil, has certain similar traits in all plants. This underlines the selective effect of the rhizosphere, which has some general consequences on plant rhizobacterial assemblages. This holds true even for bacteria belonging to broad range of classes, orders and families. Although the magnitude of the effects differs between plants groups (Fig. 1 ), we emphasize that, even when genotypic and environmental differences were taken into account, certain similarities in the selection of microorganisms common to the rhizosphere were still observed 4 , 8 , 32 . Of course, specific environmental conditions will substantially change the magnitude of the rhizosphere effect: the rice rhizosphere is a very good example. Most of the effects common to the rhizosphere of other plants were canceled out in paddy soils. Paddy soils are often anaerobic, and rice plants, therefore, have a well-developed aerenchyma. Due to the release of oxygen around the roots, the Eh value and oxygen content of the rice rhizosphere are much higher than in the bulk soil 33 , 34 and vary widely. Thus, the rice rhizosphere is inhabited by a wider phylogeny of both anaerobic and aerobic bacteria. These results indicate that environmental heterogeneities, such as root exudates, Eh, and soil moisture changes, interact to give rise to selective effects in the rhizosphere. Specific microbial taxa recruited to the rhizosphere from soil reservoirs can apparently form a distinct core microbiome 12 , 14 . The core microbiome around the roots contributes to plant growth, and fitness 17 . To date, the core microbiome of plants, whether in the rhizosphere, endosphere, or phyllosphere, has been defined primarily on the basis of taxonomic markers. However, we emphasize that more attention should be directed to identifying microbes with common functions that are selected for in the general rhizosphere environment. Thus, defining the microbiome based on function should make it easier to manipulate the community for useful purposes. The present comprehensive analysis revealed several predominant taxa that are consistently enriched in the rhizosphere, including the phyla Bacteroidetes and Proteobacteria (Fig. 2 ). This result underscores the fact that these phyla are generally adapted to C-rich conditions (common in the rhizosphere) for high metabolic activity, fast growth and propagation 30 , 35 , and are consequently very similar across diverse plant species. They are generally considered to be copiotrophs, or weedy fast-growing microbiota whose populations fluctuate opportunistically 28 . In contrast to the rhizosphere, bulk soils are generally enriched by other dominant phyla including Acidobacteria (Fig. 2b ), which are oligotrophs 36 , 37 . Interestingly, several phylum-level taxa are similar between the rhizosphere and bulk soil, but differences are seen at finer taxonomic resolution (Fig. 2 ). Consequently, the general patterns that emerge based on the selective effect of the rhizosphere depend on taxonomic resolution and fundamental niches at the level of classes and families. Microbiome formation: from structure to functions In microbial networks, highly interconnected species are grouped into modules, in which species interact more frequently and intensively than in the rest of the community. The modularity of the rhizosphere bacterial network is higher than that of the bulk soil (Fig. 3a, b ). Since modules can be interpreted as microbial niches 38 , 39 , one possible explanation would be that niche differentiation is more pronounced in the rhizosphere 25 —both spatially and temporally. Modularity is one of the main organizing principles of biological networks 40 , and the higher modularity of the rhizosphere may indicate a more complex topological structure. Despite the higher modularity, the bacterial co-occurrence network in the rhizosphere is less stable (Fig. 3c ), because the rhizosphere is characterized by very high temporal dynamics compared to the more static conditions in the bulk soil 30 , 41 . Plant species selectively enrich specific microorganisms by investing in root exudates to feed their rhizosphere microbiota 1 , 42 . The structure of the indigenous rhizosphere microbial community often varies considerably across host species 43 . In the soybean rhizosphere, the microbial community was selected through niche filtering, whereas the bulk soil community arose through neutral (stochastic) processes 44 . The rhizosphere network allocates more modules for executive functions, but fewer species for network robustness, partly reflecting the rapid element cycling in the rhizosphere 45 , 46 . The rhizosphere and bulk soil are characterized by different microbial dormancy-dominating strategies: sporulation factors and toxin–antitoxin systems (Fig. 4a, b ). Sporulation factors were more abundant in the bulk soil, while toxin–antitoxin systems were enriched in the rhizosphere (Fig. 4a, b ). During plant growth, roots actively and passively release a wide range of organic compounds into the rhizosphere. These compounds are the driving force of microbial growth and activity 29 , 47 , 48 . The sporulation factor was abundant in the bulk dryland soils but was not significant in paddies (Fig. 4b ). Hence, bacterial sporulation is more common in dryland soils because the environmental conditions in paddies are more stable and homogeneous, and paddies are always moist. The lower importance of sporulation in the rhizosphere (compared to bulk soil) indirectly confirmed the buffered amplitude of moisture variation associated with the conditions of mucilage swelling 49 – 51 . Genes related to dormancy/sporulation increase strongly with aridity 52 . Toxin–antitoxin systems consist of genes that encode a toxin protein that inhibits cell growth and an antitoxin that counteracts the toxin 53 . It is generally assumed that rRNA operons in prokaryotic organisms are able to reflect their heterotrophic strategies 54 . Copiotrophs ( r -strategists) have a high ribosome content, in part, by maintaining multiple ribosomal RNA operon copies in their genomes to achieve high growth rates 55 . The rhizosphere was inhabited by a greater number of copiotrophs (e.g., Bacteroidetes and Proteobacteria), as confirmed by higher rRNA operon counts associated with them (Fig. 4c ). These results confirm that the major groups in the rhizosphere are fast-growing bacteria, especially the phyla Proteobacteria and Bacteroidetes 56 . Oligotrophs (K-strategists) have lower operon counts than copiotrophs 55 . A lower rRNA operon copy number is common in oligotrophic microbiota indicating slower growth rates and more stable populations. A functional feature of bacteria is that the copy number of the rRNA operon increases in response to the availability of resources 24 . Organisms with multiple operons are referred to r -strategists and tend to dominate in resource-rich environments and respond more rapidly to nutrient inputs 57 – 59 . Therefore, copiotrophs will predominate when resources are abundant, such as in rhizosphere habitats where plants secrete photosynthates to produce available C and energy. As the interface between roots and soil, the rhizosphere hosts an abundant and diverse bacteriome that drives the soil C and N dynamics. Genes related to C and N transformation were all broadly higher in the rhizosphere (Fig. 5 ), due to the larger amount of plant-derived organic compounds, leading to higher microbial activity and abundance in the rhizosphere. This is directly confirmed by the fact that the rhizosphere is rich in almost all N-cycling functions (except nitrification), and the magnitude of their effects is greatly dependent on plant groups in agroecosystems (Fig. 5 ). Nitrification is prone to take place in aerobic conditions, but the rhizosphere generally suffers from oxygen deficiency 60 because roots and microorganisms consume more oxygen than the bulk soil. Likewise, activities related to the decomposition and transformation of organic compounds, such as cellulolysis, xylanolysis, ligninolysis, ureolysis, and chitinolysis were generally more intense in the rhizosphere, reflecting the higher abundance and activity of bacteria degrading these substances. Methanol oxidation and methylotrophy genes are much more abundant in the rhizosphere than in the bulk soil (except in rice paddies). Methylotrophy is higher in the rhizosphere, but in paddy soils this difference is smoothed out because of the aerobic microenvironment around rice roots and the high dilution of available reduced C in water. Functional groups of plants, such as grasses and forbs, have distinct characteristics and fill specific niches 56 , 61 . Grasses have a much denser root systems, finer roots, and higher root biomass than forbs 62 , and thus accelerate nutrient cycling by intense litter and rhizodeposition decomposition. Hence, plant functional groups are likely to be critical drivers of rhizosphere functions. Particularly in the rhizosphere, plants are continuously challenged by thousands of microbial populations, including commensals, pathogens, and symbionts. Plant pathogens and N-fixers (e.g., Rhizobium sp., etc.) are enriched in the rhizosphere (Fig. 5 ) because their reproduction and functioning depend on the supply of organic matter from the plant host. Although the rhizosphere is a dynamic environment, and the microbiome evolves rapidly in space and time, evidence is accumulating to confirm that plants shape the rhizosphere microbiome to their own benefit and skillfully utilize the microbial functional repertoire 13 . As expected, plant pathogens were more enriched in the rhizosphere than in the bulk soil, which may be due to the following reasons: (1) most plant pathogens grow saprophytically in the rhizosphere and derive their basal energy from the roots 63 ; (2) bacterial pathogens, in particular, require a wound or natural opening to penetrate into the plant before they establish a parasitic relationship; (3) only a few groups of pathogenic bacteria are considered to be soilborne, probably because non-spore forming bacteria cannot survive well in bulk soils for long periods 64 ; (4) certain pathogens, either candidate symbionts or stealthy pathogens, favor the colonization of the rhizosphere 65 , 66 . The rhizosphere is both the playground where soilborne pathogens infect plants and the battlefield where the complex rhizosphere community, including both microflora and microfauna, interacts with pathogens and influences the outcome of infection 64 . The number and diversity of deleterious and beneficial microorganisms are related to the quantity and quality of rhizodeposits and the outcome of microbial interactions in the rhizosphere. Nevertheless, identifying the equilibrium conditions for plant fitness remains a challenge, as does establishing a balance between passive attack by pathogens and active recruitment of beneficial bacteria. By integrating sequencing data from multiple studies, we have generalized the main differences in the rhizosphere and bulk soil microbiomes with respect to bacterial diversity, composition, selection of specific groups, co-occurrence network, and a very broad range of functions (Fig. 6 ). The bacterial diversity in the rhizosphere is reduced by 0.9–5.3% and represents a subset of the bulk soil community. The bacterial community in the rhizosphere is highly enriched in copiotrophs such as Proteobacteria and Bacteroidetes, while bacteria such as Chloroflexi, Acidobacteria, and Nitrospirae are significantly reduced. Because of the surplus of organic C in the rhizosphere and rapid nutrient cycling, fast-growing bacteria have an excess of functions related to C transformation and plant pathogenesis, but the functions responsible for nitrification are depleted. Indirect evidence supporting the generalizations presented here is that land use regimes and plant functional groups influence almost all rhizosphere effects on bacterial diversity and functions. Based on our results, we validly sketched the generalized rhizosphere effects on bacteriome across continents, even though the soil properties and geographic distance showed certain contributions on the bacteriome variation (Supplementary Fig. 11 ). The selective influence of the rhizosphere on the formation of microbial communities overshadows to some extent the differences in soil, plant, or climate even at continent scale. This suggests that the rhizosphere is the powerful factor shaping the composition, structure, and functions of the soil microbiome and, thus, a key factor in the cycling of biogenic elements. Therefore, the present study expands our knowledge of the critical role of the rhizosphere effects in recruiting bacterial populations. Fig. 6 Conceptual figure showing the enrichment (red) and depletion (blue) of bacterial community taxa and functions in the rhizosphere relative to bulk soil. The vertical arrows correspond to the intensity of the changes. The light peach-colored area around the root reflects the enrichment with available organics caused by exudates. With the rapid development of instrumental and molecular techniques, there have been many attempts to consider the functional traits of microorganisms at the community level based on their ecological relevance. With further expansion of available soil metagenomic/metatranscriptomic datasets and the coupling of sequencing with isotopic probing approaches, more accurate and quantitative results are expected to be very promising in the future. Further efforts may still be needed to identify functions of plants and microbial communities influencing the rate of relevant soil processes at the ecosystem level, with a focus on plant performance and anthropogenic disturbances, to identify strategies to control or reshape the rhizosphere microbiome for microbial benefits to efficient nutrient cycling and soil health."
} | 5,806 |
25267881 | null | s2 | 8,234 | {
"abstract": "Electro-active materials are capable of undergoing large deformation when stimulated by an electric field. They can be divided into electronic and ionic electro-active polymers (EAPs) depending on their actuation mechanism based on their composition. We consider electronic EAPs, for which attractive Coulomb forces or local re-orientation of polar groups cause a bulk deformation. Many of these materials exhibit pronounced visco-elastic behavior. Here we show the development and implementation of a constitutive model, which captures the influence of the electric field on the visco-elastic response within a geometrically non-linear finite element framework. The electric field affects not only the equilibrium part of the strain energy function, but also the viscous part. To adopt the familiar additive split of the strain from the small strain setting, we formulate the governing equations in the logarithmic strain space and additively decompose the logarithmic strain into elastic and viscous parts. We show that the incorporation of the electric field in the viscous response significantly alters the relaxation and hysteresis behavior of the model. Our parametric study demonstrates that the model is sensitive to the choice of the electro-viscous coupling parameters. We simulate several actuator structures to illustrate the performance of the method in typical relaxation and creep scenarios. Our model could serve as a design tool for micro-electro-mechanical systems, microfluidic devices, and stimuli-responsive gels such as artificial skin, tactile displays, or artificial muscle."
} | 399 |
30788343 | PMC6379063 | pmc | 8,236 | {
"abstract": "Ecosystems constantly face disturbances which vary in their spatial and temporal features, yet little is known on how these features affect ecosystem recovery and persistence, i.e., ecosystem stability. We address this issue by considering three ecosystem models with different local dynamics, and ask how their stability properties depend on the spatial and temporal properties of disturbances. We measure the spatial dimension of disturbances by their spatial extent while controlling for their overall strength, and their temporal dimension by the average frequency of random disturbance events. Our models show that the return to equilibrium following a disturbance depends strongly on the disturbance’s extent, due to rescue effects mediated by dispersal. We then reveal a direct relation between the temporal variability caused by repeated disturbances and the recovery from an isolated disturbance event. Although this could suggest a trivial dependency of ecosystem response on disturbance frequency, we find that this is true only up to a frequency threshold, which depends on both the disturbance spatial features and the ecosystem dynamics. Beyond this threshold the response changes qualitatively, displaying spatial clusters of disturbed regions, causing an increase in variability, and even a system-wide collapse for ecosystems with alternative stable states. Thus, spanning the spatial dimension of disturbances is a way to probe the underlying dynamics of an ecosystem. Furthermore, considering spatial and temporal dimensions of disturbances in conjunction is necessary to predict ecosystem responses with dramatic ecological consequences, such as regime shifts or population extinction.",
"introduction": "1 Introduction Understanding the stability of ecosystems, i.e., their ability to recover and persist in the face of natural and anthropogenic disturbances, is of fundamental importance to ecology and conservation ( May, 1973 ; Neubert and Caswell, 1997 ; Loreau and de Mazancourt, 2013 ). Ecosystems are spatially extended, comprised of multiple interacting communities in different locations, and therefore an important factor in understanding their stability is their spatial structure ( Levin, 1992 ; Peterson et al., 1998 ; Wang and Loreau, 2016 ). However, while the influence of space on properties such as biodiversity and food web structure has been intensely investigated ( Loreau et al., 2001 ; Chase and Leibold, 2002 ; Montoya and Sol, 2002 ; McCann et al., 2005 ), basic questions regarding spatial stability remain open. In particular, despite the fact that most disturbances (e.g., fires, pest outbreak, pollution runoff) are strongly heterogeneous in space, the impact of their spatial structure on stability is largely unknown. Similarly, their temporal dimension, e.g., their timespan or the frequency of their occurrence, is critical. Taken together, these dimensions span a vast space of possible disturbances that ecosystems can face (e.g., fires and storms). This, in part, explains why reaching a clear understanding of ecosystem stability has proven to be an extremely challenging endeavor. Research on ecosystem stability has a long history in ecology, and numerous studies have investigated how various properties of disturbances affect ecosystem responses. The importance of spatial properties of disturbances, in particular, has been assessed by a few studies of regeneration dynamics under recurrent, spatially structured disturbances ( Turner et al., 1993 ; Moloney and Levin, 1996 ; Fraterrigo and Rusak, 2008 ). These studies introduced the concept of landscape equilibrium and demonstrated how the spatial and temporal scales of disturbances can generate different stability patterns. A point not explicitly addressed in these studies, however, is the importance of rescue dynamics occurring at a regional scale when local recovery processes are too slow or fail altogether. This can occur in sufficiently connected ecosystems, following high-intensity disturbances ( Foster et al., 1998 ; Fraterrigo and Rusak, 2008 ). In fact, recovery from a disturbance is a consequence of both local and regional processes. Local processes lead to recovery due to dynamics that are internal to local communities (e.g., birth and death of individuals), while regional processes lead to recovery by bringing in individuals from neighboring communities via dispersal ( Turner, 1989 ; Leibold et al., 2004 ). These two processes mediate the large-scale system response to a disturbance, and their respective parts in this response is bound to strongly depend on the spatial connectivity of the system and, importantly, on the spatial structure of disturbances. Recent work has made this relationship more explicit, by defining three distinct regimes of recovery from a single spatially heterogenous disturbance: Isolated, Rescue and Mixing ( Zelnik et al., 2018 ). If a system is highly connected due to strong dispersal of organisms, then it is in the Mixing Regime, and the system’s behavior at large scales is essentially an extended version of a local system ( Durrett and Levin, 1994 ). At the other extreme, if dispersal is low and hence each site acts separately with its own local dynamics, then the system is in the Isolated Regime, and its large-scale behavior is an aggregation of many independent small systems ( Tilman et al., 1998 ; Yachi and Loreau, 1999 ). In between these two extremes is the Rescue Regime, where systems with intermediate connectivity show large-scale rescue dynamics due to the interaction between limited dispersal and the system’s behavior at the local scale ( Peterson, 2000 ; Dai et al., 2013 ; Wang et al., 2017 ). For instance, in the study by Dai et al. (2013) , a metapopulation of yeast exhibits a front structure which emerges due to interaction of dispersal with nonlinear local behavior of the yeast. A different example is found in the work of Wang et al. (2017) , where the correlations between local bird populations, mediated by dispersal, leads to a spatial scaling law of the variability of populations across North America. While the spatial structure of both system and disturbance plays no role in the Mixing regime, for weaker dispersal it does: in both the Isolated Regime and the Rescue Regime the spatial structure of the disturbance has significant effects as it can initiate qualitatively different responses that involve both local and regional processes ( Zelnik et al., 2018 ). This is the case in an experimental study of a predator-prey protist system, in which local extinctions are met by rescue processes, which prevent synchronization of the regional metapopulation ( Fox et al., 2017 ). We will therefore consider systems with intermediate dispersal, and focus on the effect of the spatial structure of disturbances as well as their temporal properties. Quantifying the impact of disturbances amounts to defining relevant stability measures. If the disturbance is an isolated event, a natural measure to consider is the return time to the unperturbed state ( May, 1973 ; Neubert and Caswell, 1997 ). On the other hand, in a regime of repeated disturbances (e.g., climatic events), measures of temporal variability are commonly used ( Tilman et al., 2006 ). In the presence of alternative stable states, those repeated disturbances can cause a regime shift from one state to another. One well-known example is that of lake eutrophication ( Carpenter, 2005 ) due to fertilizer runoff disturbances. Here the stability measure of interest is typically persistence, i.e., the probability that a system will remain in a desired state ( Holling, 1973 ; Pimm, 1984 ). Importantly, these stability measures reflect not only the spatial and temporal properties of the disturbance, but also the dynamical features of the perturbed ecosystem. Exploring this interplay is the focus of our study, which we will address by considering three spatial ecosystem models with increasing nonlinear local dynamics, ranging from logistic growth to bistability. Under various perturbation scenarios we will measure their stability using return time, variability and persistence. We begin by looking at the ecosystem’s recovery following a single disturbance, and show that changing the spatial structure of the disturbance reveals two basic recovery trajectories: isolated and rescue. Isolated recovery trajectories reflect the local resilience of the system, while rescue trajectories involve spatial processes, and their dominance signals the failure of local processes. We thus argue that the relationship between spatial structure and recovery contains substantial information about the local dynamics of the system, both close to and far from equilibrium. We continue by exploring the temporal axis of disturbances, and demonstrate a direct link between return time (following an isolated disturbance event) and temporal variability (under a regime of repeated disturbances). We find that for low disturbance frequency patterns of variability do not contain additional information in comparison to the patterns of return time. However, past a frequency threshold (which depends on the system’s internal dynamics) the variability patterns change. As we will argue, this signals the onset of a new dynamical regime driven by disturbances, which can lead to a regime shift—in our case a transition from a populated to a bare state (extinction). Our work demonstrates that the spatial dimension of disturbances can be used to reveal information on the ecosystem’s internal behavior. Furthermore, our results illustrate that the conjunction of the spatial and temporal properties of disturbances may lead to unforeseen dynamical responses, with drastic ecological consequences.",
"discussion": "4 Discussion Investigating the role of the spatial and temporal dimensions of disturbances in ecosystem stability, we obtained four main results: (1) In comparison with a global disturbance, a localized one of the same strength can initiate a fundamentally different, and much slower, ecosystem response, especially when local dynamics are nonlinear. (2) The return time from a single disturbance and the temporal variability caused by repeated disturbances show the same trends, even for locally intense (and therefore nonlinear) disturbances. (3) The relationship between a system’s response and the spatial extent of the disturbances it experiences reveals its underlying dynamics. For instance, a hump-shaped relationship between return time and the spatial extent of the disturbances may indicate bistability. (4) The correspondence between return time and variability breaks down for high disturbance frequencies. This discrepancy signals the occurrence of spatial interactions between disturbed regions, which, in turn, may lead to a regime shift. Although we considered simple spatially homogenous models, our results should apply to a wide range of ecosystems. Forests, savannah and shrublands might be good examples of ecosystems to which our models apply since disturbances such as fires and grazing occur frequently and are often localized, and the recovery of plant communities often follows complex succession dynamics driven by spatial processes ( Adler et al., 2001 ; Turner, 2010 ; Staver and Levin, 2012 ). Our results, however, need not be restricted to such spatially homogeneous systems. Although we built our theory using spatially uniform models, this simplifying feature is not essential to our arguments, which only require a notion of locality. Therefore, our theory may also be relevant to less homogeneous ecosystems, such as mountain lake networks, coral reefs and riverine systems. Indeed, such ecosystems undergo different disturbances that are often strongly localized, and their dynamics may be sufficiently nonlinear ( Knowlton, 1992 ; Campbell Grant et al., 2007 ; Forrest and Arnott, 2007 ). Uniquely to our work, we considered systems locally pushed far from their equilibrium, and even to a different basin of attraction. In a marine ecosystem context, this could represent coral reefs ( Nyström et al., 2000 ; Adjeroud et al., 2009 ) or rocky intertidal systems ( Sousa, 1979 ; Paine and Levin, 1981 ), which frequently undergo intense disturbances (e.g., storm damage). These locally intense disturbances can allow rescue recovery, mediated by dispersal, to dominate the ecosystem response. In the case of the bistable (AE) model this glimpse outside the basin of attraction of the populated state is the direct cause of the hump-shaped trends of variability and return time as a function of disturbance extent. In fact, the front propagation that drives rescue recovery contains information about the ecosystem’s basins of attractions, reflecting the existence of alternative stable states and the transient dynamics between them. Thus, by observing the ecosystem’s response to localized disturbances, rescue recovery allows us to probe ecosystem dynamics far from equilibrium. For instance, comparing between different disturbed marine ecosystems may give further evidence that some have alternative states (e.g., coral reefs) while for others the dynamics show a succession process (e.g., rocky intertidal systems). This reasoning could be taken further by focusing on regions where rescue recovery takes place, e.g., analyzing the plant community structure at transition zones between grassland and forest in a savanna ecosystem ( Augustine, 2003 ). Spanning the spatial dimension of disturbances could thus allow us to detect nonlinearities in ecosystem behavior, revealed by the increasing local intensity of disturbances (see Figure 2 ). One might expect that along the temporal dimension of disturbances, increasing their average frequency could also reveal nonlinear effects, since the ecosystem becomes more strongly disturbed. In fact, increasing frequency has only a trivial linear effect, as reflected by the relation we found between return time and variability (see Figure 3 ). Beyond some threshold, however, a response of a different kind emerges, due to spatial interactions between disturbed regions which aggregate in potentially large-scale clusters. This causes a higher variability than expected and can, consequently, cause a global loss of persistence or a regime shift. Taking, once again, the example of corals reefs, we could ask how the impact of both natural and anthropogenic disturbances leads to a phase-shift from hard coral to fleshy algae dominance. A regime shift due to an aggregation of unrecovered regions would occur not as a typical tipping point due to loss of resilience (e.g., due to changing temperatures), but rather due to the crossing of a threshold for disturbance frequency. Importantly, in such a scenario the two dimensions, spatial and temporal, must be considered in conjunction. The threshold beyond which aggregation occurs depends strongly on the spatial extent of disturbances and hence the associated response is not a mere superposition of responses to single disturbances. In other words, this finding highlights and explains how the interplay between the spatial and temporal dimensions of disturbances can have drastic ecological consequences, such the loss of persistence. Since our findings are purely theoretical, it would be enlightening to elucidate the prevalence of this interplay in empirical systems that have undergone regime shifts (e.g., phase-shifts in coral reefs Nyström et al., 2000 or the desertification of the once green Sahara Ortiz et al., 2000 ). As previously mentioned, in bistable systems the relationship between return time (as well as variability) and the spatial extent of disturbances is hump-shaped. This relation could be used as an indicator of bistability, assessed empirically by comparing time series of the same ecosystem in different regions with estimates of the intensity of single disturbances. Its implications for ecosystem management depend on the type of disturbances considered. Anthropogenic disturbances that are largely controlled, such as logging in forests ( Chazdon, 2003 ) or large-scale fishing ( Kaiser et al., 2006 ), can be better planned to avoid both an unpredictable yield due to high variability and an overall collapse. For many natural disturbances control is neither possible nor desired (e.g., fires in semi-arid ecosystems necessary for plant germination Wellington and Noble, 1985 ), but predicting their effects and the possibility of regime shifts is paramount ( Kéfi et al., 2007 ). In order to focus on the role of the spatial properties of disturbances and allow a clearer presentation, we conducted our analysis assuming disturbances of constant overall strength. It is straightforward to extend the analysis to more general settings, such as a random extent of disturbances and seasonal patterns (see Appendix D for details). It is particularly interesting to consider the case of different values of disturbance strength s . As shown in Figure 5 , if we randomly choose a set of points with different values of strength s and extent σ , we can use these to reconstruct a normalized version of the dependency of the different stability measures on disturbance extent. Thus we can use the different phenomena described previously, such as a hump-shape relationship as an indicator of bistability, under more general conditions, thereby making our theory more empirically accessible. Our work is a step toward a quantitative account of spatial and temporal dimensions of disturbances, and their interplay with local and regional ecosystem dynamics. This is an important goal in the context of global change. Disturbances are of increasing frequencies and occur at different scales (which is evident, e.g., in coral reefs Jackson, 1991 and forests Turner et al., 1993 ), while the spatial structure of ecosystems themselves is altered by land use change, often causing fragmentation of the landscape ( Harrison and Bruna, 1999 ). It is thus important to build a framework in which we can understand and predict the ecological impacts of this complex interplay."
} | 4,530 |
34168253 | PMC8225664 | pmc | 8,237 | {
"abstract": "We propose a new concept that utilizes the difference in Poisson's ratio between component materials as a strengthening mechanism that increases the effectiveness of the sacrificial bond toughening mechanism in macroscale double-network (Macro-DN) materials. These Macro-DN composites consist of a macroscopic skeleton imbedded within a soft elastic matrix. We varied the Poisson's ratio of the reinforcing skeleton by introducing auxetic or honeycomb functional structures that results in Poisson’s ratio mismatch between the skeleton and matrix. During uniaxial tensile experiments, high strength and toughness were achieved due to two events: (1) multiple internal bond fractures of the skeleton (like sacrificial bonds in classic DN gels) and (2) significant, biaxial deformation of the matrix imposed by the functional skeleton. The Macro-DN composite with auxetic skeleton exhibits up to 4.2 times higher stiffness and 4.4 times higher yield force than the sum of the component materials. The significant improvement in mechanical performance is correlated to the large mismatch in Poisson's ratio between component materials, and the enhancement is especially noticeable in the low-stretch regime. The strengthening mechanism reported here based on Poisson's ratio mismatch can be widely used for soft materials regardless of chemical composition and will improve the mechanical properties of elastomer and hydrogel systems.",
"conclusion": "Conclusion In summary, we have demonstrated that incorporating functional skeletons into the Macro-DN composite design improves the resulting mechanical response, especially in the low stretch region. Enhanced stiffness and toughness seen in Macro-DN composites are now known to originate from two sources: (1) the incorporated auxetic or honeycomb structures that exhibit strongly negative or positive Poisson’s ratio mismatch (Δμ), increasing deformation of the matrix prior to skeleton rupture, and (2) the preferential, repetitive rupture of the skeleton within the composite prior to matrix fracture, based on the DN principle. The method introduced here improves upon previous Macro-DN designs that increased toughness solely through the fracture of a sacrificial network. This new method simultaneously increases the work performed by the matrix, by incorporating biaxial deformation. Even if the interfacial adhesion strength between the skeleton and the matrix is poor, it is possible to improve the mechanical properties of composite structures through topological interlocking. This strengthening mechanism can be widely used for diverse combinations of hard skeletons and may serve as a guideline for the design of high strength soft/hard composites in the future.",
"introduction": "Introduction The invention of double-network (DN) hydrogels demonstrated that the incorporation of a sacrificial network can result in gels that possess the toughness required for applied use. DN gels, consisting of a nanoscale interpenetrating hard and brittle “1st network” and soft and ductile “2nd network”, exhibit high Young’s modulus and high toughness despite containing 90 wt.% water 1 – 3 . The toughening of DN gels (known as the DN principle) occurs through the preferential fracture of “sacrificial” covalent bonds within the brittle 1st network over a wide area without causing macroscopic failure of the material 4 . Fracture of the material is delayed by the integration of the 2nd network until substantial widespread damage has occurred within the 1st network, allowing for high levels of energy dissipation. Recent advancements have shown that tough hydrogels can be fabricated not only with covalent sacrificial bonds, but also with weak, reformable non-covalent sacrificial bonds such as ionic and hydrogen bonds 5 – 7 . Furthermore, the DN principle not only applies to hydrogels, but also to industrial materials such as elastomers 8 – 11 . Implementation of the DN principle has the potential to improve the mechanical properties of wide-ranging materials systems. Later, researchers applied the DN concept to macroscopic composites. Inspired by the DN principle, Feng et al . developed the first composites to explicitly demonstrate the “macroscale” DN (Macro-DN) effect, by combining a fabric mesh reinforcement with adhered VHB tape 12 . The stiff fabric mesh plays the role of the 1st network and the soft VHB tape plays the role of the 2nd network, and the composite dissipates energy by breaking the fabric mesh when force is applied, but the VHB tape enables toughness by preventing rupture of the sample. Recently, we developed Macro-DN composites made of a rigid grid skeleton imbedded in soft matrix, and these Macro-DN materials show some common features with nanoscale DN materials 13 , 14 . Similar to the fabric mesh example above, the roles of the stiff skeleton and soft matrix match those of the 1st and 2nd network of DN hydrogels, respectively: the skeleton dissipates energy by rupturing sacrificially, while the matrix maintains extensibility. These studies show that the DN principle can also be applied universally, regardless of size scale 15 . Since mechanical properties of the Macro-DN composites depend much more on material selection and geometric design of the skeleton rather than molecular design, Macro-DN composites can be developed facilely by utilizing 3D printing technology. 3D printing can directly print the reinforcing phase with the desired network geometry to enable bending, rotation, or sacrificial bonds 13 , 16 – 20 . Furthermore, Macro-DN composites can incorporate functionality in this sacrificial network, by utilizing materials such as liquid or low-melting-point alloys 14 , 21 . In our previous research on Macro-DN composites 13 , which utilized a rigid skeleton imbedded in a soft matrix, we focused on understanding the design parameters that control the DN principle on the macroscale. The skeleton was designed in a grid-lattice shape made up of numerous sections, with stiff crossbars delineating each section. The stiff crossbars prevented transverse deformation of the composites when stretched. Rupture of rigid interconnects in multiple sections occurs when the strength ratio between the reinforcing skeleton and matrix approaches 1, enabling high fracture force and large extensibility to show the maximum toughness. In such composites, the matrix of the composite carries almost no stress before breaking the rigid lattice, and the yield force and stiffness of the Macro-DN composite are comparable to that of the individual skeleton. The role of the soft matrix was to maintain global integrity of the composite after breaking of the rigid interconnects, and local deformation only occurs in the regions that have fractured. Furthermore, the maximum strength of the composite was limited to the fracture strength of the skeleton in the low stretch region, and to the matrix at ultimate fracture. The combination of these two points limits the increase in toughness to approximately a factor of two, when compared to a neat matrix sample. In order to achieve higher strength and toughness in Macro-DN composites, it is necessary to introduce a unique macroscopic mechanism in which the deformation of the skeleton and the matrix are strongly coupled even before the rupture of the skeleton. In this case, the skeleton not only dissipates energy through fracture but also acts to apply stress on the matrix to increase the strength of the composite in a synergistic manner. It is known that preferential design of a reinforcing phase in composites with interpenetrating phase structures enable additional plastic deformation, resulting in increased strength and toughness 22 – 24 . This effect has been demonstrated on both the macro- as well as the microscale 25 . Recently, it has been shown that 2D re-entrant honeycomb reinforced structures that exhibit auxetic characteristics under compression improve mechanical performance such as impact resistance when modulus mismatch exists between the two phases 17 – 19 , 26 , 27 . Here, our interest is in soft composite materials that can exhibit high strain at break with ductile characteristics under tensile deformation, along with high stress. To realize this effect, we present a strategy to design skeletons that can exhibit transverse deformations under uniaxial tensile strain prior to rupture as sacrificial bonds. We design sacrificial skeleton networks that can increase or decrease their planar area with deformation, which induces size mismatch with the soft matrix that intends to maintain an almost constant volume during deformation. The correlation between the longitudinal and lateral deformations of materials is characterized by Poisson’s ratio (μ), defined as the negative ratio of lateral contraction strain (ε x ) to the longitudinal extension strain (ε y ) of a material (or structure). The Poisson’s ratio of isovolumetric materials is μ \\documentclass[12pt]{minimal}\n\t\t\t\t\\usepackage{amsmath}\n\t\t\t\t\\usepackage{wasysym} \n\t\t\t\t\\usepackage{amsfonts} \n\t\t\t\t\\usepackage{amssymb} \n\t\t\t\t\\usepackage{amsbsy}\n\t\t\t\t\\usepackage{mathrsfs}\n\t\t\t\t\\usepackage{upgreek}\n\t\t\t\t\\setlength{\\oddsidemargin}{-69pt}\n\t\t\t\t\\begin{document}$$=$$\\end{document} = 0.5. In contrast, planar honeycomb structures have a Poisson’s ratio larger than 0.5, while planar auxetic structures that expand in the lateral direction when being stretched have a negative Poisson’s ratio 28 – 30 . Both honeycomb and auxetic structures provide simple methods to tune the mechanical coupling between the rigid skeleton and soft matrix in Macro-DN composites, which can be characterized by the difference in the Poisson’s ratio, Δμ = μ skeleton –μ matrix . In this work, we fabricated three categories of two-dimensional skeleton structures: auxetic, offset rectangle, and honeycomb, and investigated the mechanical properties of the macroscale planar composites by uniaxial tensile testing. The mismatch in Poisson’s ratio exerts biaxial stress on the soft matrix before the skeleton ruptures, and the mechanical behaviors and the rupture processes of the composites were analyzed. Taking advantage of the planar structure, we also performed real-time birefringence visualization of the stress distribution during uniaxial tension. The stiffness, fracture force, and work of extension were characterized for each composite as a function of the Macro-DN Poisson’s ratio mismatch, Δμ. We see that when the magnitude of Poisson’s ratio mismatch is high, regardless of whether it is negative or positive, the impact of reinforcement significantly increases. Furthermore, we show that these Macro-DN composites exhibit enhanced toughness due to improved matrix deformation by large Poison’s ratio mismatch with the skeleton, rather than just through sacrificial bonds. This mechanism greatly increases low stretch mechanical properties compared to the component materials, resulting in a ~ 320% increase in stiffness, a ~ 340% increase in yield force, and a ~ 200% increase in work of extension for a stretch ratio up to half of the fracture stretch ratio. From these results we propose a new strengthening mechanism for Macro-DN composites via Poisson’s ratio mismatch."
} | 2,795 |
38888209 | PMC11223495 | pmc | 8,239 | {
"abstract": "Coastal zones account for 75% of marine methane emissions,\ndespite\ncovering only 15% of the ocean surface area. In these ecosystems,\nthe tight balance between methane production and oxidation in sediments\nprevents most methane from escaping into seawater. However, anthropogenic\nactivities could disrupt this balance, leading to an increased methane\nescape from coastal sediments. To quantify and unravel potential mechanisms\nunderlying this disruption, we used a suite of biogeochemical and\nmicrobiological analyses to investigate the impact of anthropogenically\ninduced redox shifts on methane cycling in sediments from three sites\nwith contrasting bottom water redox conditions (oxic-hypoxic-euxinic)\nin the eutrophic Stockholm Archipelago. Our results indicate that\nthe methane production potential increased under hypoxia and euxinia,\nwhile anaerobic oxidation of methane was disrupted under euxinia.\nExperimental, genomic, and biogeochemical data suggest that the virtual\ndisappearance of methane-oxidizing archaea at the euxinic site occurred\ndue to sulfide toxicity. This could explain a near 7-fold increase\nin the extent of escape of benthic methane at the euxinic site relative\nto the hypoxic one. In conclusion, these insights reveal how the development\nof euxinia could disrupt the coastal methane biofilter, potentially\nleading to increased methane emissions from coastal zones.",
"introduction": "Introduction Methanogens in marine sediments produce\nup to 85–300 Tg\nof the potent greenhouse gas methane per year, which represents 7–25%\nof global methane production. 1 , 2 However, anaerobic methane-oxidizing\n(ANME) archaea consume more than 90% of the in situ-generated methane. 1 While coastal zones cover only ca. 15% of the\ntotal ocean surface area, they account for more than 75% of global\nmarine methane emissions 3 , 4 as an indirect result\nof high nutrient inputs and burial rates of organic matter. 5 Recent estimates suggest that 5–28 Tg\nof methane per year is emitted from coastal waters to the atmosphere. 6 Eutrophication—excess nutrient input—can\nfurther disrupt the balance between microbial methane production and\nconsumption in the (near) future. For instance, eutrophication can\ncause seawater oxygen depletion due to aerobic microbial respiration\ncoupled to degradation of fresh labile organic matter inputs from\nincreased primary productivity, particularly in enclosed basins with\na shallow water depth. 7 Such oxygen loss\nis anthropogenically induced in coastal ecosystems 5 , 8 and\ncould lead to decreased aerobic methane oxidation in the water column,\nincreasing net methane emissions. 9 , 10 Moreover,\neutrophication alters sediment geochemistry, such as the vertical\ncompression of the typical marine sedimentary redox zonation, 11 that could lead to the release of sulfide into\nbottom waters, 12 hence the development\nof euxinia. Additionally, high inputs of organic matter, following\neutrophication, provide increased substrates for methanogenesis. 5 While the combination of these processes is expected\nto increase methane production in sediments, it remains largely unknown\nhow such processes impact anaerobic methane removal and benthic methane\nrelease into the water column. This makes it urgent to mechanistically\nunderstand coastal sediment methane dynamics in order to build predictive\nbiogeochemical models of future changes, 13 develop appropriate management strategies, and accelerate the pace\nof climate action. ANME archaea are key players in anaerobic\nmethane removal in marine\nsediment ecosystems ranging from coastal zones to the deep sea. 14 − 18 Phylogenetically, ANME forms three major clades: ANME-1, ANME-2,\nand ANME-3, assigned a putative nomenclature at family and genus level, 19 to which we refer in this section. ANME-1, in\nthe order Methanophagales , comprises the family Methanophagaceae with at least 6 genera, and are present\nat a broad range of temperatures, from 2 to 100 °C. 20 Members of this clade have been implicated in\nboth methanogenesis and methane oxidation in estuarine sediments. 21 ANME-2, in the order Methanosarcinales , comprises the family Methanocomedenaceae , with\ntwo genera, Ca. Methanocomedens (ANME-2a) and Ca. Methanomarinus (ANME-2b), the family Methanogasteraceae (ANME-2c), and the family Methanoperedenaceae (ANME-2d).\nMembers of this clade inhabit a narrower range of temperatures (4\nto 20 °C). 20 Finally, ANME-3, also\nin the order Methanosarcinales , comprises the family Methanosarcinaceae with the genus Ca. Methanovorans , and has been reported in colder temperatures (−1 to 17 °C). 20 While ANME-1, ANME-2, and ANME-3 have been implicated\nin sulfate-dependent anaerobic oxidation of methane (S-AOM) in consortia\nwith a syntrophic sulfate-reducing partner, 20 , 22 ANME-2d can independently perform nitrate-dependent, iron-dependent,\nand manganese-dependent anaerobic oxidation of methane (N-AOM, Fe-AOM,\nand Mn-AOM, respectively). 23 − 26 ANME-2a were enriched in Fe-AOM incubations with\nsediments of the North Sea 14 and their\n16S rRNA gene-based abundance correlated to methane and Fe concentrations\nin sediments of the Bothnian Sea. 27 Moreover,\nANME-2a, 2b, 2c, 2d and ANME-3 genomes have genes predicted to encode\nmultiheme c -type cytochromes potentially implicated\nin extracellular electron transfer and Fe reduction, 19 suggesting that multiple ANME groups might perform metal-AOM. The Baltic Sea is highly impacted by eutrophication 8 , 28 and has been proposed as a model marine ecosystem indicative of\nfuture global changes related to anthropogenic impacts such as oxygen\ndepletion and environmental degradation. 29 High methane emissions to the atmosphere from several locations\nin the Baltic Sea have been documented, in the range of 0.1–3.3\nmmol m –2 day –1 , particularly from\ncoastal sites with a shallow sulfate–methane transition zone\n(SMTZ) in the sediment and relatively shallow water depth. 3 , 30 , 31 Similarly, significant methane\nconcentrations in the water column (up to 47 nmol L –1 ), 31 large benthic fluxes of methane (up\nto 2.6 mmol m –2 day –1 ), 32 and high porewater concentrations of methane\n(6 mM) 32 have been reported in the Baltic\nSea. Previous studies indicate that ANME are key players in methane\ncycling in Baltic Sea sediments, with ANME-1 and ANME-2 accounting\nfor S-AOM activity, ANME-2 potentially involved in Fe-AOM, and ANME-1\nalso implicated in methanogenesis. 27 , 33 − 35 However, a mechanistic understanding of the environmental and biological\nfactors that impact AOM in the Baltic Sea and other coastal ecosystems\nremains elusive. Here, we investigated sediments of the eutrophic\nStockholm Archipelago, 8 , 36 , 37 pursuing a better mechanistic\nunderstanding of the impacts of divergent bottom water redox conditions\non microbial methane cycling and associated sediment biogeochemistry.\nWe selected three sites with contrasting bottom water oxygen concentrations\n(oxic: [O 2 ] aq > 63 μmol L –1 ; seasonally hypoxic: [O 2 ] aq < 63 μmol\nL –1 ; euxinic: [O 2 ] aq = 0 μmol\nL –1 with free sulfide). Sediment cores taken at\nthese sites were subjected to high-resolution geochemical characterization,\n16S rRNA gene profiling, potential methane production rate measurements,\nmetagenomic analyses, and AOM rate measurements in selected sediments\nincubated with different sulfide concentrations. Our study specifically\naimed to identify the main controls on the abundance, distribution,\nand activity of ANME archaea and to elucidate the impacts of differing\nbottom water redox conditions on methane cycling in these eutrophic\ncoastal sediments.",
"discussion": "Results and Discussion Sulfate-Dependent Anaerobic Oxidation of Methane Is the Dominating\nProcess for Benthic Methane Removal At the time of sampling,\nwater column characteristics at the three study sites were coherent\nwith historical environmental monitoring data by SMHI ( Tables 1 and 2 , and Figure S1 ). 33 Oxygen penetration into the sediment followed the trend in ambient\nbottom water redox conditions, with the deepest oxygen penetration\nat Site 3 and no oxygen penetration at Site 7 ( Table S3 ). In fact, for the latter, sulfide (10 μmol\nL –1 ) was detected in the bottom water ( Tables 2 and S3 ). Macrofauna (polychaetes and bivalves) were\nobserved only at Site 3. While bottom water pH, salinity, and temperature\nwere similar across sites, the different bottom water redox conditions\nwere reflected in full profile-averaged sedimentary TOC contents,\nwith the highest TOC observed at the euxinic site, Site 7 ( Tables 2 and S3 ). However, taken together, the three sites\nare geochemically rather similar and comparable to other sites in\nthe Stockholm Archipelago that were previously geochemically characterized,\ni.e., organic-rich sediments, sulfide accumulation in the porewaters,\nand a shallow SMTZ. 32 , 36 The observed shallow SMTZ\nis a common feature of eutrophic coastal sediments. 5 Porewater sulfate decreased with depth at all sites ( Figure 2 and Table S3 ). At Site 3, it took ∼30 cm below\nthe seafloor (cmbsf) before sulfate was depleted, whereas at Sites\n5 and 7, this removal occurred already around 10 cmbsf ( Figure 2 ). Methane concentrations increased\nslightly with depth to a maximum of ca. 100 μmol L –1 at Site 3. At Sites 5 and 7, in contrast, concentrations of methane\nincreased strongly with depth, reaching a value of ∼2 mmol\nL –1 . Porewater sulfide was only present at low concentrations\n(<200 μmol L –1 ) and in a confined zone\n(10–30 cmbsf) at Site 3. At Sites 5 and 7, however, sulfide\nconcentrations increased rapidly with depth, with the strongest increase\nand highest concentrations (>1 mmol L –1 ) at Site\n7. Additionally, sediments in Site 7 had a higher approximated annual\nsulfide exposure (0.88 mmol year –1 ) in comparison\nto Site 5 (0.39 mmol year –1 ) and Site 3 (0.08 mmol\nyear –1 ) ( Table S3 ). The\nshallow SMTZ is the combined result of low salinity, hence low sulfate\nconcentrations and high rates of organic matter deposition and degradation,\nculminating in a vertical compression of the redox zonation. 61 The somewhat deeper SMTZ at Site 3 can be explained\nby its ambient redox conditions. Oxygen is perennially available throughout\nthe water column, leading to more aerobic degradation of organic matter\nin the water column and surface sediments, hence the extended depth\nof the depletion of alternative electron acceptors, such as sulfate\n( Figure 2 ). Figure 2 Porewater depth\nprofiles of sulfate (SO 4 2– ), methane\n(CH 4 ), sulfide (H 2 S), dissolved\niron (Fe 2+ ) and manganese (Mn 2+/3+ ), ammonium\n(NH 4 + ), and the sum of nitrate and nitrite (NO x ) at our study sites in the Stockholm Archipelago:\nSite 3 (Sandofjärden), Site 5 (Lilla Värtan), and Site\n7 (Skurusundet). The arrow to the left indicates decreasing bottom\nwater (BW) oxygen concentrations from Site 3 to 7. Cmbsf; centimeters\nbelow the seafloor. At Sites 3 and 5, a peak in dissolved Fe was observed\ndirectly\nbelow the sediment–water interface, followed by a rapid decrease\nto values of around zero. At Site 3, dissolved Fe concentrations increased\nagain when sulfide was depleted at a depth. Dissolved Fe was nearly\nabsent at Site 7. A peak in dissolved Mn was observed near the sediment–water\ninterface at all three sites, with maximum concentrations decreasing\nwith increasingly reducing sediments from Site 3 to Site 5 to Site\n7. At Site 3, dissolved Mn in the porewater increased again at a depth\nafter a subsurface minimum. Ammonium increased with depth at all sites,\nwith the highest concentrations (up to 2.5 mmol L –1 ) and most rapid increase at Sites 5 and 7. NO x (nitrate and nitrite) concentrations ranged from 2.5 to 12.5\nμmol L –1 in the bottom waters and decreased\nrapidly with depth in the sediment at all sites ( Figure 2 and Table S3 ). The calculated downward flux of sulfate into the\nSMTZ ( Table 3 ) was\nhighest at Sites\n5 and 7 (1.5 and 1.3 mmol m –2 d –1 ), but still substantial (0.9 mmol m –2 d –1 ) at Site 3. The calculated upward flux of methane into the SMTZ\nat Site 3 was nearly absent, and around 0.5 mmol m –2 d –1 at Sites 5 and 7, which should be regarded\nas an absolute minimum estimate due to degassing of methane during\nsampling. 43 , 44 There was no benthic methane efflux at Site\n3. The benthic methane efflux was about seven times larger at Site\n7 relative to Site 5 (∼1 vs 0.15 mmol m –2 d –1 , respectively). Table 3 Diffusive Fluxes of Sulfate and Methane\n(mmol m –2 day –1 ) as Calculated\nfrom Porewater Profiles (Intervals in Parentheses) and, for the Benthic\nFlux, from Porewater and Bottom Water Concentrations (See Text) a site downward\nsulfate flux into SMTZ upward methane\nflux into SMTZ b benthic methane\nefflux site 3 (Sandöfjarden) 0.88 (9.5–19 cm) 0.08 (29–34 cm) 0 site 5 (Lilla Värtan) 1.49 (1.75–7.5 cm) 0.54 (8–13 cm) 0.15 site 7 (Skurusundet) 1.27 (2.5–4.5 cm) 0.50 (0–15 cm) 1.02 a SMTZ; sulfate–methane transition\nzone. b Methane fluxes into\nthe SMTZ at\nSites 5 and 7 are likely underestimated because of degassing during\nsampling. In sediments of eutrophic, low-oxygen coastal systems,\nS-AOM is\nexpected to dominate methane removal. 1 Sulfate\nis presumably also the major terminal electron acceptor for AOM by\nANME-2 archaea in the investigated sediments of the Stockholm Archipelago\nat all three sites. Estimated diffusive fluxes of sulfate and methane\ninto the SMTZ ( Table 3 ) suggest that S-AOM accounts for at least 40% of the observed sulfate\nreduction. Given the potential degassing of methane, especially for\nsediment intervals with high methane concentrations, 44 in situ S-AOM rates are likely higher. The differences\nin calculated benthic effluxes of methane between our three sites\nsuggest that the removal of methane becomes less effective with more\nreducing bottom water redox conditions ( Table 3 ). To investigate if alternative terminal\nelectron acceptors for AOM\nsuch as Fe and Mn oxides were available, we determined the concentration\nand reactivity of the Fe and Mn oxides present in the sediment. The\nsequential extractions dissolved between 70 and 80% of the total sediment\nFe at the three sites ( Figure 3 and Table S3 ). Except for the\nsurface sediment at Site 3, the poorly ordered Fe(III)-oxides (e.g.,\nferrihydrite and lepidocrocite) content was generally low. Crystalline\nFe(III)-oxides (e.g., goethite and hematite) accounted for a substantial\nfraction of the extractable Fe at all sites (15–30%). The fraction\nconsisting of a mixture of Fe(II)-carbonates and monosulfides, merely\nconsisting of Fe(II)-monosulfides in this setting with abundant porewater\nsulfide, 36 was the dominant Fe fraction\nat depth in the sediment at Site 3. At Sites 5 and 7, this fraction,\nas well as Fe(II)-pyrite, accounted for the majority of the extractable\nFe at depth. Figure 3 Solid-phase iron and manganese speciation (μmol\ng –1 dry sediment) depth profiles for the three study\nsites. The arrow\nat the bottom indicates decreasing bottom water (BW) oxygen concentrations\nfrom Site 3 to 7. cmbsf; centimeters below the seafloor. About 50 to 60% of the total sedimentary Mn was\nextracted in the\nanalyzed steps, i.e., steps 1, 2, and 5 (see the Supporting Information section on Solid-Phase Analysis), at\nall three sites. At Site 3, poorly ordered Mn(III/IV)-oxides (e.g.,\nbirnessite and pyrolusite) dominated the extracted Mn pool with only\na minor (<10%) contribution of pyrite-associated Mn. At Site 5,\npoorly ordered Mn(III/IV)-oxides were never dominant, and concentrations\nwere low at depth. Sedimentary Mn could generally be divided into\ntwo relatively equal fractions of Mn(II)-carbonates (rhodochrosite)\nand pyrite-associated Mn. At Site 7, pyrite-associated Mn dominated\nthe extracted Mn pool with only a minor (<10%) contribution of\npoorly ordered Mn(III/IV)-oxides ( Figure 3 and Table S3 ). In coastal systems, electron acceptors other than sulfate may drive\nAOM, such as nitrate and nitrite 23 , 60 as well as\npoorly ordered Fe(III)- and Mn(III/IV)-oxides. 14 , 61 , 62 At our sites, nitrate and nitrite are exclusively\npresent in low concentrations in the surface sediment ( Figure 2 ) and are therefore unlikely\nto substantially contribute to AOM activity. At Sites 5 and 7, poorly\nordered Fe(III)- and Mn(III/IV)-oxides are nearly absent ( Figure 3 ). Additionally,\nat both these sites, most of the reactive Fe and Mn is sulfidized\n( Figure 3 ), in line\nwith the ambient bottom water redox conditions and relatively high\nporewater sulfide concentrations ( Table 2 and Figure 2 ), previously also observed for other sites in the\nStockholm Archipelago. 36 Hence, there is\nonly limited potential for Fe- and Mn-AOM, which seem to play a larger\nrole in oligotrophic rather than eutrophic coastal ecosystems. 63 − 65 However, a role for Fe-AOM cannot be fully excluded, as crystalline\nFe(III)-oxides and even recalcitrant Fe(II/III)-oxides ( Figure 3 ) may also play a role in Fe-AOM 66 and S-AOM. 67 Magnetite,\nfor example, was shown to stimulate Fe-AOM activity and ANME-2a enrichment\nin incubations with North Sea sediments, 14 and goethite-dependent AOM has been suggested as a significant methane\nsink in paddy soils, in which hematite and magnetite-AOM were also\ndetected. 68 High Methane Production Potential in the Hypoxic and Euxinic\nSites Sites 5 and 7 showed particularly high potential methane\nproduction rates (up to 2.3 ± 0.3 and 3.3 ± 0.4 μmol\nmethane g –1 d –1 respectively at\n2 cm depth). By contrast, at Site 3, potential methane production\nrates did not exceed 0.22 ± 0.006 μmol methane g –1 d –1 at 18 cm ( Figure 4 and Table S1 ).\nTo examine the microbial diversity and metabolic potential, sediments\nwere subjected to DNA extractions and high-resolution 16S rRNA gene\nsequencing, with selected samples also used for metagenomic sequencing.\nArchaeal 16S rRNA gene sequences were used to generate amplicon sequence\nvariants (ASVs), which were clustered at the family level for relative\nabundance visualization ( Figure 4 ). Methanoregulaceae and Methanosaetaceae represent the two most abundant putative methanogenic families.\nWhile Methanoregulaceae had the highest relative\nabundances of 16% in Site 3 at 34 cm, 38% in Site 5 at 50 cm, and\n34% in Site 7 at 30 cm, Methanosaetaceae reached\n11% in Site 3 at 19 cm, 20% in Site 5 at 42 cm, and 24% in Site 7\nat 38 cm. Other identified putative methanogenic families within Euryarchaeota included Methanosarcinaceae , Methanofastidiosaceae , Methanomicrobiaceae , Methanospirillaceae , Methanobacteriaceae , and Methanocorpusculaceae , poorly resolved Methanocellales and Methanomicrobiales families, Methanocellaceae , and Methanothermobacteriaceae , and the putative methanotroph family Methanoperedenaceae . Within the phylum Verstraetearchaeota , the putative\nmethanogenic families Methanomethyliaceae and Methanomethylophilaceae were identified, and, within the\nphylum Thermoplasmatota , Methanomassiliicoccaceae . An ANME-2a-2b family had the highest relative abundances, among\narchaeal sequences, of 32% at 2.5 cm depth at Site 3, 12% at 26 cm\nin Site 5, and 1% at 50 cm depth in Site 7. The putative ammonium-oxidizing Crenarchaeota family Nitrosopumilaceae reached\n70% relative abundance in Site 3 at 1.25 cm and was minor in the other\ntwo sites. Of 7192 archaeal ASVs, 71% could not be classified at the\nfamily level, and their summed relative abundances varied between\n22 and 86%. However, only 6% of archaeal ASVs could not be classified\nat the phylum level. The large proportion of unclassified archaeal\nASVs at the family level could indicate taxonomic novelty or, alternatively,\noverestimation of biodiversity, a common limitation of 16S-based studies. 69 Figure 4 Abundance, distribution, and activity of key microbial\ngroups in\nsediments from the three study sites presented with selected geochemical\ndata. Relative abundances (%) of archaeal families, based on 16S rRNA\ngene sequencing, are color coded according to the legend to the left\nand match depths as in other panels. The designation “fam”\nindicates a poorly resolved family. The thickness of bars corresponds\nto the depth resolution indicated in Materials and\nMethods . The normalized genome coverage (ngCOV) of metagenome-assembled\ngenomes (MAGs) representing methanotrophs (one genome) and methanogens\n(four genomes) is displayed next, followed by methane production rates\nin μmol of methane wet sediment g –1 d –1 as measured in triplicate methanogenic incubations,\nin which error bars are generally smaller than black circles. Sulfate\nand methane porewater concentrations are expressed in mmol L –1 . The arrow to the right indicates decreasing bottom water (BW) oxygen\nconcentrations from Site 3 to 7. cmbsf; centimeters below the seafloor. In total, 144 metagenome-assembled genomes (MAGs)\nof high (>90%\ncomplete, <5% contaminated) and medium quality (>50% complete,\n<10% contaminated) were reconstructed from 12 coassembled samples\nand screened for methane metabolism marker genes. No particulate or\nsoluble methane monooxygenase-encoding genes, which are diagnostic\nof aerobic methane oxidation potential, were found in binned and unbinned\ncontigs. Five genes encoding a methyl-coenzyme M reductase alpha subunit\n( mcrA ) related to methane production were identified\nin the following four genomes. MAG 009 Methanoregulaceae had potential\nfor hydrogenotrophic methanogenesis, and MAG 010 Methanosarcinaceae\nhad potential for methanogenesis from H 2 and CO 2 , formate, acetate ( acss ), H 2 and methanol\n( mtaA ), and H 2 and mono-( mtmBC ) and trimethylamine ( mttC ), but contained two mcrA genes. While MAG 015 Methanomassiliicoccales had potential\nfor methanogenesis from H 2 and methanol ( mtaA ), MAG 016 Methanomassiliicoccales had potential for methanogenesis\nfrom H 2 and methanol ( mtaA ), di( mtbC ), and trimethylamine ( mttC ). We also\nidentified an unbinned mcrA sequence with a best\nblastp hit to Candidatus Methanofastidiosum methylthiophilus\n(KYC53403.1 NCBI accession number), which has been proposed to perform\nmethanogenesis from methanethiol, dimethylsulfide, 3-methylmercaptopropionate,\nand 3-mercaptopropionate. 70 We used\nNMR values of genes and normalized genome coverage (ngCOV)\nof MAGs as a qualitative proxy for microbial abundances with the sole\npurpose of comparisons between samples in this study and interpret\nthese data strictly in the context of sediment biogeochemistry and\nmicrobial activity rates. NMR values of mcrA genes\nindicated that MAG 010 Methanosarcinaceae, MAG 015 Methanomassiliicoccales,\nand MAG 009 Methanoregulaceae accounted for the most significant methanogen mcrA NMR changes across sites ( Figure S2 ), with highest NMR values in Site 5, then Site 7, and lowest\nin Site 3. The summed normalized genome coverage (ngCOV) of four methanogen\nMAGs was highest at Site 5 (3.6× at 34 cm), followed by Site\n3 (3× at 22 cm) and Site 7 (2.6× at 22 cm, Figure 4 ). Putative methanogen\nabundances (hypothesized from MAG coverages\nand 16S rRNA gene-based relative abundances) and potential methane\nproduction rates have contrasting profiles and do not positively correlate\n( Figure 4 ). An explanation\nfor these results is that sequencing data do not reflect a potentially\nhigher methanogen biomass or activity in surface sediments (0–10\ncm). Alternatively, larger pools of labile organic substrates generated\nfrom organic matter degradation could be available in surface sediments\n(as observed in other aquatic ecosystems 71 ), which are depleted at depth, resulting in decreasing potential\nmethane production rates in deeper sediment layers. This might become\napparent at the gene expression level, which could correlate with\nmethane production activity, which is potentially detectable in future\nmetatranscriptomic studies. In Site 3, where sulfate was detected\nuntil ca. 30 cm, sulfate reduction-driven competitive inhibition of\nmethanogenesis was expected, 72 and low\npotential methane production rates at this site indicate that this\nexpectation was fulfilled despite the relatively high MAG coverages\nand 16S-based relative abundances of methanogens. In Sites 5 and 7,\nhigh potential methane production rates in surface sediments might\nalso reflect larger pools of labile organic carbon and the rapid depletion\nof sulfate. These are conditions that could favor methanogens at potentially\nlower abundances in top sediments to be more active than in deeper\nsediments, where they could be more abundant but have less substrate\navailability. Additionally, the observation that the highest potential\nmethane production rates were measured in surface sediments concomitant\nwith relatively high sulfate concentrations suggests cryptic methane\ncycling, in which methane is consumed as soon as it is produced within\nthe SMTZ, detectable via radiotracer or stable isotope studies. 73 Methanogenesis from noncompetitive substrates\nis supported by our data since genomic potential for methanol and\nmethylamine-driven methanogenesis was identified in three out of four\nmethanogen MAGs. This could partially fuel AOM in Site 5, where the\nANME-2 MAG coverage was significant (3.85×) within the SMTZ.\nMethylotrophic methanogenesis has been previously implicated in cryptic\nmethane cycling in similar ecosystems. 74 , 75 Alternatively,\ncompetitive methanogenesis could coexist with sulfate reduction, as\npreviously observed in estuarine systems, potentially due to the abundance\nof substrates. 76 This is also supported\nby our data, given the relative abundances of sequences affiliated\nwith Methanoregulaceae and Methanosaetaceae . Anaerobic Methane-Oxidizing Archaeon Constitutes the Benthic\nMethane Biofilter Based on phylogenetic analyses, MAG 011,\naffiliated with the archaeal order Methanosarcinales , was identified as representative of an ANME-2 archaeon ( Figure S3 ). This was the only detected genome\nrepresenting a methane-oxidizing microorganism in our sequencing data\nsets, indicating that the benthic methane biofilter (the biological\nprocess of methane removal in sediments) was potentially constituted\nof a single organism. MAG 11 had the highest ngCOV at Site 5 (9.5×\nat 22 cm) below the SMTZ ( Figure 4 ), but reached only 0.3× at Site 3 at 2 cm and\n0.06× at Site 7 at 2 cm. This genome had 66% average amino acid\nidentity to genome MZXQ01 (NCBI accession number), which represents\nan ANME-2b archaeon obtained from sediment of the Hydrate Ridge North\nmethane seep 77 classified as Ca . Methanomarinus sp. nov. 1 19 ( Figure S4 ). MAG 011 was 98.4% complete\nand 2% contaminated and had potential for a full reverse methanogenesis\npathway as well as acetate production or assimilation via an acetyl-CoA\nsynthase ( acss ) gene ( Figure 5 and Table S4 ).\nMAG 011 was further analyzed in detail to elucidate its metabolic\npotential. Genes encoding proteins involved in the reverse methanogenesis\npathway were mostly present as single copy ( Table S4 ), with a few exceptions: (i) hdrABC were\npresent in three copies, (ii) a second copy of mtrAH was downstream of mtrX , while mtrEDCBAFGH subunits were present in a separate contig, (iii) three copies of\nthe gene encoding the formylmethanofuran-tetrahydromethanopterin N-formyltransferase\n( ftr ) were present, and (iv) both molybdenum- and\ntungsten-dependent formylmethanofuran dehydrogenases were present\n( fmdCABDE and fwdGBDC ), with fwdC present as a separate single subunit and also as a\ntwo-copy fwdC gene fusion ( Table S4 ). Figure 5 Metabolic reconstruction of MAG 011 ANME-2 based on loci specified\nin Table S3 . The gray area represents the\npseudoperiplasm, and the outermost circle represents the S-layer.\nNumbers that follow # indicate the number of heme-binding motifs.\nAbbreviations are as follows: F 420 , coenzyme F 420 ; Fd, ferredoxin; CoM, coenzyme M; CoB, coenzyme B; MP, methanophenazine;\nHdrABC, cytoplasmic heterodisulfide reductase; Frh, F 420 -reducing hydrogenase; MvhD, methyl viologen-reducing hydrogenase\nsubunit D; HdrDE, periplasmic heterodisulfide reductase; Mcr, methyl-coenzyme\nM reductase; H 4 MPT, tetrahydromethanopterin, Mtr, formylmethanofuran-H 4 MPT N-formyltransferase; Mer, F 420 -dependent methylene-H 4 MPT reductase; Mtd, F 420 -dependent methylene H 4 MPT dehydrogenase; Mch, Methenyl-H 4 MPT cyclohydrolase;\nFtr, formylmethanofuran: H 4 MPT formyltransferase; Fmd,\nmolybdenum-dependent formylmethanofuran dehydrogenase; Fwd, tungsten-dependent\nformylmethanofuran dehydrogenase; Cyt, cytochrome; RNF, Rhodobacter nitrogen fixation complex; Fpo, F 420 :methanophenazine\noxidoreductase; S-LP, S-layer protein; OmcZ, outer membrane cytochrome;\nFsrN 2 C, F 420 -dependent sulfite reductase; FrhB-FqoF,\nhypothetical F 420 - and Fd-oxidizing electron-confurcating\nhydrogenase; CODH/ACS, carbon monoxide dehydrogenase/acetyl-CoA synthase\ncomplex EC: 1.2.7.4/2.3.1.169); acss, acetyl-CoA synthetase (EC: 6.2.1.1);\nTCA, tricarboxylic acid cycle; SDH, succinate dehydrogenase. Nine candidate genes encoding electron-carrying\nferredoxins were\nidentified, as well as several FrhB/FdhB/FpoF paralogues ( Table S4 ). A single subunit fpoF could be part of the F 420 H 2 dehydrogenase\ncomplex fpoDCBAONMLKJ1J2IH , and an frhB - fqoF (K00441, K22162) gene fusion could encode\na protein to couple ferredoxin oxidation to F 420 H 2 production, as potential Fpo/Fqo-dependent ferredoxin oxidation\npathways. 19 Moreover, three other genes\nwith homology to FrhB of Candidatus Methanoperedens\nnitroreducens strain BLZ1 78 were found:\n(i) the first as a single subunit, (ii) the second immediately upstream\nof mvhD , hdrA2 , another mvhD , then hdrABCC , and (iii) the third\nas 2x frhB - fsrC fusion (K00441–K00441–K21816).\nThe genes fsrNC encode an F 420 -dependent\nsulfite reductase in Methanocaldococcus jannaschii ( 79 ) (EC: 1.8.98.3), which detoxifies sulfite\nwhile reducing it to sulfide. Furthermore, the succinate dehydrogenase\nmembrane subunits sdhCD were absent, while sdhAB were present and sdhB was fused with tfrB (K00239, K00240–K18210), which encodes the CoM/CoB-fumarate\nreductase subunit B (EC: 1.3.4.1) characterized in Methanobacterium\nthermoautotrophicum strain Marburg. 80 In M. thermoautotrophicum , TfrA harbors FAD-binding\nmotifs and the catalytic site for fumarate reduction, while TfrB harbors\none [2Fe–2S] cluster, two [4Fe–4S] clusters, and the\ncatalytic site for CoM–S–H and CoB–S–H\noxidation. Therefore, we hypothesize that in ANME-2 represented by\nMAG 011, electrons from succinate oxidation could be used to generate\nheterodisulfide for the first step in methane oxidation instead of\nflowing to the electron transport chain. Finally, a complete Rnf complex,\ninvolved in ferredoxin recycling and proton gradient generation in\nANME, 81 was identified, as well as a downstream\nc 7 family octaheme cytochrome as previously reported in\nANME-2. 82 Other cytochromes were also identified\nin this genome: three c 3 -family cytochromes containing\n11 or 12 heme-binding motifs, an S-layer protein containing 21 heme-binding\nmotifs, and two FeGenie-identified outer membrane cytochrome omcZ sequences with 7 and 8 heme-binding motifs ( Table S4 ), which could mediate extracellular\nelectron transfer to a syntrophic partner or metallic terminal electron\nacceptor. Sulfide Toxicity Hinders Methane Removal by ANME Based\non the highest ngCOV of ANME-2 MAG 011, we selected sediments from\nSite 5 at the depth interval of 8–28 cm to experimentally test\nthe hypothesis that sulfide inhibits AOM activity in these sediments\n( Figure 6 and Table S2 ). We preincubated sediments with 13 C-methane until AOM activity was detected. Then, we added\n0, 0.5, 1, 2, or 4 mM sulfide to triplicate incubations and monitored\nAOM rates as well as final sulfide concentrations. Additionally, we\nincluded one control to which we added 2 mM sulfide at the beginning\nof the preincubation. Average AOM rates were highest when no sulfide\nwas added (31.3 nmol of methane g dry sediment –1 day –1 ), decreasing with increasing sulfide concentrations:\n23.2, 16.9, 14.5, and 12.1 nmol of oxidized methane g dry sediment –1 day –1 when, respectively, 0.5,\n1, 2, and 4 mM of sulfide were added to incubations ( Figure 6 and Table S2 ). The average AOM rate in the control incubation that received\n2 mM of sulfide at the beginning of the preincubation was 10.5 nmol\nof oxidized methane g dry sediment –1 day –1 . This resulted in an average of 3.62 mM sulfide at the end of the\nexperiment and inhibition of 67% of AOM activity relative to incubations\nthat received no sulfide, which had an average of 0.55 mM of sulfide\nat the end of the experiment. In preincubated bottles with no added\nsulfide, additions of 0.5, 1, 2, and 4 mM of sulfide resulted, respectively,\nin average final sulfide concentrations of 0.8, 1.12, 1.88, and 3.41\nmM, and inhibition of 26, 46, 54 and 61% of AOM activity, respectively\n( Figure 6 and Table S2 ). Figure 6 Average AOM rates measured in Site 5 sediments\nincubated with varying\nadded sulfide concentrations (0, 0.5, 1, 2, or 4 mM). While black\ncircles indicate that samples were preincubated with 13 C-methane only, the gray circle indicates that samples were preincubated\nwith 13 C-methane and 2 mM of sulfide. Error bars provide\nthe standard deviations across three biological replicates. Results from this experiment corroborate our key\nmicrobiological\nobservation, the differential distribution of potential ANME-2 abundances\nacross samples based on MAG 11 normalized genome coverages. Low coverages\nof MAG 11 were estimated in sediments of Site 3, in agreement with\nthe low methane concentrations measured in situ and low potential\nmethane production rates measured ( Figure 4 ). At Site 5, MAG 11 coverages were the highest,\nwithin and below the SMTZ, matching abundant methane and sulfate substrates\nfor S-AOM activity ( Figure 4 ) and their calculated fluxes into the SMTZ ( Table 3 ) as well as the highest potential\nmethane production rates. However, at Site 7, where methane and sulfate\nwere also abundant and had similar high fluxes into the SMTZ ( Table 3 ), putative ANME-2\nabundances were near zero, based on both 16S analyses and MAG 11 coverages. We acknowledge that shallowing of the SMTZ and methanogenesis in\nsurface sediments likely contributed to higher methane escape from\nSites 5 and 7 relative to Site 3. 5 Such\nhigh benthic fluxes of methane are in accordance with previously reported\nvalues for the Stockholm Archipelago. 32 However, differences in the SMTZ or in methanogenesis in surface\nsediments could not account for the 7-fold higher benthic methane\nefflux in Site 7 relative to Site 5, given that these two sites had\nnearly identical depths and thicknesses of the SMTZ (0–10 cm)\nand similar potential methane production rates in top sediments ( Figure 4 ). Instead,\nour experimental results in combination with the virtual\ndisappearance of ANME-2 sequences at Site 7, but their persistence\nat Site 5, suggests that the disruption of the methane biofilter likely\naccounted for this 7-fold difference. We infer that the putative distribution\nof ANME-2 at our study sites is most likely linked to high sulfide\nconcentrations (0.5–1.2 mmol L –1 ) and especially\nthe higher annual exposure to sulfide (0.88 mmol year –1 ) at Site 7 compared with that at Site 5 (0.39 mmol year –1 ), which could have directly caused sulfide toxicity in ANME-2 cells.\nWe note that the relationship between sulfide and AOM rates is nonlinear\n( Figure 6 ), which could\nexplain why the difference in sulfide exposure between the two sites\n(∼2-fold) does not mathematically account for the difference\nin benthic methane efflux (7-fold) between the two sites. In addition,\nwe cannot rule out an effect of other (abiotic) factors such as differences\nin the type of organic matter and sediment composition in general\nbetween the two sites. These results agree with an early study on\nenvironmental controls on ANME abundances, which found that sulfide\nconcentrations negatively correlated with 16S-based ANME-2 abundances\nwhile positively correlated with ANME-1 abundances, potentially due\nto sulfide-oxidizing bacteria co-occurring with ANME-1 but not with\nANME-2. 83 Additionally, sulfide could hamper\nthe enzymatic activity of the F 420 -dependent sulfite reductase\n(Fsr) via product inhibition, leading to sulfite toxicity, as well.\nSulfite is known to inhibit methanogenesis, 84 and the Fsr enzyme, first described in M. jannaschii , detoxifies sulfite by reducing it to sulfide, which can then be\nassimilated. While sulfite toxicity generally occurs due to its intracellular\nreaction with proteins and sulfhydryl groups, 79 , 85 in methanogens specifically, sulfite reacts in vitro with and inactivates\nthe key methanogenic enzyme, methyl-coenzyme M reductase. 86 , 87 The recent crystal structure of Fsr from Methanothermococcus\nthermolithotrophicus revealed Fsr as the simplest\nsulfite reductase crystallized so far, with similar traits to assimilatory\nsulfite reductases, having, interestingly, a higher preference for\nnitrite (apparent Km of 2.5 μM) over sulfite (apparent Km of\n15.6 μM). 88 F 420 -dependent sulfite reductases have been found to\nbe highly abundant sulfur metabolism proteins in an ANME-2 metaproteomics\nstudy, 77 which also reported the inhibition\nof AOM activity by ANME-2a/2c in methane seep sediments incubated\nwith 1 mmol L –1 sulfite, 1 mmol L –1 polythionate, and 0.25 mmol L –1 polysulfide. The\nauthors concluded that the role of F 420 -dependent sulfite\nreductases in ANME-2 is more likely sulfite detoxification rather\nthan sulfur assimilation, which could occur via several other ANME-2\nenzymes. 77 Furthermore, while methanogens\nare known to withstand 3–5 mmol L –1 sulfide\nlevels and assimilate sulfur from sulfide, 89 ANME archaea have been reported to be inhibited by 3–4 mmol\nL –1 sulfide under the low sulfate concentrations\nthat we measured in our study (4 mmol L –1 range),\nbut not under high (21 mmol L –1 ) sulfate concentrations. 90 Our experimental results ( Figure 6 ) provide further evidence for a dose-dependent,\nsulfide-driven inhibition of AOM activity under ∼4 mmol of\nL –1 sulfate. MAG 11 ANME-2 had several genes\nencoding multiheme c -type cytochromes ( Figure 5 ) including S-layer proteins,\nlowly expressed in Fe-AOM-performing\nANME-2d, 25 and OmcZ-like proteins suggested\nas the ANME mechanism for extracellular electron transfer, 19 which could be used for electron transfer to\na sulfate-reducing partner or to metal oxides. These multiheme c -type cytochromes identified in ANME might be targets of\nsulfite toxicity. 91 These previous\nstudies suggest that a threshold sulfide concentration,\npotentially dependent on sulfate concentrations, might exert thermodynamic\nand toxicity controls on AOM activity. Our results match these previous\nobservations but also indicate that sulfide exposure (total sulfide\nin mmol year –1 ), which differed more between Sites\n5 and 7 than sulfide concentrations (mmol L –1 at\nthe time of sampling), could play a role in the lower putative ANME-2\nabundances at Site 7 ( Figure 4 ). Overall, these results support our experimental, biogeochemical\nand metagenomic evidence for the proposed mechanism of sulfide toxicity\nas a key control on putative ANME-2 abundances and activity in coastal\nsediments, suggesting that the expansion of euxinia in coastal areas 92 might increase benthic methane release into\nthe water column and potentially coastal methane emissions to the\natmosphere. This is particularly relevant for relatively shallow coastal\nsites such as Site 7 in the Stockholm Archipelago. Genes encoding\nF 420 -dependent sulfite reductases have been found in ANME-1,\nANME-2, and ANME-3 genomes, 19 suggesting\nthat sulfide-driven sulfite toxicity may be a commonly encountered\nenvironmental pressure by ANME and, therefore, a widespread control\non AOM activity. In conclusion, our data suggest that ANME-2\narchaea might be able\nto compensate for methane increases under hypoxic conditions but are\nunable to thrive under euxinic conditions because of sulfide-driven\ntoxicity. This disruption of the methane biofilter results in increased\nbenthic methane release into the water column in coastal ecosystems\nseverely impacted by eutrophication and bottom water deoxygenation.\nFurther studies should investigate if increased methane concentrations\nin euxinic waters result in increased emissions of methane to the\natmosphere. Moreover, future studies are required to characterize\nthe methane-oxidizing activity of ANME archaea under changing bottom\nwater redox conditions as well as the metabolism and terminal electron\nacceptors utilized."
} | 10,208 |
30038449 | null | s2 | 8,241 | {
"abstract": "1. Despite commonly used to unveil the complex structure of interactions within ecological communities and their value to assess their resilience against external disturbances, network analyses have seldom been applied in plant communities. We evaluated how plant-plant spatial association networks vary in global drylands, and assessed whether network structure was related to plant diversity in these ecosystems. 2. We surveyed 185 dryland ecosystems from all continents except Antarctica and built networks using the local spatial association between all the perennial plants species present in the communities studied. Then, for each network we calculated four descriptors of network structure (link density, link weight mean and heterogeneity, and structural balance), and evaluated their significance with null models. Finally, we used structural equation models to evaluate how abiotic factors (including geography, topography, climate and soil conditions) and network descriptors influenced plant species richness and evenness. 3. Plant networks were highly variable worldwide, but at most study sites (72%) presented common structures such as a higher link density than expected. We also find evidence of the presence of high structural balance in the networks studied. Moreover, all network descriptors considered had a positive and significant effect on plant diversity, and on species richness in particular. "
} | 355 |
37520908 | null | s2 | 8,242 | {
"abstract": "Antibacterial properties of copper against planktonic bacteria population are affected by surface microstructure and topography. However, copper interactions with bacteria in a biofilm state are less studied. This work aims at better understanding the difference in biofilm inhibition of bulk, cold-sprayed, and shot-peened copper surfaces and gaining further insights on the underlying mechanisms using optical and scanning electron microscopy to investigate the topography and microstructure of the surfaces. The biofilm inhibition ability is reported for all surfaces. Results show that the biofilm inhibition performance of cold sprayed copper, while initially better, decreases with time and results in an almost identical performance than as-received copper after 18h incubation time. The shot-peened samples with a rough and ultrafine microstructure demonstrated an enhanced biofilm control, especially at 18 hr. The biofilm control mechanisms were explained by the diffusion rates and concentration of copper ions and the interaction between these ions and the biofilm, while surface topography plays a role in the bacteria attachment at the early planktonic state. Furthermore, the data suggest that surface topography plays a key role in antiviral activity of the materials tested, with a smooth surface being the most efficient."
} | 334 |
30404263 | PMC6189994 | pmc | 8,244 | {
"abstract": "This paper presents an innovative mixing technology for centrifugal microfluidic platforms actuated using a specially designed flyball governor. The multilayer microfluidic disc was fabricated using a polydimethylsiloxane (PDMS) replica molding process with a soft lithography technique. The operational principle is based on the interaction between the elastic covering membrane and an actuator pin installed on the flyball governor system. The flyball governor was used as the transducer to convert the rotary motion into a reciprocating linear motion of the pin pressing against the covering membrane of the mixer chamber. When the rotation speed of the microfluidic disc was periodically altered, the mixing chamber was compressed and released accordingly. In this way, enhanced active mixing can be achieved with much better efficiency in comparison with diffusive mixing.",
"conclusion": "4. Conclusions In this paper, the application of the flyball governor in the micro-mixing function on a centrifugal platform was explored. By controlling the spinning speed of the flyball governor to fluctuate periodically, the sample fluids in the micromixer were chaotically mixed by the repeated compression and releasing actions of the pin pressing against the covering membrane of the mixer chamber. By visual inspection and digital analysis of the photo images of the fluid sample in the mixer chamber to compare the standard deviations of the gray intensities, it was confirmed that this chaotic mixing method enables effective mixing of liquids in 15 s. The technology can be used alone or integrated with other devices such as micropump and pinch-valves actuated using the same flyball governor. It therefore bears significant potential for more complicated lab-on-CD applications.",
"introduction": "1. Introduction Micromixing is a critical process in miniaturized analysis systems. However, mixing of fluids at the microscale faces a big challenge because viscous effects dominate at small scales, where the flow is laminar, and the mixing between different streams in the flow mainly depends on the molecular diffusion. Unfortunately, the diffusive mixing is slow compared with the convective mixing; thus, the mixing length for molecular diffusion is always prohibitively long which negates most of the benefits of miniaturization. In recent years, to reduce the mixing time, many efficient chaotic micromixers have been explored. The concept of the chaotic mixer is to generate chaotic advection via stretching and folding fluids. Generally, micromixers can be broadly classified as two types: passive micromixers and active micromixers. In passive micromixers, the flow field is perturbed by changing the geometry of channels or adding geometric obstacles such as the square-wave micromixer [ 1 ], the zigzag micromixer [ 2 ], and the staggered herringbone micromixer [ 3 ]. In active micromixers, fluids are always perturbed by using an external energy source such as mechanical pulsation [ 4 ], acoustic vibration [ 5 ], magnetic force [ 6 ], electrohydrodynamic force [ 7 ] and electroosmotic force [ 8 ]. These reported strategies can improve mixing performance for conventional microfluidic systems to some extent. In the past few decades, centrifugal microfluidic systems have emerged as an important branch of the microfluidic systems. The liquid mixing has also become a challenging issue for these types of micro systems. Because of the limitation of the spinning platform, the aformentioned approaches cannot be easily implemented for centrifugal microfluidic systems. First, the passive mixer usually requires high pressure due to the resistance in the fluidic channel. In the centrifugal microfluidics, the pressure exerted on the liquid generally can only be applied by the centrifugal force as the disc rotates. It is not practical to achieve the required pressure level in most cases. For the active mixers not based on centrifugal forces, the internal energy sources are difficult to integrate with centrifugal microfluidic platforms. Considerable effort therefore has been dedicated to the development of the micromixing technologies suitable for centrifugal microfluidic platforms. The most common method is known as “shake-mode mixing”, which is to continuously alter the rotating direction of the disc [ 9 ]. The action of the liquid due to the frequent reversal of rotational direction can improve the mixing effect. However, a consequence of shake-mode mixing is that the disk is momentarily stationary when the rotational direction is altered, which can possibly affect the operation of other units on the disc such as the priming process for the siphon valves. Some other methods have also been reported. Noroozi et al. [ 10 ] generated a reciprocating flow between two chambers of a polydimethylsiloxane (PDMS) chip. However, the mixing time is still too long. Kong et al. [ 11 ] blew compressed gas into the mixing chamber and agitated chaotic mixing. The limitations of this method are that the external gas may contaminate the sample liquid as well as the additional complexity of controlling the external gas. In this paper, we report a micromixing technology for the centrifugal microfluidic platform. This technology is based on a flyball governor system described in our previous reported work [ 12 , 13 ]. The mixing chamber is periodically compressed simply by altering the spinning speed of the disc to enhance the mixing effect. The operation of this mixer is in a non-contact fashion without sample contamination issues and with no external energy source required.",
"discussion": "3. Experimental Results and Discussions 3.1. Fabrication of the Microfluidic Disc The microfluidic disc was fabricated in four steps as shown in Figure 3 . The first step was to fabricate the SU-8 master mold with soft lithography method. A thick layer of SU-8 100 photoresist (MicroChem, Westborough, MA, USA) was first spin-coated on a four-inch silicon wafer. Next, fluidic patterns on a mask were transferred to the photoresist in UV exposure followed by the soft bake of photoresist. After the postbake and development, the master mold was obtained for the PDMS casting fabrication step. A mixture of PDMS (Sylgard184, Dow Corning, Midland, MI, USA) was prepared by mixing the base and curing agent in a 10:1 ( m / m ) ratio and casting it on the master mold. After being cured at 85 °C in the oven for 1 h, the PDMS layer was peeled off the master mold and the holes were punched using a sharpened blunt needle. The top and bottom PDMS layers were obtained through similar procedures. In this study, a 4-inch PMMA disc was used as the substrate. The central hole, through which the shaft of the motor passes, and connection holes were drilled by computer numerical control (CNC) machinery. After the PDMS layers were fabricated, the three parts were aligned and directly bonded together [ 14 ]. The bonded components were gently pressed to eliminate air bubbles. The reversible bond was performed without any chemical or plasma treatment. The experiments showed that the direct bonding between the PDMS and PMMA was strong enough to fulfill the requirements of our experiment and no leakage was observed during the operation. Two different sizes of pins were designed, one at 4.2 and 8 mm, respectively. 3.2. Test of the Prototype Micromixer Experiments were conducted to test the mixing efficiency of the chaotic micromixer based on flyball governor actuation using a pin size of 4.2 mm. Photo images in Figure 4 demonstrated four major steps of a mixing cycle. First, samples of 50 μL of red dye solution and 50 μL of deionized (DI) water were introduced into the loading chambers as shown in Figure 4 a, with the red dye on the left chamber and the DI water on the right chamber, respectively. The mixing chamber was initially compressed. The disc was first spun at 800 rpm to remove the compression on the mixing chamber and propel the liquids into the mixing chamber as shown in Figure 4 b. Next, the spinning speed decreased to 600 rpm so that the pin pressed the chamber again to cause rapid mixing as shown in Figure 4 c. When the pin was driven away, it is clear that the liquids were better mixed in Figure 4 d. For the experimental system, it was found that 600 rpm was an appropriate speed. It was found that, at this speed, the pin is about to fully compress the chamber. With the speed significantly higher than 600 rpm, the compression effect became too weak. If the speed is significantly lower than 600 rpm, the acceleration process was increased accordingly and the mixing time was also increased. Figure 5 shows the corresponding sequence of spinning speed variation during each cycle of the mixing operation. The cycle is repeated until the liquids are completely mixed. As shown in Figure 1 , a wireless camera is installed above the microfluidic disc and rotates with it. The camera is used to monitor the movements of the samples and transmit the digital images wirelessly to a computer for data processing and display. Figure 6 gives the sequential images obtained by the wireless camera. These images demonstrated the effectiveness of the mixing after one cycle, two cycles, three cycles, and four mixing cycles, respectively. The total time for completing four mixing cycles was 15 s. To quantify effectiveness of the mixing system, the standard derivation of the pixel intensity of the image of the fluid sample in the mixing chamber was used to numerically represent the state of the mixing process. After each cycle, the image of mixer chamber was captured and stored. For each image, the area of the image for the mixing chamber was digitally cut out as the region of interest. These images were then imported into Matlab for further processing. The gray intensities for all the pixels were obtained. The standard derivation for each image was then calculated. A smaller standard derivation stands for better mixing effect. Figure 7 shows the histograms and the standard derivations after one, two, three, and four cycles, respectively. As the number of pressing-releasing mixing cycles increased from one to four, the gray-scale distributions of the images became more uniform and the standard deviation of the images dropped from 0.284 to 0.167. It also can be noted that, as mixing continues, the distribution of pixel intensities shifts towards unity in the histograms. It also can be noted that using the image analysis method to estimate mixing effective is limited by the camera resolution and the irregular shape of the liquid sample in the mixing chamber. However, the standard deviation of the gray intensities still provides a useful estimation of the mixing performance. For the sake of comparison, experiments were also conducted for the mixing effect of pure diffusion for the same centrifugal platform without using the compressing-releasing mechanism. The same amounts of sample fluids were transferred into the mixing chamber and spun with the disc for 15 s. Because of no compression-releasing actuation, the mixing is therefore primarily based on diffusion under the centrifugal environment. The photo image in Figure 8 shows the mixing result. As can be observed, the liquids stayed mostly unmixed. The standard deviation is calculated to be 0.25. By comparison, the standard deviation of the mixing results shown in Figure 6 d after 15 s was at 0.167, which is significantly better. A study was also conducted to investigate the effect of the operational parameters on the mixing results. First, different rotational speeds of disc were tested. The experimental results showed that the pin started to lose contact with the covering membrane of the mixer to perform the “release” operation at around 800 rpm. Higher rotating speed was not helpful to deliver better mixing performance due to the extra time needed for longer deceleration process in the “compress” operation. Therefore, the highest rotational speed of disc was set to be at 800 rpm. To study the effect of the rotation pattern of the disc on the mixing efficiency, an extra “holding phase” was added in the compression-release mixing process. In this approach, the compression by the pin on the mixing chamber was maintained for a short period (1 s) and then released instead of repeating the fast compression-release process as presented in Figure 5 . Standard deviations of the mixing images for active mixing with one second of “holding phases” after one, two, three, four, and five cycles were obtained, and the results are presented in Figure 9 . As can be obtained from the results in Figure 9 , the extra holding phase did not improve the mixing effectiveness. In the holding phase, the mixing only relies on diffusion. These experimental results further confirmed that the improvement of the mixing effect was primarily caused by the chaotic mixing mechanism rather than the diffusive one. For design purposes, the effect of the pin size was also studied. An 8 mm pin was used to replace the 4.2 mm pin used in the previous experiments. The comparison of the results is demonstrated in Figure 10 . It can be observed that the mixing performance was improved. The main reason is that the larger pin can produce a stronger compression effect to boost the chaotic mixing. However, the size of the pin is limited by the size of the mixer chamber. As a result, to choose a pin size not much smaller than that of the mixer chamber is therefore preferred. This mixing technique works on the chaotic mixing caused by the external mechanical system, and the primary advantage of this mixing method is that it does not require extra entities integrated into the disc such as beads or magnets. The technique can effectively avoid any possible contact contamination of the samples because the actuation of the mixer was achieved on the outside of the covering membrane. Additionally, this mixing technique also relies on the function of the flyball governor for the pinch-valves and inward-pumping technologies reported in our earlier work [ 12 , 13 ]. It is therefore quite easy to integrate the mixer with the pinch-valving and inward pumping units to achieve complex fluid handling on a centrifugal microfluidic platform based on flyball governor actuation. One possible drawback of the mixing method is its applications in dealing with cells or beads may be limited because of possible damage during the compression process. Nevertheless, this mixing method can provide an efficient, non-contact way of mixing liquids on centrifugal microfluidic platforms, adding another possible alternative for the existing mixing technologies."
} | 3,660 |
26462113 | PMC4603686 | pmc | 8,246 | {
"abstract": "Two main mechanisms are thought to affect the prevalence of endophyte-grass symbiosis in host populations: the mode of endophyte transmission, and the fitness differential between symbiotic and non-symbiotic plants. These mechanisms have mostly been studied in synthetic grass populations. If we are to improve our understanding of the ecological and evolutionary dynamics of such symbioses, we now need to determine the combinations of mechanisms actually operating in the wild, in populations shaped by evolutionary history. We used a demographic population modeling approach to identify the mechanisms operating in a natural stand of an intermediate population (i.e. 50% of plants symbiotic) of the native grass Festuca eskia . We recorded demographic data in the wild over a period of three years, with manipulation of the soil resources for half the population. We developed two stage-structured matrix population models. The first model concerned either symbiotic or non-symbiotic plants. The second model included both symbiotic and non-symbiotic plants and took endophyte transmission rates into account. According to our models, symbiotic had a significantly higher population growth rate than non-symbiotic plants, and endophyte prevalence was about 58%. Endophyte transmission rates were about 0.67 or 0.87, depending on the growth stage considered. In the presence of nutrient supplementation, population growth rates were still significantly higher for symbiotic than for non-symbiotic plants, but endophyte prevalence fell to 0%. At vertical transmission rates below 0.10–0.20, no symbiosis was observed. Our models showed that a positive benefit of the endophyte and vertical transmission rates of about 0.6 could lead to the coexistence of symbiotic and non-symbiotic F . eskia plants. The positive effect of the symbiont on host is not systematically associated with high transmission rates of the symbiont over short time scales, in particular following an environmental change.",
"conclusion": "Conclusion In conclusion, we have demonstrated that the combination of a small positive effect of the endophyte on host fitness and imperfect vertical transmission can generate an intermediate steady-state prevalence of an endophyte in a natural population. Our findings also reveal that the link between the effect of a symbiont and its transmission may not apply over the ecological timescale, particularly in the presence of ecological disturbances. Finally, our results highlight the need for additional empirical studies of the physiological processes connecting plant growth, reproduction and endophyte transmission.",
"introduction": "Introduction Symbioses have been implicated in many of the major ecological and evolutionary innovations in the history of life [ 1 ]. For instance, the mitochondria of modern eukaryotes developed from an alpha-proteobacterium internalized by cells 1.45–2 billon years ago [ 2 , 3 ]. However, mitochondria and chloroplasts have become fixed across host populations and generations, whereas this is not the case for contemporary symbioses such as the fungus Neotyphodium in grasses (for a review [ 4 ]) or the bacterium Wolbachia in arthropods [ 5 ]. Consider the fungus from the genera Neotyphodium and Epichloë (Clavicipitaceae, Ascomycota) as an illustration. These vertically transmitted fungal endophytes are prevalent in cool-season grasses. They develop in the aerial tissues of the grass and are transmitted to the next generation of host plants via the seed. Fungal endophytes may confer several benefits on the host, increasing sexual reproduction rates, competitiveness and resistance to abiotic and biotic stresses, such as drought (see [ 6 ]) and herbivores (e.g. [ 7 ]). However, these vertically-transmitted fungal endophytes commonly display a mosaic of prevalence, ranging from 0 to 100%, in host populations [ 8 – 11 ]. Does this imply that contemporary symbioses are not sufficiently beneficial to their hosts to have become fixed? This hypothesis is intuitive and consistent with commonly understood principles of the mode of transmission and effects of symbiosis: exclusive vertical transmission would be expected to favor mutualism and to generate high frequencies of symbiosis in host populations [ 12 , 13 ]. Investigations of the interplay between symbiont transmission and effects in the generation of current prevalence patterns are required to improve our understanding of why some symbioses are fixed in host species, whereas others are not. Theoretically, three principal mechanisms shape the overall pattern of variation of endophyte frequencies in a plant population: i) the outcome of symbiosis, in terms its benefits or harm to the host (e.g. [ 14 ]), ii) the mode of transmission (e.g. vertical and/or horizontal) and relative rates of transmission (e.g. [ 15 ]), and iii) the migration of symbiotic (S) and non-symbiotic (NS) organisms (i.e. via seeds) between neighboring populations [ 16 ]. Theoretical models suggest that imperfect transmission leads to the disappearance of the symbiont from the host population even when it is highly beneficial [ 17 , 18 ]. Recent empirical data are consistent with this hypothesis. Using a population modeling approach, Yule et al . [ 19 ] demonstrated that vertical transmission rates below a certain threshold could lead to symbiont extinction, even if the symbiont increased the net growth of Agrostis hyemalis populations. The proximal explanation for this is straightforward: the magnitude of the beneficial effect does not sufficiently compensate for the rate of symbiont loss. The distal explanation is much more complicated: why eliminate a beneficial symbiont, even if it is only slightly beneficial? The elimination of an advantageous endophyte over an evolutionary time scale is counterintuitive [ 12 , 13 ]. To our knowledge, this scenario of the elimination of an advantageous symbiont due to imperfect vertical transmission has never been demonstrated in a natural population, shaped by evolutionary history . It therefore remains to demonstrate whether and how it occurs in natural populations. We considered two scenarios in which a positive effect of the endophyte was counteracted by weak transmission. These two scenarios did not result in the same probability of a fungal endophyte of grasses becoming fixed across host populations and generations. Under the first scenario, despite fluctuations over short time scales, there was a strong link between the transmission and effect of the symbiont. The positive effect was thus associated with weak transmission due to an episodic stress or disturbance. Evidence in favor of this “disturbance” hypothesis would be provided by a shift from high to low in the endophyte transmission rates. This shift should occur when environmental conditions are modified and in a population where the endophyte has a positive effect on the host. In the second scenario, the endophyte also increased host fitness, but its rate of transmission was low due to a trade-off between the benefit accrued and transmission. We called this scenario the “internal-limitation” hypothesis; it is an intrinsic characteristic of the symbiosis that precludes it transmission. As an example of this scenario, symbiosis may increase plant fitness by increasing vegetative growth rate, but high plant growth rates decrease the ability of the symbiont to colonize all the reproductive tillers of the plant (i.e. dilution effect; [ 20 ]). Evidence in favor of this scenario would be provided by a net positive effect of the symbiosis counteracted by a low transmission rate in a population with an intermediate steady-state endophyte prevalence. These scenarios lead to different evolutionary trajectories for the grass-endophyte symbiosis. The “disturbance” scenario enlarges the range of ecological conditions in which a symbiont can persist and be fixed across host generations. By contrast, in the “internal-limitation” scenario, the grass-endophyte symbiosis is unlikely to become fixed in host populations and generations with its current characteristics. Fixation is likely to occur only if a mutation arises that can disrupt the trade-off between symbiont advantage and transmission. We analyzed the link between endophyte prevalence, effects and transmission in a natural population of an alpine grass, Festuca eskia , with intermediate levels of endophyte colonization. F . eskia harbors an asexual form of the endophytic fungus Epichloë festucae , and the proportion of symbiotic individuals in the population may range from 0 to 99%, depending on the location [ 10 ]. This plant species appears to be a simple and relevant model for investigations of the combination of mechanisms underlying the endophyte-grass system. Gene flow between populations is limited in this species [ 10 ], making it unnecessary to include migration processes in the model. We identified a population with an intermediate level of endophyte colonization (i.e. 50% of the plants symbiotic) over at least the last five years [ 10 ]. We used population modeling methods to estimate host-endophyte fitness. Given the connection between demography and fitness [ 21 ], population modeling appears to be an appropriate way of evaluating the outcome of symbiosis over the entire life cycle of the plant [ 19 , 22 ]. The use of this demographic approach in a wild population made it possible to assess the relative dynamics of symbiotic and non-symbiotic plants due to both environmental conditions and the life-history traits of the population. We focus here on the weighting of demographic parameters for determining the frequency of symbiotic and non-symbiotic plants, rather than on the predictive dimension in terms of population density. We addressed the following questions: 1) How are intermediate levels of endophyte colonization generated in a natural alpine grass population? 2) Are the transmission rates and effects on the host of the endophyte linked during ecological disturbances? 3) Which demographic traits can account for the differences between S and NS plants?",
"discussion": "Discussion To our knowledge, this study is the first to describe the mechanistic links between endophyte prevalence, effects on host fitness and transmission in a natural grass population. Our findings suggest that symbiotic and non-symbiotic plants can coexist in a stable manner within grass populations, due to the combination of a positive effect of the endophyte on host fitness, and imperfect vertical transmission. We also demonstrated that changes to nutrient resources dissociated the transmission rate of the endophyte from its effects, leading to the disappearance of the endophyte from the host population. These findings are clearly consistent with the “disturbance” hypothesis. The transmission and effects of the symbiont can be dissociated over short time scales in the presence of a disturbance (a change in soil nutrient levels in this study). However, they are also consistent with the “internal-limitation” hypothesis. Under this hypothesis, a positive effect of the symbiosis would be counteracted by weak transmission, even in the lack of an environmental disturbance, due to a trade-off between symbiont benefit and transmission. Our results showed a positive effect of the endophyte on population growth rate, in the presence of transmission rates of about 0.63 to 0.87. A transmission threshold of 0.1–0.2 was calculated, below which the symbiont frequency in the population was zero. The transmission rates observed in this study are consistent with those reported in the literature (about 0.92 for Epichloë species and 0.75 for Neotyphodium species, see [ 33 ]). However, our result raises questions about how “weak” transmission should be defined. Exclusively perfect vertical transmission is known to be rare in symbiotic systems. With the exception of mitochondria and chloroplasts, all vertically transmitted symbionts display some plasticity in their rates of vertical transmission (see [ 34 ]). The use of terms such as “imperfect transmission” in endophyte-grass symbiosis assumes that perfect transmission is the baseline condition, but quantitative evaluations of transmission rates have shown that these rates range from 0 to 1 [ 33 , 35 ]. Further studies are clearly required to establish the physiological link between plant growth, fecundity and endophyte transmission. How are intermediate levels of endophyte colonization achieved in a natural alpine grass population? Our findings identify a positive effect of the endophyte and imperfect vertical transmission as factors accounting for the stable coexistence of symbiotic and non-symbiotic plants of F . eskia . Indeed, in our model, the positive effect of the endophyte on symbiotic plants was enough to account for the preponderance of endophytes in grass populations. However, this positive effect was counteracted by intermediate rates of vertical transmission (i.e. 0.63 to 0.87). Our results are consistent with the predictions of several theoretical models. Those of Ravel et al . [ 36 ] and Gundel et al . [ 18 ] predict the persistence of non-symbiotic grasses in a population, under an assumption of non-propagation of the endophyte. We demonstrated that symbiotic and non-symbiotic plants could coexist in a stable manner within grass populations. Intermediate endophyte prevalences are the norm, rather than the exception, in grass species [ 9 , 33 ], but only a few empirical studies have focused on populations with intermediate symbiotic frequencies. Instead, most studies have focused on individuals from populations with high symbiotic frequencies. For example, Yuel et al . [ 19 ] studied native grasses for which 97% of the seeds contained the symbiotic endophyte. Davitt et al . reported an endophyte prevalence of 96% and Kannadan and Rudgers [ 37 ] reported an endophyte prevalence of about 74 to 100%. Populations with intermediate frequencies have generally been considered not to be in equilibrium and to display transient dynamics [ 18 ], due to environmental fluctuations or seed immigration [ 16 , 38 ]. Our work thus clarifies the status of populations with intermediate symbiotic frequencies, and provides a rationale for further studies of the specific trajectories underlying the establishment of intermediate frequencies. Are these two mechanisms linked in conditions of ecological disturbance? Our findings indicate that the two mechanisms identified act independently following a change in soil resource level. When fertilizer was applied (F+), the fungal endophyte enhanced F . eskia fitness, but vertical transmission rates were slightly lower than in the absence of fertilizer, resulting in a frequency of E . festucae of 0% in the population. The apparent paradox between the enhancement of host fitness by the endophyte and the decrease in endophyte transmission observed in our study indicates that these two mechanisms are not linked by a monotonous relationship over the ecological time scale. It is therefore not possible to estimate one mechanism directly from the other. From an ecological perspective, our results are consistent with those of Gibert et al . [ 35 ], showing, for several F . eskia populations with 11 to 90% S plants, an apparent discrepancy between the effect of the endophyte on the host and the rate of vertical endophyte transmission. Which demographic traits can explain the difference between S and NS plants? The prospective analysis (elasticity) identified three demographic parameters as particularly likely to affect population growth rate in Festuca eskia : the probability of a seedling surviving to the juvenile stage (G J1 ), the probability for a year-one juvenile surviving and reaching the adult stage (G A1 ) and the mean number of seeds produced per plant (F). However, only a component of F, e 0 the probability of seedling emergence from the seed, explained the observed differences in population growth rates between NS and S plants in both sets of conditions. This finding is quite logical for a symbiosis in which seeds are the vector for the transmission of the symbiont across host generations. Inconsistent effects of endophyte on plant germination rates have been reported in the literature (see [ 39 ]). It would therefore be of interest to investigate the beneficial effects of the endophyte (resistance to drought stress, etc.) through this demographic parameter. Theoretical and empirical studies have predicted the existence of a threshold vertical transmission rate, below which the positive effect of the endophyte cannot compensate for the rate of symbiont loss [ 17 – 19 ]. Here, we document the occurrence and magnitude of such a transmission threshold in a natural system: at vertical transmission rates below 0.1–0.2 under natural condition–the precise threshold depending on the life-cycle stage considered—the frequency of symbionts in the population was zero in our study. This result is consistent with both the theoretical results of Ravel et al . [ 17 ], and the empirical results of Yule et al . [ 19 ], who predicted that symbiotic grasses could persist in the population, provided that transmission rates exceeded 0.1. Our results also demonstrated that this threshold was higher under F+ treatment than under control conditions, suggesting that it could change across environmental gradient. Our results also indicate that vertical transmission from adult to seed (TA) is more critical for the persistence of symbiosis than vertical transmission from seedling to juvenile (TS) in a F . eskia population. In our study, the vertical transmission threshold beyond which symbiosis persisted in the population was higher for TA (i.e. 20%) than for TS (i.e. 10%) under natural conditions. This result is consistent with previous studies reporting the tiller-seed transition (TA) as the transition during which symbiosis is most frequently lost in grass-endophyte systems (see [ 33 ]). Yet, the relative importance of the vertical transmission parameters appears to be dependent on environmental conditions. Indeed, in our study the vertical transmission threshold beyond which symbiosis persisted in the population was higher for TS (i.e. 20%) than for TA (i.e. 10%) under fertilized conditions, suggesting that the different vertical transmission parameters have to be studied simultaneity. Model validation The assumption of a fixed survival rate of 0.95 in adult, in addition to the use of a sowing experiment, resulted in an overestimation of population growth rates in our study. The absolute values of lambda were > 1 (i.e. expanding population). However, the ranking of population growth rates between symbiotic, non-symbiotic, natural and fertilized conditions is more important than the absolute value of growth rate in this study. And this ranking was not affected by the choice of an arbitrary value for survival (here 0.95, but simulations were performed with a range of values, from 0.5 to 0.99, result not shown). Finally, the endophyte prevalence values estimated by our model are consistent with observation from the Festuca eskia population in Guzet,"
} | 4,799 |
38806482 | PMC11133408 | pmc | 8,248 | {
"abstract": "We report a breakthrough in the hardware implementation of energy-efficient all-spin synapse and neuron devices for highly scalable integrated neuromorphic circuits. Our work demonstrates the successful execution of all-spin synapse and activation function generator using domain wall-magnetic tunnel junctions. By harnessing the synergistic effects of spin-orbit torque and interfacial Dzyaloshinskii-Moriya interaction in selectively etched spin-orbit coupling layers, we achieve a programmable multi-state synaptic device with high reliability. Our first-principles calculations confirm that the reduced atomic distance between 5 d and 3 d atoms enhances Dzyaloshinskii-Moriya interaction, leading to stable domain wall pinning. Our experimental results, supported by visualizing energy landscapes and theoretical simulations, validate the proposed mechanism. Furthermore, we demonstrate a spin-neuron with a sigmoidal activation function, enabling high operation frequency up to 20 MHz and low energy consumption of 508 fJ/operation. A neuron circuit design with a compact sigmoidal cell area and low power consumption is also presented, along with corroborated experimental implementation. Our findings highlight the great potential of domain wall-magnetic tunnel junctions in the development of all-spin neuromorphic computing hardware, offering exciting possibilities for energy-efficient and scalable neural network architectures.",
"introduction": "Introduction As a new paradigm for parallel processing, neuromorphic computing (NC) has attracted intensive attention worldwide due to the great potential in artificial intelligence (AI) and big data analysis applications with overwhelming performance than the conventional von Neumann architecture 1 – 5 . Deep neural networks (DNNs), mimicking the biological structure and working principles of human brains, have been widely applied in various areas such as image 6 , 7 , speech 8 and video recognition 9 , and data classification 10 , demonstrating superior advantages among the applications requiring unprecedently increased speed and capacity for training huge data sets. In general, DNNs consist of multiple layers connected through synapses with updated weights. The summation of products of inputs and corresponding synaptic weights is calculated first and then applied with an activation function to get the output of such specific layer which could further act as the input to the next layer. Synapses with linear weight modulation and neurons with non-linear activation functions are basic elements of most importance constructing a neural network 11 . Explicitly, compact hardware implementation of efficient and reliable bio-inspired synapse and neuron devices is one of the major challenges limiting the development of the NC chip 2 , 12 . Notably, hardware implementation of neuromorphic devices based on emerging nonvolatile memories (NVMs) offers significant performance advantages when combined with traditional complementary metal-oxide semiconductor (CMOS) technology. The integration of NVMs and CMOS electronics provides benefits such as nonvolatility, scalability, direct mapping of synaptic weights, as well as facilitating functions such as data thresholding, conversion, and trimming required for each layer of a neuromorphic DNN 7 , 13 – 18 . There are typical reports on the realization of operations in DNNs using different types of NVMs, such as resistive random-access memory (RRAM) 19 , 20 , phase change memory (PCM) 21 , ferroelectric RAM (FeRAM) 22 , flash memory 23 , and magnetic RAM (MRAM) 24 . While these NVMs show promise for neural network applications, they also come with inherent challenges related to nonlinearity, energy efficiency, area overhead, and reliability 3 . These challenges make it difficult to customize the NVMs, resulting in a loss of learning accuracy and hindrances in implementing specific either synaptic or non-linear activation functions. These issues pose significant confrontations for the practical implementation of NVMs in neural network applications 25 . Nevertheless, each type of NVM possesses its own advantages and disadvantages, with some being slower, having limited endurance, larger area footprints, or only supporting 1-bit operations. There is pressing need to explore the uniqueness among different memory features and synergistic integration with CMOS in order to achieve optimal performance in neuromorphic DNN computing. Importantly, spintronic devices with rich, reproducible and controllable magnetization dynamics, which can emulate functions of synaptic and various types of neurons 25 – 36 , have been receiving increasing attention worldwide in recent years. Among the spintronic devices, the domain wall (DW) magnetic tunnel junction (MTJ) leveraging DW dynamics, which can be precisely manipulated by all electrical methods 37 – 41 as an information token, is an ideal candidate for application to both linear weight updating and nonlinear activation functions in neural networks because of its intrinsic linear relationship between junction magneto-resistance and programing stimulus. Particularly, in recent demonstrations of CoFeB/MgO based MTJs, the tunnel magnetoresistance (TMR) ratio has significantly improved (>200% at room temperature) while keeping read and write voltage at relatively low values (~0.5 V), manifesting the promise of this technology 42 , 43 . In the present work, a distinct type of DW perpendicular MTJs (pMTJs) based multi-state synaptic device is experimentally demonstrated, in which a series of DW pinning centers (PCs) is introduced by selective etching of the spin-orbit coupling (SOC) layer to tailor the interfacial Dzyaloshinskii–Moriya interaction ( i DMI) strength in the PC regions. The uniformly spaced PCs result in controllable and stable multi-states that can be linearly manipulated by a magnetic field or pulsed electrical current. Next, a novel sigmoid activation function generator is explored based on the same design scheme and fabrication process flow as the developed synaptic device. PCs in the activation function generator are non-linearly distributed which results in a sigmoidal-like resistance state switch driven by the synergetic effect of spin-orbit torque (SOT) and tunable i DMI as elaborated by extensive first-principles ab initio investigations. The systematic experimental and theoretical calculation results with micro-magnetic and circuit-level co-simulations complementarily verified the feasibility of the proposed sigmoidal activation function generator. It is essential to emphasize that the present research primarily focuses on advancing spintronic components, specifically spintronic synapses and neurons, within a larger framework that integrates CMOS electronics. Our DW-pMTJs based all-spin synaptic and sigmoidal neuron prototype shows great potential in energy-efficient neuromorphic hardware development with high performance in a standard CMOS-process-technology-compatible way.",
"discussion": "Discussion Spurring growth of generative AI models with ever-more-sophisticated devices, hardware implementation of synaptic and neuron devices poises one of the major challenges and constrains the scalability and high-dense integration of neural networks due to the large on-chip area and high-power consumption. We experimentally demonstrated a unique spin synaptic device by introducing a series of effective DW PCs achieved by selective heavy metal partial etching with modulated local i DMI. A pioneering proof of concept of dynamic tailoring DW pinning via ion-beam-etching-strain modulated antisymmetric exchange strength is validated at the HM/FM interface in the MTJ. The extensive cross-sectional high-resolution transmission electron microscopy images and energy source of local i DMI from first-principles calculations are being conducted to further comprehend the mechanism responsible for i DMI engineering. Reliable state-by-state switching corroborates the high controllability, stability, and repeatability of the multi-states introduced by DW pinning and depinning dynamics in the designed PCs. The linear multi-states against the magnetic field are direct mapping of the synaptic function. We further verified the feasibility of constructing a sigmoidal spin-neuron using the same scheme as that of a synaptic device experimentally. By further complementary micro-magnetic and circuit-level co-simulation, a sigmoid activation function generator has been successfully demonstrated based on a DW-pMTJ driven by synergistic SOT and i DMI with an energy consumption of 36.3 fJ/pulse. A neuron circuit design with a compact sigmoidal cell area and low power consumption is also presented, along with corroborated experimental implementation. Compared to state-of-the-art counterparts of neural activation function generator, our developed architecture is competitive as shown in Figure S9 benchmark diagram. The overall energy consumption is less than 508 fJ/operation including the reset process with a competitive firing rate up to 20 MHz. Such developed devices are compatible with the current standard CMOS technology and the magnetoresistive random-access memory process, without any exotic material, complicated structure, and extra masks in comparison with the state-of-the-art, offering a promising and applicable candidate for neuromorphic devices and chips application in new trajectories with a device-circuit perspective and the envision design of all-spin neuron circuits. By offering this comprehensive elucidation, it is expected to provide more nuanced insights into the intricate integration of spintronic and CMOS technologies within the realm of neuromorphic computing."
} | 2,427 |
37155857 | PMC10194018 | pmc | 8,249 | {
"abstract": "Significance Dissimilatory sulfate reduction (DSR) is one of the oldest and most prominent microbial metabolic pathways on Earth. It is generally accompanied by zero-valent sulfur (ZVS) that is involved in several cryptic pathways in marine and terrestrial environments. In this study, we identified a to-date unknown DSR pathway or sulfate-to-ZVS conversion mediated by sulfate-reducing microorganisms. This finding provides further insights into the sulfur cycle, which may help reveal details about cryptic element cycling pathways and improve our understanding of the sulfur metabolism in early Archaean microorganisms.",
"discussion": "Results and Discussion Characterization of ZVS Generated from SRM-Mediated DSR. Recently, we identified ZVS generation in a sulfate-reducing and methanogenic bioreactor, in which known sulfide-oxidizing microorganisms and available sources of oxidants were excluded ( 20 ). We hypothesized that ZVS could be derived from the DSR rather than through the reoxidation of sulfide under anaerobic conditions. To test this hypothesis, we examined ZVS generation in six typical strains of phylogenetically diverse SRMs isolated from varied niches ( SI Appendix , Fig. S1 ), including Desulfovibrio vulgaris Hildenborough (DvH) as a model SRM. In these pure cultures, ZVS generation occurred along with sulfate-to-sulfide reduction and peaked when sulfate reduction terminated, resulting in 6.9 to 12.9% (a mean value of 8.9%) sulfate reduction to ZVS, e.g., 8.4 ± 1.5% in DvH ( Fig. 1 A and SI Appendix , Fig. S2 ). Thereafter, ZVS levels decreased and stabilized in the late stationary phase ( Fig. 1 A and SI Appendix , Fig. S2 ). The ZVS generation was further corroborated by the appearance of yellowish color (because of elemental sulfur) in SRM cultures without amendment of Fe(II/III) ions ( Fig. 1 B ). Such yellowish color could be over-shadowed by black FeS precipitates in the SRM-cultivating LS4D medium. Quantification of the total and extracellular ZVS showed that 8.5 to 38.1% of the ZVS was intracellular sulfur ( Fig. 1 C and SI Appendix , Fig. S3 ). Desulforamulus ruminis DL (DLT) as an SRM with a gram-positive-type cell wall structure ( 22 ) generated comparatively more abundant intracellular sulfur relative to that of gram-negative DvH, i.e., 38.1 ± 3.7% and 11.3 ± 1.8% intracellular sulfur in DLT and DvH cultures, respectively, fed with a molar ratio of lactate (C) to sulfate (S) of 1:1 ( Fig. 1 C ). This difference in intracellular sulfur accumulation could be attributed to the difficulty in crossing-cell-wall transportation of ZVS in DLT. Fig. 1. Generation of zero-valent sulfur (ZVS) from DSR. ( A ) Dissimilatory sulfate reduction (DSR) and ZVS generation in Desulfovibrio vulgaris Hildenborough (DvH). ( B ) Observed yellowish color in DvH cultures without amendment of Fe(II/III). DvH without Fe(II/III), DvH in the LS4D medium without the amendment of Fe(II/III); Abiotic control, LS4D medium with sulfate- and sulfide-amendment but without DvH inoculation; Biotic control, DvH in the LS4D medium. ( C ) The total/extracellular ZVS generation of DvH and DLT ( Desulforamulus rumins DL) in cultures amended with two different molar ratios of lactate (C) to sulfate (S), i.e., C/S ratios of 1:1 and 2:1. ( D ) The temporal patterns of ZVS compositional profiles in DvH. Data are presented as mean value of triplicate cultures. ( E ) Representative Raman spectrum of elemental sulfur in DvH, containing prominent peaks of the elemental sulfur (S 8 ) at 151.6, 220.8, and 473.6 cm − 1 ; a.u., arbitrary units. ( F ) DSR-derived ZVS and sulfide, as well as the ratio of ZVS to reduced sulfur (i.e., ZVS and sulfide), in DvH cultures with varied pH, temperature (Temp.), C/S ratio, and salinity (%, w/v). Error bars represent SDs of triplicate cultures. ZVS could be present as elemental sulfur ( S n , n ≥ 1 0 ) and/or polysulfide ( S n , n ≥ 2 2 - ), qualitative and quantitative analyses of which are challenging ( 20 , 21 ). In this study, multiple complementary methods were used to identify and quantify total ZVS ( 23 , 24 ), elemental sulfur ( 25 , 26 ), and polysulfide ( 27 ) under anaerobic conditions. The elemental sulfur accounted for 84.1 to 90.4% (a mean value of 87.2%) of the total ZVS at ZVS peak time ( Fig.1 D and SI Appendix , Fig. S3 ) and mainly constituted of S 8 , S 7 , and S 6 ( Fig. 1 D and SI Appendix , Fig. S4 A ). The high ratio of S 8 (>84.0% of the elemental sulfur) was confirmed by the strong Raman peaks for S 8 ( Fig. 1 E ), which could be further converted into S 6 and S 7 ( SI Appendix , Fig. S4 A and B ) ( 28 ). In contrast, a diverse range of polysulfide species (from di- to nona-sulfide) were identified to be generated in DSR ( SI Appendix , Fig. S4 C ), of which the total amount only accounted for 9.6 to 15.9% (a mean value of 12.8%) of the total ZVS ( Fig. 1 D and SI Appendix , Fig. S3 ). The temporal changes of the elemental sulfur and polysulfide followed a similar trend in DvH ( Fig. 1 D and SI Appendix , Fig. S4 D ). The nona-sulfide could be derived from the biogenetic and pH-dependent reaction of S 8 and sulfide, which was further transferred into di- to octa-sulfide species ( 29 ). Notably, extracellular ZVS shared a similar composition profile with that of the total ZVS in the DvH culture ( SI Appendix , Fig. S4 D ). Environmental stresses (e.g., limited reducing equivalent, high pH, low temperature, and high salinity) could change the metabolism and, consequently, sulfate reduction of SRMs ( 12 ). We compared the molar ratio of ZVS to reduced sulfur (ZVS and sulfide) and ZVS compositions in DvH cultures under growth conditions with varied pH, temperature, C/S ratio, and salinity ( Fig. 1 F ). Adjustment of C/S ratios by changing molar concentrations of lactate (C) and sulfate (S) altered the ratios of ZVS to reduced sulfur in the range of 8.2 to 13.1% ( Fig. 1 F and SI Appendix , Fig. S5 ). In contrast, high pH (pH = 9.0) and low temperature (20 °C) slightly changed the ratio of ZVS to reduced sulfur, compared to control cultures ( Fig. 1 F ). Strikingly, high salinity (2%, w/v) remarkably increased the ratio of ZVS to reduced sulfur, e.g., 8.2 ± 0.3% in cultures with no salinity vs. 24.0 ± 3.5% in cultures with 2% (w/v) salinity ( Fig. 1 F ). S 8 was generated as the predominant ZVS species under these saline conditions ( SI Appendix , Fig. S5 ). The comparatively higher ratio of ZVS to reduced sulfur under saline conditions hint a higher flow ratio of sulfate-to-ZVS in marine environments relative to fresh waters, being consistent with observations of a large amount of ZVS in marine sediments ( 21 , 30 ). ZVS Generation from DSR rather than Reoxidation of Sulfide. To eliminate the possibility that ZVS could have been generated from the oxidation of sulfide under anaerobic conditions ( 3 , 31 ), we carried out experiments using radiolabeled sulfate ( 35 S-sulfate) or sulfide ( 35 S-sulfide) in DvH cultures. If sulfide-to-ZVS re-oxidation occurred, one would expect i) a lag time between sulfide- and ZVS-generation curves in the 35 S-sulfate-fed cultures and ii) an accumulation of radiolabeled ZVS ( 35 S-ZVS) along with decreases in 35 S-sulfide in the 35 S-sulfide-amended cultures ( 32 , 33 ). This prediction was contradicted by our observations of synchronous generation of sulfide and 35 S-ZVS in 35 S-sulfate-fed cultures ( Fig. 2 A and SI Appendix , Fig. S6 ) and by the absence of 35 S-ZVS accumulation or 35 S-sulfide consumption in 35 S-sulfide-amended cultures ( Fig. 2 B ). Additional evidence of the negligible role of sulfide-to-ZVS oxidation in our experiments comes from transcription and translation analyses. The biotic sulfide-to-ZVS oxidation can be catalyzed by sulfide-quinone reductase (Sqr) and flavocytochrome c sulfide dehydrogenase ( 3 , 31 , 33 ), potentially harbored by SRMs ( 31 ). Although DvH contains a sqr -like gene ( SI Appendix , Fig. S7 ), transcriptome and proteome analyses of DvH cultures fed with lactate and sulfate at their molar ratios of 1:1 and 2:1 showed high and comparable transcriptional/translational levels of dissimilatory sulfite reductase-encoding genes ( dsrABC ), especially the highly transcribed and translated dsrC gene, while the transcription and translation of sqr -like gene were negligible ( Fig. 2 C and SI Appendix , Fig. S8 and Table S1 ). The qPCR quantification of temporally transcribed dsrC and sqr -like genes demonstrated no correlation between the sqr -like gene transcription and ZVS generation ( Fig. 2 D ), further ascertaining the negligible role of Sqr in mediating reoxidation of sulfide-to-ZVS in DvH cultures. Collectively, 35 S-radiosulfur-isotope-labeled experiments and transcription/translation analyses unambiguously eliminate the possibility of sulfide-to-ZVS oxidation and reinforce our discovery of synchronous generation of elemental sulfur S 8 , along with sulfide, from DSR in SRMs and subsequent partial conversion to polysulfide ( Fig. 2 E ). In this pathway, DsrC-trisulfide as an important intermediate from DsrAB-mediated sulfite reduction ( 17 ) could be critical to its subsequent conversion to both ZVS and sulfide, which awaits future studies. Fig. 2. Isotopic and transcriptomic analyses of ZVS generation from DSR. ( A ) 35 S-ZVS accumulation in DvH cultures fed with 35 S-sulfate. ( B ) 35 S-ZVS in DvH cultures fed with both sulfate and 35 S-sulfide. ( C ) Transcription of key sulfur metabolism-related genes in DvH cultures fed with different molar ratios of lactate to sulfate, i.e., C/S ratios of 1:1 and 2:1. Statistical significance is based on the T test (n = 3), which is denoted by asterisks (*), P < 0.05. ( D ) qPCR-quantified transcription of dsrC and sqr genes in DvH. ( E ) A proposed model for the ZVS generation from DSR in SRB. Error bars represent SDs of triplicate cultures. Biogeochemical Implications of DSR-Derived ZVS. The SRMs-mediated DSR plays a key role in the past and present global cycle of sulfur with profound influence on the global cycles of other elements as well ( 1 , 2 , 34 ). In the modern marine environment, about 11.3 teramoles of sulfate are estimated to be reduced yearly, which accounts for the oxidation of 12 to 29% of the global organic carbon flux to the sea floor ( 2 ). If 8.9% of the sulfate reduction is channeled to ZVS, there will be around 1.01 teramoles of DSR-derived ZVS that may support a wide range of ZVS-metabolizing microorganisms, e.g., Geobacter , Pelobacter , and Dethiobacter ( SI Appendix , Table S2 ) under thermodynamically favorable conditions ( SI Appendix , Table S3 ). In our previous studies, we isolated a ZVS-reducing Geobacter (Geo) with acetate as both a carbon source and electron donor ( 35 ), which was used to form a DvH-Geo coculture fed with lactate and sulfate in this study. Compared to the DvH pure culture, lower concentrations of sulfide and slightly higher concentrations of ZVS were generated in the DvH-Geo coculture ( Fig. 3 A ), suggesting an improved sulfate reduction in the coculture relative to the DvH pure culture. In contrast to no cell growth in the pure Geo culture fed with lactate and sulfate, both Geo and DvH in the coculture grew with the sulfate reduction and ZVS generation at the highest cell concentrations of 2.22 × 10 5 and 3.97 × 10 8 16S rRNA gene copies per ml, respectively, at the stationary phase ( Fig. 3 B ). The low cell concentration of Geo in the coculture could be due to the limited amount of ZVS and low energy yield derived from ZVS reduction ( SI Appendix , Table S2 ) as well as trace amount of acetate generated by DvH. The two populations may form a network through carbon- and sulfur-metabolisms, suggesting the DvH-derived ZVS-dependent cell growth of Geo ( Fig. 3 C ). The analysis of 16S rRNA gene amplicon sequencing-based metadata of worldwide marine and terrestrial samples showed that similar metabolic networks between varied lineages of SRMs and ZVS-metabolizing microorganisms (S 0 MMs, including ZVS-reducing-microorganisms/S 0 RMs and ZVS-disproportionating-microorganisms/S 0 DMs) were ubiquitous in natural environments ( Fig. 3 D and SI Appendix , Table S4 ). Notably, many DSR sites being in the absence of taxonomically characterized S 0 MMs had both polysulfide reductase-encoding gene ( psr ) and dsr genes in their metagenomic data ( Fig. 3 D and SI Appendix , Table S5 ), indicating the coexistence of taxonomically uncharacterized S 0 MMs with SRMs. Therefore, both the 16S rRNA and metagenomic data analyses indicated the global distribution of coexisted SRMs and S 0 MMs. The multicomponent interaction between sulfur-based metabolizing microorganisms is also essential for us to precisely decipher geological records toward a deeper understanding of our Earth’s earliest microbial ecosystems ( 10 , 36 ). Fig. 3. DSR-derived ZVS supported cell growth of ZVS-metabolizing microorganisms (S 0 MMs). ( A ) DSR and ZVS generation in a coculture of DvH and Geobacter lovelyi LYY (Geo). ( B ) Cell growth of DvH and Geo in the coculture with pure cultures as the controls. Pure culture controls of DvH and Geo were prepared under the same growth conditions with the DvH-Geo coculture. ( C ) Proposed syntrophic interactions between DvH and Geo in the coculture. ( D ) The global distribution of cooccurred SRMs and ZVS-metabolizing microorganisms (S 0 MMs, i.e., ZVS-reducing-microorganisms/S 0 RMs and ZVS-disproportionating-microorganisms/S 0 DMs) in marine and terrestrial environments by meta-analysis. The metadata were retrieved from previous studies based on 16S rRNA gene amplicon sequences (colorized bubbles) and metagenome sequencing data (gray bubbles). See SI Appendix , Tables S4 and S5 for the detailed metadata information. Error bars represent SDs of triplicate cultures. In summary, we unraveled a thus far unrecognized pathway of sulfate-to-ZVS in the SRMs-mediated DSR process. This pathway may lay the foundation for better understanding of the fate and role of ZVS in the cryptic sulfur cycle in oxygen minimum zones ( 37 ), ANME-SRMs-based methane oxidation ( 4 – 6 , 34 , 38 ), sedimentary pyrite synthesis ( 16 ), and even the sulfur metabolism of early Archaean microorganisms ( 36 ). The identification of DSR-derived ZVS also fuels future exploration of the underappreciated sulfur metabolism in worldwide marine and terrestrial environments for the complete understanding of how sulfur biogeochemical cycling shapes our planet’s surface, atmosphere, and climate."
} | 3,652 |
23515612 | null | s2 | 8,250 | {
"abstract": "Ant protection of extrafloral nectar (EFN)-secreting plants is a common form of mutualism found in most habitats around the world. However, very few studies have considered these mutualisms from the ant, rather than the plant, perspective. In particular, a whole-colony perspective that takes into account the spatial structure and nest arrangement of the ant colonies that visit these plants has been lacking, obscuring when and how colony-level foraging decisions might affect tending rates on individual plants. Here, we experimentally demonstrate that recruitment of Crematogaster opuntiae (Buren) ant workers to the EFN-secreting cactus Ferocactus wislizeni (Englem) is not independent between plants up to 5 m apart. Colony territories of C. opuntiae are large, covering areas of up to 5,000 m(2), and workers visit between five and 34 EFN-secreting barrel cacti within the territories. These ants are highly polydomous, with up to 20 nest entrances dispersed throughout the territory and interconnected by trail networks. Our study demonstrates that worker recruitment is not independent within large polydomous ant colonies, highlighting the importance of considering colonies rather than individual workers as the relevant study unit within ant/plant protection mutualisms."
} | 320 |
26172732 | null | s2 | 8,251 | {
"abstract": "High-throughput experimental techniques and bioinformatics tools make it possible to obtain reconstructions of the metabolism of microbial species. Combined with mathematical frameworks such as flux balance analysis, which assumes that nutrients are used so as to maximize growth, these reconstructions enable us to predict microbial growth. Although such predictions are generally accurate, these approaches do not give insights on how different nutrients are used to produce growth, and thus are difficult to generalize to new media or to different organisms. Here, we propose a systems-level phenomenological model of metabolism inspired by the virial expansion. Our model predicts biomass production given the nutrient uptakes and a reduced set of parameters, which can be easily determined experimentally. To validate our model, we test it against in silico simulations and experimental measurements of growth, and find good agreement. From a biological point of view, our model uncovers the impact that individual nutrients and the synergistic interaction between nutrient pairs have on growth, and suggests that we can understand the growth maximization principle as the optimization of nutrient synergies."
} | 303 |
36760734 | PMC9886077 | pmc | 8,253 | {
"abstract": "Soil fungi play indispensable roles in all ecosystems including the recycling of organic matter and interactions with plants, both as symbionts and pathogens. Past observations and experimental manipulations indicate that projected global change effects, including the increase of CO 2 concentration, temperature, change of precipitation and nitrogen (N) deposition, affect fungal species and communities in soils. Although the observed effects depend on the size and duration of change and reflect local conditions, increased N deposition seems to have the most profound effect on fungal communities. The plant-mutualistic fungal guilds – ectomycorrhizal fungi and arbuscular mycorrhizal fungi – appear to be especially responsive to global change factors with N deposition and warming seemingly having the strongest adverse effects. While global change effects on fungal biodiversity seem to be limited, multiple studies demonstrate increases in abundance and dispersal of plant pathogenic fungi. Additionally, ecosystems weakened by global change-induced phenomena, such as drought, are more vulnerable to pathogen outbreaks. The shift from mutualistic fungi to plant pathogens is likely the largest potential threat for the future functioning of natural and managed ecosystems. However, our ability to predict global change effects on fungi is still insufficient and requires further experimental work and long-term observations. Citation: Baldrian P, Bell-Dereske L, Lepinay C, Větrovský T, Kohout P (2022). Fungal communities in soils under global change. Studies in Mycology \n 103 : 1–24. doi: 10.3114/sim.2022.103.01",
"conclusion": "CONCLUSIONS While ongoing climate change has had seemingly no dramatic effects on soil fungal communities, and neither fungal biomass nor fungal diversity in soils appear to be dramatically affected, experiments simulating the main global change effects predict significant shifts in fungal community composition and the share of fungal guilds. The differences in the size of the realised niche of plant-beneficial ECM fungi compared to that of plant pathogens suggests that the fitness of vegetation may decrease as ecosystems experience increased spread of plant pathogens and potentially higher frequencies of outbreaks. This issue is perhaps the one that deserves most attention ( Fig. 1 ). Interestingly, responses of soil fungi to various aspects of global change can be predicted based on different ecological features. While differential responses of ECM fungal species to global changes such as N deposition can be predicted from their extracellular enzymatic capabilities related to organic nitrogen accessibility, response of AM fungal species depends on their differential colonisation traits. Global change effects on ecosystems are highly context dependent and there are undoubtedly ecosystems where changes will be more pronounced. Where global change relieves existing limitations, such as the coldest or N-limited areas, novel limitations will arise, such as increased desertification or induced P-limitation, respectively. Unfortunately, these systems are rarely the subject of research. Experimental manipulations in underexplored systems are thus most welcome. Although the experiments combining multiple factors are relatively frequent ( Yang et al. 2021a ), they are in most cases applying unrealistic treatment intensities and so far too rare to allow generalisations. Since global change factors act in combination and their effects are not simply additive ( Rillig et al. 2019 ), it would be more than welcome to see results of long-term manipulations based on complex predictions of multiple global change factors for given localities. Since it will never be possible to perform manipulations everywhere, long term collection of observational data is needed that would help to describe trends in the soil mycobiome. Global and regional initiatives intending to capture all available types of fungal community data, combined with paired environmental metadata, across time ( Andrew et al. 2017 , Větrovský et al. 2020 ) have the potential to scale our understanding of global change effects on soil fungi to a global level.",
"introduction": "INTRODUCTION Over the past century, CO 2 levels have steadily increased, and global temperatures have risen accordingly. The climate is predicted to continue to change, with increased variability in rain and temperature extremes, both inter- and intra-annually ( IPCC 2014 , Lee et al . 2021 ), and affect the whole biosphere including soils. In addition to the changing climate, it is the change of global atmospheric nitrogen (N) deposition that is perhaps the most threatening global phenomenon. It has increased from 34 Tg N/y in 1860 to 93.6 Tg N/y in 2016 ( Ackerman et al. 2019 ) and is predicted to continue increasing worldwide as the result of human activity. Whether soils will become a source or sink of greenhouse gases under future climate scenarios is difficult to predict due to unclear changes in soil carbon and nitrogen pools, and differences in microbial responses between ecosystems and locations ( Jansson & Hofmockel 2020 ), but there is a justified concern that soils will be heavily affected. Fungi are eukaryotic microorganisms that play multiple fundamental roles related to the future of soil health. As major decomposers of organic matter, mutualists, or pathogens, fungi significantly influence plant health, carbon mineralisation and sequestration, and act as important regulators of the soil carbon balance ( Crowther et al. 2016 ). It is thus important to determine how climate and other global change factors affect future soil fungal communities. The responses of the plant associated guilds to global change factors will likely be of particular interest due to their effects on plant communities. Mycorrhizal fungi act as mutualistic symbionts to plants, providing access to critical nutrients and can ameliorate abiotic stressors associated with climate change, such as heat and drought ( Redman et al. 2002 , Kivlin et al. 2013 ). Plant pathogenic fungi, on the other hand, may opportunistically attack plant hosts that are under stress due to the rapid change in their environment ( Juroszek et al. 2020 , Desaint et al. 2021 ). Therefore, soil fungi, particularly plant associated guilds, mediate the effects of global change on natural vegetation and agricultural crops in multiple ways. In addition to direct effects, climate change can indirectly affect soil fungi through shifts in soil chemistry and vegetation structure ( Tedersoo et al. 2014 , Větrovský et al. 2019 , Zhou et al. 2020 ). It is thus important to understand how global change affects soil fungi. Even though this question has been repeatedly addressed in many contexts and settings in the past, it is still difficult to give a general answer. Soil is the habitat with the highest fungal diversity ( Baldrian et al. 2021 ) and generalisations based on the observed response of individual species are difficult. This high diversity is associated with high levels of functional redundancy in the communities of saprotrophic as well as symbiotic fungi ( Žifčáková et al. 2017 ). Consequently, loss of some species may in theory be replaced by other taxa. However, the critical level of species loss with consequences for ecosystem processes remains largely unknown. Additionally, the diversity, and dependence on plant hosts, of fungal lifestyles ( i.e ., free-living saprotrophs, mutualistic symbionts and plant pathogens) affect fungal species responses to climate change. In this review, we will discuss the links between soils, plants, and fungi to explore the paths by which global change affects fungi and their roles in soils. We will also estimate taxon realised niche space to make predictions about the relative sensitivity of various fungi to global change. Lastly, we will use the accumulated information from experimental manipulations of ecosystems to find general patterns in fungal responses to individual global change factors. For simplicity, we will cover only selected global change processes, namely the increasing CO 2 levels, warming, reduction in precipitation and N deposition ( Fig. 1 ) since these effects are general and long-lasting. While there is a whole suite of other important phenomena linked with global change, such as land use change, biological invasions, increased fire frequency or increased phosphorus (P) input, these factors are very often geographically local or appear at limited temporal scales which makes the predictions of their effects on fungi difficult. This review adds to our knowledge of belowground communities’ responses to global change by focusing on soil fungi, comparing the possible and current responses of plant pathogens to that of mycorrhizal symbionts, leveraging estimates of fungal guilds realised niches to predict their responses, and only synthesising studies that impose realistic global change manipulations. Fungi and their climatic niche Utilisation of the niche concept is one approach to predicting the response of fungi to climate change: if we understand the constraints for fungal life, we can identify and localise the environments where they can live. The concept of the ecological niche provides a framework for understanding resource partitioning by organisms and emergent patterns of coexistence and distribution ( Macarthur & Levins 1967 ). Realised niches define the conditions under which organisms can survive and reproduce in the presence of biotic interactions while fundamental niches are defined in the absence of biotic interactions. While the realised niche can be derived from a species’ distribution and abundance across habitat properties ( Veresoglou et al. 2012 , Davison et al. 2021 ), characterisation of the fundamental niche is more difficult, because it requires experimental investigation of responses to environmental gradients ( Lekberg et al. 2007 ). However, knowing parameters of the fundamental niches of species would be a valuable tool for the prediction of species’ responses to changing abiotic environments. The fundamental niche provides information on species’ potential responses without the influence of biotic interactions, which must also be expected to change along with abiotic changes ( Blois et al. 2013 ). In a global metastudy of soil fungal occurrences using available high-throughput sequencing data, climatic factors contributed, on average, 40–80 % of total explained variability, substantially more than the soil and vegetation properties ( Větrovský et al. 2019 ). Though climatic factors are generally found to be among the most important drivers of global fungal composition, their relative importance varies between studies. For example, Bahram et al. (2018) found that soil carbon-to-nitrogen ratio was the most important driver of fungal abundance, taxonomic and gene composition while Tedersoo et al. (2014) found that soil pH was a major driver of many fungal guilds. Of the climatic factors tested, Větrovský et al. (2019) found that mean temperature of driest quarter, precipitation seasonality, mean temperature of wettest quarter, precipitation of coldest quarter and diurnal temperature range were most often the strongest predictors of individual species distributions. Here we used mean annual temperature (MAT) and mean annual precipitation (MAP) to define species realised niches because these metrics are the most widely used and intuitive defining features of biomes and local climates, are known to affect both soil biota and plants ( Jetz et al. 2012 , Thompson et al. 2017 ) and MAT was identified as the strongest predictor of the local distribution of macrofungi within Norway ( Wollan et al. 2008 ). If we define the breadth of the realised climatic niche as the range of MAT / MAP where 90 % of occurrences are observed, fungal species typically inhabit soils within 5–15 °C difference in MAT and 300–1 200 mm difference in MAP ( Větrovský et al. 2019 ), although niche breadth varies largely among individual taxa ( Fig. 2 ). When we compared the 200 most common soil fungi (taxa occurring in > 99 samples worldwide) based on their membership in ecological guilds, the mean annual temperature at the location of occurrence was lowest for ectomycorrhizal (ECM) fungi followed by ericoid mycorrhizal (ERM) fungi, saprotrophs, and plant pathogens while there was less variation between guilds in the observed mean annual precipitation ( Table 1 ). More importantly, the size of the realised temperature and precipitation niche (the range of MAT and MAP between the first and the ninth decile of all observations) was smaller in ECM fungi than in saprotrophs, ERM fungi, and plant pathogens ( Table 1 ; Fig. 2 ; Větrovský et al. 2019 ). Narrow breadth of the temperature niche in ECM fungi across climatic gradients was also observed within a smaller geographic extent spanning Japan ( Miyamoto et al. 2018 ). Since plant pathogens tend to inhabit warmer areas, and individual species extend both into drier and wetter climates than the ECM fungi ( Fig. 2 ), warming will likely more negatively affect plant-beneficial fungi than plant pathogens ( Větrovský et al. 2019 ). Supporting our prediction of increased soil pathogens, a recent global model of current and projected distributions of plant pathogens showed likely increases in pathogen abundance with MAT predicted to be the major driver ( Delgado-Baquerizo et al. 2020a ). Furthermore, there is evidence that the niches of pathogens may lack trade-offs between biotic and abiotic niche breadths ( Chaloner et al. 2020 ) and may be more labile than that of plant mutualists such as AM fungi ( Bebber & Chaloner 2022 ) suggesting that pathogens may adapt more rapidly to future climates than plant mutualists. It should be noted that the niche concept can be, in theory, extended to other global change factors as well. For example, the response of ectomycorrhizal fungi to nitrogen availability is known for several taxa ( van der Linde et al. 2018 ). However, the limited number of species with reasonable information on their niche breadth, and missing data on local N availability (which exhibits much higher spatial variability than climate), make this concept at present unusable for predicting responses to altered N. Ecological guilds of fungi and global change As already discussed, global surveys of soil fungal occurrences in the GlobalFungi database ( Větrovský et al. 2020 ) show that members of various fungal guilds differ in the size of their climatic niche. Moreover, the level of dependence on vegetation varies from obligate biotrophs to free-living fungi. Due to this, global changes are expected to affect various ecological guilds of soil fungi (ECM fungi, AM fungi, ERM fungi, plant pathogens and saprotrophs) differently, affecting their relative share or community composition. These shifts may subsequently result in changes in various ecosystem processes such as decomposition rate or plant performance. Importantly, climate change-driven shifts in plant communities may lead to shifts in the host availability affecting those fungi that have a narrow host range. With increasing warming, some alpine communities have seen the replacement of forbs with deep rooted grasses ( Liu et al. 2018 ) and increasing nitrogen deposition can lead to reduced species richness though this effect depends on ecosystem characteristics, such as mean annual precipitation ( Clark et al. 2007 ). Altered environmental conditions promote not only natural range shifts of plants species ( Rudgers et al. 2014 ), but also enable naturalisation of alien plant species outside their native distribution range ( Seebens et al. 2015 ). Such events can affect local ecosystems and their fungal components in several ways: by competition for resources, by the introduction of novel fungal species (such as mycorrhizal symbionts or pathogens), or by selective recruitment of root-associating fungal species already present in the local pool by the alien plants ( Rudgers et al. 2020 , Vlk et al. 2020a ). Because of all these factors, changes in local fungal communities are expected as has been already observed for plant introductions ( Vlk et al. 2020b ). Due to the complex effects of N on soil chemistry and vegetation, and the fact that mutualistic mycorrhizal fungi mediate its transfer to plants, change in atmospheric deposition is perhaps the factor with greatest importance for guild composition of soil fungi ( Fig. 1 ). Indeed, nitrogen addition to 25 grasslands distributed across four continents led to the increase of fungal pathogens, although it did not significantly affect AM fungi and saprotrophs. These guild level responses were primarily mediated through nutrient-induced shifts in plant communities ( Lekberg et al. 2021 ). On the other hand, no consistent shifts in guild composition were observed across N-supplemented forests in the USA ( Moore et al. 2021 ). Among the various aspects of global change, changes in climate lead to severe ecosystem alterations. Forests are already facing increasing lengths of heat waves with unprecedented increases of temperature in high latitudes combined with long drought periods. This high level of climate stress likely increases the vulnerability of forests to disturbances including tree dieback and forest fires ( Fig. 1 ; Allen et al. 2010 ). These severe forest disturbances were shown to result in a shift of fungal communities from those dominated by ectomycorrhizal fungi in undisturbed forests to those dominated by saprotrophs in disturbed forests ( Štursová et al. 2014 , Rodriguez-Ramos et al. 2021 ) as a response to changes in primary productivity. \nMycorrhizal plant symbionts\n Geographic distributions of plants with various mycorrhizal symbioses show climate-driven patterns. Temperature-related factors have been found to be the main predictors of the distributions of plant species forming AM, ECM, and ERM symbiosis. Recent models show AM plants to be favoured by warm climates, while dominance of ECM plants (and to some extent ERM plants) is more favoured by colder climates ( Barcelo et al. 2019 ). Ectomycorrhizal symbiosis dominates forests in which seasonally cold and dry climates inhibit decomposition and is the predominant form of symbiosis at high latitudes and elevation. AM trees dominate in grasslands and the warm-and-wet climates of tropical forests where enhance decomposition is typical ( Steidinger et al. 2019 ). Warming can significantly alter the distribution of mycorrhizal host plants, with likely subsequent impacts on the proportion of various guilds of mycorrhizal fungi. In addition to warm climates, AM fungal colonisation has been found to be strongly related to soil carbon-to-nitrogen ratio and highest at sites featuring continental climates with mild summers and a high availability of soil nitrogen ( Soudzilovskaia et al. 2015 ). In contrast, the intensity of ectomycorrhizal infection in plant roots maybe more related to soil acidity, soil carbon-to-nitrogen ratio and seasonality of precipitation and is highest at sites with acidic soils and relatively constant precipitation levels ( Soudzilovskaia et al. 2015 ). As such, root colonisation by both guilds is predicted to respond to climatic factors and N deposition. AM fungi primarily rely on inorganic forms of N ( Phillips et al. 2013 ) or small organic N compounds ( Whiteside et al. 2012 ). In contrast, some ECM fungi are thought to rely more heavily on organic N sources ( Phillips et al. 2013 ), having a greater capacity to invest in N-degrading extracellular enzymes that access complex organic forms of N in soil, such as proteins and chitin ( Fernandez & Kennedy 2016 ). ECM fungi are thus more associated with slower decomposition of soil organic matter and increased soil carbon (C) storage ( Averill et al. 2014 , Averill & Hawkes 2016 , Fernandez & Kennedy 2016 ), potentially by competing with free-living soil microbes for organic N resources. These distinctions between AM and ECM fungi lead to two important predictions: (a) that inorganic N inputs to ecosystems will favour AM-associated trees at the expense of ECM-associated trees, and (b) that inorganic N-driven declines in ECM fungal abundance will reduce the belowground C storage capacity of the forest biome ( Fig. 1 ). Indeed, recent nitrogen deposition across USA favoured the expansion of AM trees at the expense of ectomycorrhizal trees, and was spatially correlated with reduced soil carbon stocks ( Jo et al . 2019 ). This implies that future changes in nitrogen deposition may further turn the balance between AM and ECM fungi in forest ecosystems ( Averill et al. 2018 ). \nEctomycorrhizal fungi\n Despite the potential for climate change driven replacement of ECM with AM trees, most ecosystems are dominated by either ECM plant symbionts (in most temperate and boreal forests worldwide) or AM symbionts (in natural grasslands, croplands and tropical forests). Therefore, relative abundance of each guild or the change of within-guild species composition are the most likely responses. While shifts in dominant mycorrhizal type mediated by global changes will likely result in changes in nutrient cycles and soil carbon storage, consequences of potential shifts of within guild species composition are less clear. Based on the assessment of present climatic drivers of ECM fungal distribution, under future climate scenarios North American Pinaceae forests are predicted to see as high as 26 % declines in ECM fungal species richness within 50 years, although there is a high level of regional variation ( Steidinger et al. 2020 ). Furthermore, ECM fungal diversity across Japan was also demonstrated to significantly decrease with MAT ( Miyamoto et al. 2018 ), suggesting potential decreases with warming. The observation of the ECM fungal community shift on Betula papyrifera and Abies balsamea saplings in a warming experiment ( Fernandez et al. 2017 ) suggests that warming may change the future composition of the ECM fungal subcommunity. Since N supply to plants is one of the major roles of ECM fungi, N deposition likely affects ECM fungal communities. With increasing nitrogen availability, fungi that obtain nitrogen from complex soil organic sources using metabolically costly pathways – e.g ., Cortinarius , Piloderma and Tricholoma – are likely at a disadvantage compared to fungi that use inorganic nitrogen, such as Elaphomyces or Laccaria ( Lilleskov et al. 2011 ). In a large survey of ECM fungi associated with forest trees in Europe, several ECM fungi responded to N throughfall deposition. Fungi that use organic nitrogen tended to be negative indicators for nitrogen deposition, while fungi that use inorganic nitrogen tended to be positive indicators. Conifer specialists – particularly those with abundant hyphae and rhizomorphs – were more negatively affected by increasing nitrogen than generalists and broad-leaf specialists ( van der Linde et al. 2018 ). In the future, N deposition will likely affect ECM fungi and promote shifts from nitrophobic species ( e.g ., Russula vinosa, Lactarius rufus ) to nitrophilic species ( e.g ., Scleroderma citrinum, Amanita rubescens, Russula ochroleuca ) ( Fig. 1 ; van der Linde et al. 2018 ). In theory, mutualistic fungi could accompany host plants in climate-induced migration ( Rudgers et al. 2020 ). In a study of the upward migration of tree individuals above the tree line, low ECM diversity was observed in the roots of migrating trees indicating that the altitudinal shift in the ECM fungal community lags behind climate-driven tree migration. ECM fungal dispersal limitation is thus an important factor controlling this process and possibly retarding vegetation shifts ( Alvarez-Garrido et al. 2019 ). Similar conclusions were found in a study of invasive pines that clearly showed plant invasions can be limited by the dispersal of ECM fungi ( Nunez et al. 2009 ). \nArbuscular mycorrhizal fungi\n Similar to ECM fungi, AM fungi also fully depend on their symbiotic host plants as a sole source of carbon ( Tisserant et al. 2013 ) and therefore any environmental shifts may affect abundance, species richness and AM fungal community composition directly as well as indirectly by altering their host plants. A recent review of the response of AM fungal species richness and community composition to various aspects of global change found that elevated CO 2 will likely have no effect on AM fungal richness, and responses to N deposition, warming, and changed precipitation will likely be highly context dependent ( Cotton 2018 ). The effects of the above-mentioned extrinsic factors associated with global change are translated into community composition of AM fungi via differential responses of each species, which are determined by their intrinsic characteristics, such as specific growth patterns, morphology or anatomy. AM fungi greatly vary in root colonisation traits such as extent and structure ( Klironomos & Hart 2002 ), and soil hyphal traits such as extent, density and structure ( Powell et al. 2009 ). Interestingly, the increase of CO 2 concentration, as well as increases in N availability, leads to lower relative abundance of AM fungal taxa from the Gigasporaceae and Diversisporaceae families, which produce high levels of extraradical mycelia, while relative abundance of the Glomeraceae taxa, which are characterised by extensive intraradical colonisation, tend to increase ( Cotton 2018 ). This shift in community traits suggests lower investments in potentially costly nutrient acquisition traits with increasing nutrient availability. The community level responses to environmental conditions combined with various intrinsic characteristics indicate that niche optima and niche width may differ among the species of AM fungi. Large sampling campaigns, enabled by an onset of high-throughput sequencing methods, provide sufficient data to model parameters of species ecological niches. While Acaulosporaceae has a realised niche optima in low temperature conditions, Gigasporaceae has a realised niche optima in high temperature and high precipitation conditions ( Davison et al. 2021 ). Additionally, the width of the AM fungal temperature niche appears to be limiting, seeming to be narrower than in other fungal guilds ( Větrovský et al. 2019 , Davison et al. 2021 ). These findings indicate that changes of MAT and MAP can particularly affect the composition of AM fungal communities. Contrary to diversity, the abundance of AM fungi seems to be more consistently affected by changes in N availability and shifts in CO 2 concentration. While the majority of studies report a decrease in AM fungal abundance with enhanced nitrogen ( e.g., \n Shen et al. 2014 , Chen et al. 2017 , Treseder et al. 2018 , Zhang et al. 2018 , Han et al. 2020 , Jia et al. 2020a , Ma et al. 2021a ), a few found no effect ( Lilleskov et al. 2019 , Karst et al. 2021 ). The addition of N can benefit AM fungi if it exacerbates plant P limitation ( Johnson 2010 ), but may be suppressive if nitrophilic, ruderal plants replace plants that allocate more C to AM fungi ( Isbell et al. 2013 ). Thus, the responses likely depend on the extent to which nutrient addition alleviates plant deficiencies and alters plant communities. A meta-analysis examining the global effects of nutrient enrichment on AM fungal and plant diversity showed that AM fungal diversity, rate of root colonisation, and extraradical biomass typically decreased with N addition, while spore abundance and hyphal length were unaffected. These results were consistent among forests, grasslands, and agro-ecosystems ( Ma et al. 2021a ). The short-term fertilisation effect of elevated CO 2 concentrations mostly stimulated AMF abundance ( e.g., \n Treseder 2004 , Antoninka et al. 2011 , Zavalloni et al. 2012 , Sun et al. 2017 , Dong et al. 2018 ). Importantly, while stimulation of AM fungal abundance with increased CO 2 is expected, considering that plant productivity depends on nutrient supply by AM fungi, the increase of temperature and shifts in precipitation will likely affect AM fungal abundance thanks to a greater climate niche partitioning of AM fungi. \nPlant pathogens\n Analyses of fungal guild niche breadth indicates that plant pathogens may better cope with climate change than other fungal guilds ( Chaloner et al. 2020 ). Conditions that affect pathogen overwintering and dispersal are of essential importance due to pathogen lifestyles, survival in soils, and outbreaks triggered by climatic and plant host signals. Global warming in areas with seasonal temperature variation has increased pathogen survival during winters and increased the length of vegetation seasons leading to faster pathogen spread or stronger outbreaks ( Harvell et al. 2002 ). As an ongoing consequence of warming, movement of crop pests to higher latitudes has already been observed. Since the 1960s, fungal crop pests were observed to move polewards at a pace of some 5 km/y, more rapidly than most other crop pests ( Bebber et al. 2013 ). Warming appears to be the most important driver of plant pathogen abundance. Climatic factors, especially the MAT and precipitation seasonality were the most important predictors of the relative abundance of plant pathogens across 235 global sites. Under future climate change and land-use scenarios, relative abundance of plant pathogens is predicted to increase ( Delgado-Baquerizo et al. 2020a ). A nine-year warming experiment in a dryland on the Iberian peninsula showed higher relative share of pathogens, higher relative abundance of Alternaria and higher absolute abundance of Alternaria in warmed plots ( Delgado-Baquerizo et al. 2020a ). While the increase in relative abundance, or sporulation, of plant pathogens may increase the risk of a disease outbreak, direct causal links may be difficult to find. It is possible that negative responses of mycorrhizal fungi and neutral or positive responses of pathogens to climate change can subsequently manifest in negative responses of vegetation. More importantly, climatic events seem to be predictive factors of fungal disease outbreaks with high humidity and high temperature being the most common factors ( Romero et al. 2022 ). Pathogens may also use the opportunity to attack weakened host communities such as forest ecosystems after dieback caused by drought or heat stress ( Fig. 1 ; Anderegg et al. 2013 ). In natural systems, pathogens appear to be more abundant in resource-rich environments ( Reynolds et al. 2003 , Revillini et al. 2016 ), and nutrient addition ( e.g . fertilisation) has been linked to increased disease incidence in plants ( Walters & Bingham 2007 , Veresoglou et al. 2013 ) which may increase the risk of pathogen spread or outbreaks at elevated atmospheric N deposition. The effect of CO 2 increase on pathogens is less clear, however, concentrations of spores of several pathogens were increased by elevated atmospheric CO 2 (eCO 2 ) in a Populus tremuloides plantation in air and litter. Although the responses of fungi were not uniform, significant increases were found in the potential pathogenic genera Alternaria , Cladosporium and Fusarium ( Klironomos et al. 1997 ). Plant pathogen community composition may not intrinsically affect ecosystems because it is often individual taxa that cause disease outbreaks. The effects of global change on individual plant pathogen taxa may thus be more important than the guild-level effects. Based on historical observations of higher Alternaria spp. spore concentrations at warm temperatures, spore concentrations are predicted to increase with warming in the United Kingdom ( Maya-Manzano et al. 2016 ) and future climate models suggest increased prevalence of Alternaria brassicae in North Germany ( Siebold & Tiedemann 2012 ). In several instances, eCO 2 increased spore production by Alternaria spp. several-fold ( Klironomos et al. 1997 , Wolf et al. 2010 ). Considering disease severity, both warming and eCO 2 has been shown to increase Alternaria leaf spot severity on rocket, cauliflower and cabbage ( Pugliese et al. 2012 , Siciliano et al. 2017 ). To conclude, while differential response of ECM fungal species to global changes such as N deposition can be predicted from their extracellular enzymatic capabilities related to organic nitrogen accessibility, response of AM fungi depends on their differential colonisation traits. Species traits of saprotrophs or pathogens related to their response to global changes are much less clear and therefore predictions of global change effects on these two guilds are much more difficult. Fungal response to global change factors and lessons learned from manipulated studies Our present understanding of the response of fungi to global change is based on several lines of support: (1) ecological theory and the predictions based on the known niches of fungal species, (2) predictions of responses to indirect factors affected by global change, such as the change of soil chemistry, vegetation composition, or ecosystem productivity, (3) extrapolation of observations of changes in fungal communities across time and space, (4) experimental simulation of future conditions and the analysis of fungal response. Since there is a lack of long-term observations on soil fungi under conditions of real-time climate change and the extrapolation of such observations may be problematic, experimental manipulations simulating global change factors appear to be the best tool to predict the future of soil fungi. Experimental approaches have several limitations that must be considered when interpreting results. Each of the experiments has at least three important aspects that affect the observations: (1) the duration of treatment, (2) the intensity of manipulation, and (3) the local conditions. Over the duration of treatment, several components of the system respond so that direct, and/or indirect, effects change in time and adaptations emerge. The plant communities likely respond first with altered productivity, while change in composition comes later ( Smith et al. 2009 ). Importantly, the effects of short-term warming and/or precipitation experiments can be eclipsed by site specific year-to-year variation in climatic conditions. The intensity of manipulation is another critical issue. In many experiments, especially those simulating N deposition, the magnitude of treatments is considerably larger than those predicted by current models. Equally important, the target biome and local condition at the experimental sites can interact with the global change treatments. Moreover, soil fungi as the responding community are extremely diverse in terms of alpha and beta diversity ( Baldrian et al. 2021 ) which limits the cross-ecosystem interpretation of community effects. Unfortunately, the experimental results reported so far show high levels of geographic bias with most studies in forests and grasslands of the temperate zone ( Tables 2 – 5 ). These biases in sampling mean that surprising results from underexplored biomes, such as massive CO 2 fluxes from warmed plots recorded in the Panama tropical rainforest, cannot be ignored. Such fluxes largely exceeded model predictions and indicated high sensitivity of local soil C stocks to warming ( Nottingham et al. 2020 ). Here, we review the results from experimental simulations of climate change factors 1) elevated CO 2 , 2) warming, 3) reduction of precipitation, and 4) increased N deposition ( Fig. 1 ). We ended up with 138 studies that applied realistic treatment types and levels (see each section) and reported at least one of the below response variables ( Supp. S1 ). Though our survey is not exhaustive, we believe it is representative of the current state of knowledge. We decided to focus on the commonly studied fungal responses biomass, diversity, guild share, and changes in community composition. Though these responses are interconnected ( e.g. , changes in fungal diversity will likely lead to changes in composition), we decided to survey all factors to highlight the current focuses of research into the responses of fungi to climate change factors. The analyses of diversity, guild share, and changes in community composition largely rely on meta-barcoding sequencing, which we recognise as suffering from biases such as primer bias and the use of relative abundances ( Quinn et al. 2018 , Alteio et al. 2021 ), it is still the best tool for understanding fungal communities ( Nilsson et al. 2018 ). All recorded responses are taken directly from the results sections and therefore represent current interests in the field. \nIncrease of CO 2 concentration\n Elevated CO 2 partially underlies global increases in plant productivity ( Nemani et al. 2003 ). Furthermore, experimentally elevated atmospheric CO 2 concentrations (eCO 2 ) have led to short term increases in plant biomass production, allocation of carbon to roots and to soil ( Adair et al. 2011 ) and consequently soil respiration. The higher C allocation belowground can fuel the breakdown of labile organic matter by copiotrophic microorganisms. Therefore, microbial biomass and heterotrophic respiration will likely increase ( Fig. 1 ; Naylor et al. 2020 ). At longer time scales, eCO 2 has been shown to increase microbial decomposition of soil organic matter (SOM) through priming ( van Groenigen et al. 2014 ). Direct effects on individual fungi are unlikely since CO 2 concentration in soil pores is higher than in the atmosphere and varies in space and time. Furthermore, eCO 2 may affect fungal propagation and dispersal. Under an 2×-ambient CO 2 treatment in a Populus tremuloides plantation, the concentration of airborne fungal propagules, mostly spores, increased fourfold. Analysis of decomposing leaf litter (likely the main source of airborne fungal propagules) indicated that fungi produced fivefold more spores ( Klironomos et al. 1997 ). Furthermore, increased total sporocarp biomass was observed in an eCO 2 experiment ( Andrew & Lilleskov 2009 ). Since fruiting and sporulation is the main mode of dispersal of soil fungi, consequences of this observation – if confirmed in additional systems – may be important. Across the studies we surveyed, eCO 2 experiments report either no change or increased biomass and diversity of all fungi, and only single cases of reduced AM fungal diversity and change in guild composition. Most experiments report change in the fungal community composition but there were no consistent observations of enriched or suppressed taxa ( Fig. 3 , Table 2 ). Though we found no clear relationship between fungal responsiveness and experimental length ( Figs 3 , 4 ), a meta-analysis of 11 studies found a positive relationship between increased fungal richness due to eCO 2 and experimental length ( Veresoglou et al. 2016 ). A recent global meta-analysis found no relationship between experimental length and the responsiveness of fungal biomass, but found that eCO 2 decreased the F/B ratio across 31 studies ( Sun et al. 2021 ). In our survey, the longest experiments showed contrasting effects on soil chemistry. A forest-based experiment reported significant decreases in pH, organic matter content, and P and increased water content ( Weber et al. 2013 ) which may all potentially affect fungi. However, a grassland experiment of a similar length reported no significant change in soil chemistry ( Maček et al. 2019 ). \nWarming\n In agreement with the increasing catalytic performance of soil enzymes with increasing temperature ( Baldrian et al. 2013 ), C turnover across global biomes has been shown to increase with temperature ( Carvalhais et al. 2014 ). Temperature sensitivity of soil C loss appears higher in cold regions ( Crowther et al. 2016 , Koven et al. 2017 ) and probably the most extreme response is expected in the permafrost where thawing dramatically increases organic matter transformation and the emissions of CO 2 and CH 4 ( Jansson & Tas 2014 ). The expected C losses are large since the soils in cold regions host large C stocks ( Crowther et al. 2019 , García-Palacios et al. 2021 ). Additionally, warming has led to the loss of plant species unable to tolerate new environmental conditions ( Freeman et al. 2018 ) or outcompeted by invaders better adapted to the new conditions ( Alexander et al. 2015 ). These shifts in plant species composition may alter the quality of the carbon input into the system ( Harte et al. 2015 ). Shifts in fungal saprotroph communities in response to both increased access to extant carbon and novel carbon inputs will have important implications for global responses to climate change ( García-Palacios et al . 2021 ). The responses of soil fungal communities to warming likely depends on the local climatic conditions, such as MAT. Not surprisingly, in the Antarctic, at the lower limit of fungal temperature tolerance, air temperature is the strongest and most consistent predictor of soil fungal diversity and, with current rates of warming, a 30 % increase in fungal diversity is predicted by 2100 ( Newsham et al. 2016 ). However, this diversity response to warming is probably not universal since the highest level of fungal diversity is predicted in cold areas ( Větrovský et al. 2019 ). Similar to soil fungi, the highest diversity of bacteria in global surveys has also been observed at locations with relatively low MAT (around 10 °C; Thompson et al. 2017 ) and temperate regions ( Bahram et al. 2018 ), although bacterial biomass in soils does not seem to be affected by warming ( Lladó et al. 2017 ). Short-term and prolonged warming may have differing effects. An initial loss of labile soil carbon in one of the longest running warming experiments in the Harvard Forest was later followed by increased degradation of more recalcitrant carbon compounds. Sustained warming for 26 years resulted in the depletion of soil organic carbon (SOC) with corresponding reductions in microbial biomass ( Melillo et al. 2017 ). Based on a meta-analysis, warming initially increases soil respiration, but the magnitude of observed effect declines significantly as warming progresses and in fact, after 10 years of warming, soil respiration in experimentally warmed plots was similar to controls. Microbial acclimation, community shifts, adaptation, or reductions in labile C may ameliorate warming effects on soil respiration in the long-term. Accordingly, long-term soil C losses might be smaller than those suggested by short-term warming studies. The share of experiments where fungal biomass increased versus decreased with warming have been found to be roughly equivalent and no significant change in the fungal to bacterial (F/B) biomass ratio were observed across studies ( Romero-Olivares et al. 2017 ). The F/B ratio was also unaffected after 7–25 yr of warming across 12 experiments in the Alpine and Arctic tundra ( Jeanbille et al. 2021 ). Temperature also alters fungal fruiting with consequences for dispersal. Across Europe, timing of fruiting has been shown to vary by 25 d among latitudes and 30 d among altitudes suggesting a strong temperature effect ( Andrew et al. 2018 ). Present-day autumn fruiting of fungi has been shown to occur later than in the past, and the fruiting season length has increased, similar to the vegetation season ( Kauserud et al. 2012 ). There has also been shown to be a significant shift in fruiting of saprotrophic and ectomycorrhizal fungi towards higher altitudes in the Swiss Alps between 1960 and 2010 as a consequence of warming ( Diez et al. 2020 ). Warming was the most frequently applied treatment in our survey (47 % of studies) and as such gives the best opportunity for generalisations. Importantly, warming was most frequently reported to alter total fungal biomass and a substantial fraction of the observations indicate negative effects, especially between 3–5 yr of application. In longer-lasting experiments, however, the effects on fungal biomass were less pronounced and AM fungi seem to be even less affected. Both negative and positive effects on total fungal diversity were reported but no effects were reported for experiments running for more than three years; furthermore, the decrease of AM fungal diversity was also observed only in the short term ( Figs 3 , 4 , Table 3 ). Many individual experiments reported significant effects on fungal guild composition, which were, however, context-dependent. The only exception is the effect on plant pathogens where all reports showed their increase ( Table 3 ). Most warming experiments also reported change in fungal community composition, often within the ectomycorrhizal guild ( Fernandez et al. 2017 , van Nuland et al. 2020 ) and a decrease of the Glomeraceae was recorded within the AM fungi ( Cao et al. 2020a , b ). Interestingly, almost all studies with experimental lengths longer than 10 yr or any experimental length with warming treatments larger than 2 °C reported significant changes in fungal community composition. In partial support of our survey, a recent global meta-analysis found that warming decreased fungal richness but that there was no significant effect of experimental length on this response ( Li et al. 2022 ). There were no reports of important changes in soil nutrient content or pH but some of the long-term experiments report the decrease of the F/B biomass ratio ( Gutknecht et al. 2012 ) and lower transcription of hydrolytic enzymes ( Romero-Olivares et al. 2019 ), two factors that may be connected since fungi are important producers of enzymes in soils ( Starke et al. 2021 ). \nReduction of precipitation\n Since soil C turnover across global biomes increases with precipitation ( Carvalhais et al. 2014 ), any change in precipitation likely affects C cycling. Responses of plant communities to increased variability in precipitation have ranged from high ecosystem stability in the face of intra-annual variability ( Jones et al. 2016 ) to increasing functional diversity with increased inter-annual variability ( Gherardi & Sala 2015 ). Even when there is very little recorded change in plant community diversity, significant changes in species composition through reordering have been recorded ( Jones et al. 2017 ). While climate models predict both decreases and increases in precipitation across global locations ( IPCC 2014 ), drought effects on ecosystems are likely much more dramatic. Increases in the durations of drought are expected to be a major consequence of future climate and increased desertification is predicted for most semi-arid or arid regions in the coming decades ( Huang et al. 2016 ). Based on a recent meta-analysis, terrestrial ecosystem productivity was decreased by drought across all ecosystems ( Wang et al. 2021a ). The response of productivity to drought are more pronounced with higher drought intensity and longer duration, and consistent across biomes and climates. Drought can significantly decrease soil moisture, soil C content, soil C:N ratios, and microbial biomass C, whereas it tends to increase soil pH. The relative proportion of fungal biomass (F/B ratio) however, frequently increases with drought ( Delgado-Baquerizo et al. 2020b , Wang et al. 2021a ). The diversity and abundance of soil bacteria and fungi have been shown to decrease in drylands as aridity increased, being largely driven by the negative impacts of aridity on soil organic carbon content ( Maestre et al. 2015 ). Since most global change models predict changes in precipitation, experimental manipulations of precipitation are relatively frequent. Unfortunately, such manipulations are highly diverse and range from reduction and addition to redistribution. Both reduction and addition are frequently combined often without a clear link to a model prediction for the ecosystem under study ( Knapp et al. 2015 ). Moreover, many experiments use manipulations that are likely outside the model predictions with reductions or additions > 50 % and the relevance of such manipulations is thus unclear. For simplicity, we surveyed the effects of precipitation reduction since drought seemed to have more profound ecosystem consequences ( Table 4 ). In our survey, no negative effects of precipitation on total fungal biomass were reported with most experiments reporting no effect on any response variable ( Fig. 3 , Table 4 ). In studies of AMF, there was decreased hyphal, spore density, and root colonisation in a forest system in connection with soil acidification ( Maitra et al. 2019 ) and a reduction in root colonisation in a perennial cropping system ( Emery et al. 2022 ). Reduction of precipitation most frequently did not affect the diversity of fungi and AM fungi and decrease of total fungal diversity was never observed in manipulations lasting three or more years ( Figs 3 , 4 , Table 4 ). Contrary to our survey, a recent global meta-analysis found that precipitation reduction led to increased fungal richness with the effect size increasing with experimental length, though precipitation reduction had no effect on fungal diversity ( Li et al. 2022 ). Importantly, precipitation reduction typically shifted the share of fungal guilds with the reduction of AM fungi and increase of plant pathogens being frequently reported. Changes in fungal community composition were also relatively frequent ( Table 4 ). Increase of the F/B ratio was observed in a heathland experiment ( Haugwitz et al. 2014 ). Changes in soil chemistry were typically not found, not even for the long-lasting experiments. \nIncreased atmospheric N deposition\n Many plant communities are N limited ( LeBauer & Treseder 2008 ), and additional N can thus promote plant productivity if P content is non-limiting ( Fay et al. 2015 ). Additionally, N deposition may reduce plant species richness though this effect depends on ecosystem characteristics, such as MAP ( Clark et al. 2007 ). For example, N addition may increase plant species richness in ecosystems with high MAP ( Komatsu et al. 2019 ). In addition to effects on vegetation, N has multiple effects on soil chemistry, including acidification ( Lekberg et al. 2021 ). Though a recent global meta-analysis found that N reduced overall soil fungal richness ( Zhou et al. 2020 ), the effects of N deposition on soil fungi can be, like plant community responses, context dependent. Across N-addition studies in the US forests, fungal biomass and richness increased with simulated N deposition at sites with low ambient deposition but decreased at sites with high ambient deposition ( Moore et al. 2021 ). Along local fertility gradients, total fungal biomass was highest in soils with the lowest nutrient availability and tree productivity ( Nilsson et al. 2005 ). Higher N availability promotes bacterial growth due to their higher N demand. Especially in the N-limited boreal soils, N addition results in a decrease of the F/B ratio by 25–70 % ( Frey et al. 2004 , Wallenstein et al. 2006 , Maaroufi et al. 2015 ). There appears to be a general consensus that N deposition increases soil C sequestration due to the decline in SOM decomposition via the reduction of fungal abundance and decomposer activity in many different soil environments, including temperate and boreal forests ( Frey et al. 2014 , Maaroufi et al. 2015 ). Since, similar to plants, many fungi respond to P availability in soil and it is an important driver of fungal abundance in soils without N limitations ( Odriozola et al. 2021 ), increased N content may act on fungal productivity and community composition indirectly through P limitation ( Fig. 1 ). Within our survey, the goal of the majority of N addition experiments was to simulate predicted increases in atmospheric deposition, but many used unreasonably high amounts of fertilizer, ignored ambient N deposition rates, and virtually none of them referenced a model that predicts future deposition, whose extent shows high local variation. It is currently estimated that the vast majority of forests are subject to total N deposition lower than 25 kg N/ha/y ( Schwede et al. 2018 ) and it is unrealistic to expect that the increase in future is several-fold. We have thus considered only the results of experiments where N addition was lower than 75 kg N/ha/y. The effects of N addition on fungal biomass in soil were variable. For AM fungi, decreased spore density, root colonisation, and biomass were much more frequent than positive effects ( Fig. 4 ). In forest ecosystems, decrease of fungal biomass and root colonisation appears typical ( Ma et al. 2021b ). Both increases and decreases in diversity of fungi or AM fungi were observed ( Figs 3 , 4 , Table 5 ). This lack of consistency in diversity responses is somewhat supported by the effects of increased N on fungal richness varying between global meta-analyses with increased N either decreasing richness or having no effect ( Zhou et al. 2020 , Li et al. 2022 ). Changes in the representation of fungal guilds were a common consequence of N addition. In most long-term N addition experiments, the share of ECM fungi was significantly reduced ( Table 5 ) with a shift to nitrophilic taxa such as Rusula vinacea ( Morrison et al. 2016 , Tahovská et al. 2020 ). The consequences of longer N enrichment (> 4 yr) were relatively complex and include acidification and increased N availability ( Choma et al. 2017 , Wang et al. 2021b ), decreased F/B ratio ( Gutknecht et al. 2012 , Wang et al. 2015 ) and decreased activity of enzymes decomposing recalcitrant plant biopolymers lignin and cellulose ( Freedman et al. 2015 , Hesse et al. 2015 ). Although vegetation responds to N addition as well, the change of soil chemistry appeared to be the immediate driver of fungal community composition ( Zheng et al. 2014 , Zhou et al. 2020 , Wang et al. 2021b ). \nCombined effects and model predictions\n Current models predict that the effects of global change factors will act simultaneously in most terrestrial habitats and the resulting effect of global change thus reflects their combination. Furthermore, shifts in plant community composition are likely determined by interactions between multiple climate change drivers ( Avolio et al. 2021 ). Between 1990 and 2014, global heterotrophic soil respiration and its ratio to total soil respiration increased, probably in response to the combined effects of global change factors ( Bond-Lamberty et al. 2018 ). This suggests that climate-driven losses of soil carbon are currently occurring across many ecosystems, with a detectable and sustained trend emerging at the global scale, although the underlying mechanisms cannot be easily identified. Simulation of the global change effects until the year 2090 using available data from 1950 indicates that climate change acts mostly indirectly, through other environmental variables, e.g ., changes in the soil pH ( Guerra et al. 2021 ). The effects of global change factors on fungi thus may depend either on the relative importance of each individual factor under local conditions or on the combined effects of multiple factors."
} | 13,934 |
33535536 | PMC7867074 | pmc | 8,255 | {
"abstract": "As the need for non-renewable sources such as fossil fuels has increased during the last few decades, the search for sustainable and renewable alternative sources has gained growing interest. Enzymatic hydrolysis in bioethanol production presents an important step, where sugars that are fermented are obtained in the final fermentation process. In the process of enzymatic hydrolysis, more and more new effective enzymes are being researched to ensure a more cost-effective process. There are many different enzyme strategies implemented in hydrolysis protocols, where different lignocellulosic biomass, such as wood feedstocks, different agricultural wastes, and marine algae are being used as substrates for an efficient bioethanol production. This review investigates the very recent enzymatic hydrolysis pathways in bioethanol production from lignocellulosic biomass.",
"conclusion": "7. Conclusions and Future Perspectives Currently, the enzymatic hydrolysis process is still a narrow way to the efficient production of bioethanol because of the high cost of many enzymes as well as the inhibitory properties of compounds that reduce the efficiency of the glucose production. Further research and new protocols are required to increase cellulose to glucose conversions by finding suitable lignocellulosic biomass structures that can improve bioethanol production. However, the main obstacle is the complex structure of lignocellulosic materials, which are the crystallinity of cellulose and issues that are related to lignin. All those properties make the enzymatic hydrolysis a challenging process. Therefore, suitable pretreatment protocols must be developed that can increase the efficiency of enzymatic activity, which can improve involving substrates for cellulolytic enzymes. Limitations are present in all steps of the process: pretreatment, enzymatic hydrolysis, fermentation, and saccharification processes need considerable new research strategies to improve the economics and efficiency of the process. Enzymatic hydrolysis is a cost-efficient process that increases the value of new by-products, derived from conversion, that are microbiologically-wise safe ingredients in food or products, as well as have increased nutritional and functional value. In addition, enzymatic hydrolysis still has many possibilities for improving enzyme production, its recycling, as well as genetic screening. While current biomass utilization has lignin being used for powering process energy necessities, it also gives lignin new possibilities in the industry of biorefineries. On the other hand, replacing different substrates with lignocellulosic and algal biomass is a step forward in using renewable sources, which reduce the current demands for food crops. Moving to the fourth generation of bioethanol production using cultivated algae will provide improvements that will benefit both environment and the production industries, which can promote more economical strategies for bioethanol production. However, there are numerous variables that affect the efficiency of conversion, such as source of biomass used for the production of a biofuel, pretreatment method, source of the enzyme, and its mixture used in the enzymatic hydrolysis. All of these features must be taken into account when designing the lignocellulosic conversion process to optimize its conditions.",
"introduction": "1. Introduction The over-exploitation of our planet’s resources has worsened our environment, which is nowadays suffering from climate change more than ever. Elevated gas emissions, the greenhouse effect, and global warming have all contributed to the search for renewable sources, which are in harmony with world’s energy needs. Lignocellulosic biomass is a sustainable alternative that produces new-generation bio-based chemicals, such as biofuels, food additives, enzymes, and others [ 1 , 2 , 3 ]. Lignocellulosic biomass includes all kinds of agricultural wastes, forestry residues, and feedstocks, as well as marine algae, and it can be provided on a large-scale platform from all kinds of materials [ 4 , 5 ]. In general, lignocellulosic biomasses consist of lignin, cellulose, and hemicelluloses, some organic extracts and inorganic components, which are turned into ash after combustion. All those components make lignocellulosic biomass a complex group of polymers that are naturally recalcitrant to enzymatic conversion. Lignocellulosic biomass materials are constituted of renewable substrates used for bioethanol production, where such materials play a role in contributing to environmental sustainability [ 6 ]. Lignocellulosic biomass consists mostly of polymer sugars (celluloses and hemicelluloses) and lignin [ 7 , 8 ]. It can be broken down into simple sugars by enzymatic hydrolysis or chemically by sulfuric or other acids [ 9 ]. Due to the process that requires less energy in mild conditions, enzymatic hydrolysis is becoming a more suitable pathway in biomass hydrolysis [ 10 ]. It is an important step in converting cellulose to glucose in pretreated biomass, which is carried out by cellulose enzymes in temperature range from 40 to 50 °C, with a pH range from 4 to 5 [ 11 ]. The degree of pretreated biomass, such as lignin removal, enzyme loading, and duration of hydrolysis is highly dependent on the enzymatic hydrolysis efficiency, since the process is also highly affected by cellulose crystalline structure [ 12 ]. The enhancement of the enzymatic hydrolysis process is possible by adding non-ionic surfactants, which can change the surface properties of cellulose, as well as reduce enzyme loading. Such non-ionic surfactant is found to be polyethylene glycol (PEG), which can reportedly increase the convertability of lignocellulosic biomass for more than 30% [ 11 , 13 , 14 ]. Biofuels based on biomass have many advantages over fossil fuels: besides contributing to fuel diversity, different biofuels are accessible by different common biomass sources, have an environmentally friendly impact and potential, and provide many benefits in terms of economy and environment for all users of biofuels. Such biofuels are biodegradable and immensely contribute to sustainability. In addition, biofuels add value to migrating greenhouse gas (GHG) emissions, which provide a cleaner and more sustainable energy source with reduced air pollution. By using biomass feedstocks for bioethanol production, such actions of biomass usage enable the emerging development of rural areas in different countries, as well as increase of agricultural income. Such developing countries have more available land with favorable climate conditions and therefore minimum or at least lower labor costs. Another advantage with large-scale biofuel production for developing countries is the reduction of its oil import dependence, which contributes to international competitiveness. When referring to the production of “good” bioethanol (bioethanol being the most commonly used biofuel for transportation worldwide) in terms of reducing the GHG emissions, it is important to replace polluting fossil fuels with more environmentally friendly lignocellulosic biomass. To ensure beneficial properties of biofuels, all kinds of by-products in the production process should be properly and efficiently utilized in order to minimize the GHG effect, as well as maximize their energy. In addition, emissions (such as carbon dioxide and nitrous oxide) should be kept to a minimum in terms of pollution and fertilizers, respectively. Moreover, biofuels such as bioethanol can help reduce the carbon dioxide escalation by replacing the fossil fuels and recycling the carbon dioxide being released when combusted as fuel [ 15 ]. However, in the ever-growing biofuel industry, sustainable energy systems and energy efficiency have an important decisive part, especially when renewable energy potentials compete with high energy demands. Many different sustainability assessments have been performed over the years, describing various schemes, such as emergy, exergy, techno-economic analysis, energy accounting, and life cycle assessments (LCA). Such schemes are being employed in all biofuel production and consumption systems [ 16 ]. As fossil fuels are massive energy sources around the globe, increasing the atmospheric concentration of GHG is still a threat contributed by fossil fuels, which result in global warming and climate change. Many renewable energy policies have been in progress to reduce the carbon-intensive energy carriers. Future low-carbon strategies support different renewable energy resources; many also use industrial waste heat. As many barriers exist for the incorporation of industrial waste heat into district heating systems, Renewable Energy Directive 2018/2001/EU (RED-II) was established to address such issues, which suggest the simplification of market entries and accesses to district heating networks for third parties [ 17 , 18 ]. The advanced exergy analysis-based methods are most promising for the development of sustainable biofuel systems, especially its extensions, such as exergoeconomic and exergoenvironmental approaches, which provide more details about economic, environmental, and technical features of energy conversion systems. On the other hand, the emergy concept quantifies the energy available previously in direct and indirect forms. More details about such analysis are presented in an opinion paper by Tabatabaei et al. [ 16 ]. As a result of the potential that biomass is offering, many technologies are developing toward biomass conversion into biofuels, which have the great advantages of lowering carbon emissions as well as oil dependency due to its production from renewable and organic sources [ 19 ]. As seen from Figure 1 , the USA produces more than 50% of all bioethanol, while Europe’s share represents only 6%; also, each country’s share is less than 5%, while Brazil is the second largest bioethanol-producing country. Bioethanol is the most commonly used biofuel, which is an alternative to fossil fuel and is mainly produced by the hydrolysis of cellulose from lignocellulosic biomass and by the fermentation of sugars of different lignocellulosic sources. The biodegradability and reduced toxicity of bioethanol, for which biomass is used as a primary substrate as well, are its main advantages over fossil fuels [ 11 , 15 ]. Advanced criteria divide the liquid biofuels based on four generation biofuels, depending on the feedstock material being used and utilized for its production ( Figure 2 ). The primary source used for first-generation biofuels is mostly food crops, which were prepared from alcohol. The crop feedstocks used were starchy sourced materials, such as sugar cane and others. However, such feedstock presented some obstacles in the form of high market prices, since they require more chemical fertilizers to increase the biofuel yield. Second-generation biofuels are based on non-edible lignocellulosic biomasses, including whole parts of plants, such as leaves, steam, or bark, but they include also wood chips, different grasses, saw dust, paper pulp, organic wastes, and different forestry and agricultural residues. Polymeric substances, as well as cellulosic substances, are advanced sugar molecules that are found in lots of plants. Grain alcohol is obtained from these substances and is a by-product that can be used as a biofuel. The technology used for the production of second-generation-based biofuels was designed and adjusted to overcome limitations that occurred in first-generation biofuels, since they were also used and utilized as food supplements—meaning, decreasing the production of grain-based alcohol and maximizing the amount of biofuels so they can rival the competitive prices of fossil fuels. In comparison, more gas emissions are saved with lignocellulosic starch alcohol than in first-generation fuels. Considering the existing problem with land biomass feedstocks for the production of biofuels, third-generation biofuel production finds its resources in marine biomass. Such biomass requires much less land area and is good at decreasing the greenhouse emissions into the environment. For such purposes, algal biomass cultivation and farming has increased, since they give additional resources for demanding biofuel production. Improvements in the metabolic production of such biofuels enable the removal of non-fuel components as well as decrease the production costs. However, fourth-generation biofuels are the result of research and development in the fields of biotechnology, biochemistry, plant biology, geosynthesis, and its applications in metabolic and genetic engineering, as they try to cover carbon capture and storage techniques by developing advanced methods for the production of biofuels. Therefore, different bio or genetically engineered biomass feedstocks, such as algae, trees, and plants are developed that are capable of storing and managing carbon release. Thermal energy and power is being utilized by sustainable resources in form of wind, solar, geothermal and hydro energy [ 20 , 21 , 22 ]. Photobiological solar and electrofuels, which are breaking innovative ground with the straightforward conversion of solar energy to biofuels, are also considered as fourth-generation biofuels. Resources for such biofuel production are cheap and available and are a product of the developmental progress of engineered crops through genetic engineering and the emerging field of synthetic biology [ 23 ]. This review paper presents an overview of very recent bioethanol production processes by enzymatic hydrolysis from different lignocellulosic biomass sources, such as marine algae, agricultural residues, and forest feedstocks."
} | 3,431 |
28820125 | null | s2 | 8,258 | {
"abstract": "The kingdom Fungi comprises species that inhabit nearly all ecosystems. Fungi exist as both free-living and symbiotic unicellular and multicellular organisms with diverse morphologies. The genomes of fungi encode genes that enable them to thrive in diverse environments, invade plant and animal cells, and participate in nutrient cycling in terrestrial and aquatic ecosystems. The continuously expanding databases of fungal genome sequences have been generated by individual and large-scale efforts such as Génolevures, Broad Institute's Fungal Genome Initiative, and the 1000 Fungal Genomes Project (http://1000.fungalgenomes.org). These efforts have produced a catalog of fungal genes and genomic organization. The genomic datasets can be utilized to better understand how fungi have adapted to their lifestyles and ecological niches. Large datasets of fungal genomic and transcriptomic data have enabled the use of novel methodologies and improved the study of fungal evolution from a molecular sequence perspective. Combined with microscopes, petri dishes, and woodland forays, genome sequencing supports bioinformatics and comparative genomics approaches as important tools in the study of the biology and evolution of fungi."
} | 307 |
30370317 | null | s2 | 8,259 | {
"abstract": "Bacteria in nature live in complex communities with multiple cell types and spatially-dependent interactions. Studying cells in well-mixed environments such as shaking culture tubes or flasks cannot capture these spatial dynamics, but cells growing in full-fledged biofilms are difficult to observe in real time. We present here a protocol for observing time-resolved, multi-species interactions at single-cell resolution. The protocol involves growing bacterial cells in a near monolayer in a microfluidic device. As a demonstration, we describe in particular observing the dynamic interactions between "
} | 151 |
34165134 | PMC8265773 | pmc | 8,260 | {
"abstract": "Communication assemblies between biomimetic nanocapsules in a 3D closed system with self-regulating and self-organization functionalities were demonstrated for the first time. Two types of biomimetic nanocapsules, TiO 2 /polydopamine capsules and SiO 2 /polyelectrolytes capsules with different stimuli-responsive properties were prepared and leveraged to sense the external stimulus, transmit chemical signaling, and autonomic communication-controlled release of active cargos. The capsules have clear core–shell structures with average diameters of 30 nm and 25 nm, respectively. The nitrogen adsorption–desorption isotherms and thermogravimetric analysis displayed their massive pore structures and encapsulation capacity of 32% of glycine pH buffer and 68% of benzotriazole, respectively. Different from the direct release mode of the single capsule, the communication assemblies show an autonomic three-stage release process with a “jet lag” feature, showing the internal modulation ability of self-controlled release efficiency. The control overweight ratios of capsules influences on communication-release interaction between capsules. The highest communication-release efficiency (89.6% of benzotriazole) was achieved when the weight ratio of TiO 2 /polydopamine/SiO 2 /polyelectrolytes capsules was 5 : 1 or 10 : 1. Communication assemblies containing various types of nanocapsules can autonomically perform complex tasks in a biomimetic fashion, such as cascaded amplification and multidirectional communication platforms in bioreactors.",
"conclusion": "Conclusions In summary, we have successfully demonstrated autonomic communications between different nanocapsules in one system. This provides the possibility of self-regulating and self-organization functionalities and finally shifts control from external stimuli to internal chemical communication between capsules. The colonies of different capsules can exhibit more complicated functions by incorporating communication shuttles and internal modulation ability. We believe such a rational design idea would lead to the development of synthetic systems that can perform autonomic complex tasks in a biomimetic fashion. Also, the combination of uniquely response behaviour and communication opens up exciting opportunities in the design of soft functional materials that are capable of signal transduction. For example, a wide variety of biomimetic communication platforms capable of cascaded amplification and bidirectional communication can be introduced. Or, soft robotic materials have to depend on the ability to exchange signals and react upon them often in complex environments. Furthermore, different inducers and cargos can be encapsulated inside the capsules to realize more complicated autonomic applications, from biological technology to materials science. It can even provide internal physiological cycle changes for the robotic systems.",
"introduction": "Introduction Biological regulatory networks use chemical signaling molecules to drive cell communications and coordinate their actions to obtain quorum-based functionalities. 1–3 Communicating behaviors play a pivotal role in self-organizing processes across all-natural objects, from single cells to individual organisms. For instance, signaling molecules can be recognized by the cellular membrane and transmitted to the genome; 4,5 bacteria release autoinducers communicate with other individuals through processes like quorum sensing. 6,7 The cell-to-cell communication in natural bacteria is the most typical and has been studied extensively. The cell-to-cell interactions involve producing, releasing, detecting, and responding. Recognition of the triggers, chemical signals, receptors, target genes, and mechanisms of signal transduction is leading to a comprehensive understanding of communication between cells. 8 Communicating behaviors between artificial materials and bacterial cells have received considerable attention during the last years. 9–12 Theoretical and computational models have demonstrated communication ability caused by chemical signaling. 13–15 However, according to our best knowledge, there is no wholly artificial biomimetic system that can represent or simulate the process of generating, transforming, and processing chemical signals. Biomimetic nanocapsules have been of great interest in a wide range of scientific areas such as drug delivery, 16,17 catalysis, 18,19 analytical applications, 20 self-healing coatings 21 and multifunctional autonomic materials. 22 Nanocapsules consisting of hollow or porous structures can encapsulate various active cargos ( e.g. , drugs, enzymes, biocides). The nanocapsules have a shell that isolates the encapsulated material from the surrounding environment and has controlled release properties of the cargo. Generally, nanocapsules’ fabrication requires the loading of active cargo and the formation of a stable shell with controlled permeability. 23 Over the last couple of years, various smart materials ( e.g. , nanoparticles, polymers, proteins) have been introduced for the formation of capsule shells. 24–26 Types of stimuli strategies have also been designed in many research works to trigger reversible or irreversible shell transformations/deformation such as pH change, 27 ionic strength, 28 light, 29 and ultrasonication. 30,31 However, all of them focus on individual capsules reacting to external environmental stimuli. There are no examples of two or more different capsules cooperating with each other as well as no mentioning of the capsules’ behavior in communicating assemblies. Artificially designed assemblies of biomimetic nanocapsules will play a significant role in a new generation of smart materials, which enables precise temporal control in a 3D environment and reproducing natural events during material exploitation. Inspired by the biological regulatory networks, we propose a strategy for the rational design of programmable functional assemblies of biomimetic nanocapsules. Compared with a single capsule, these assemblies allow different capsules to work cooperatively achieving complicated and versatile functions. Among these assemblies, the single capsule can sense the external environment changes, and, by releasing initiating cargo, chemically inform surrounding capsules for their further active response. The assemblies of communicating biomimetic nanocapsules can exhibit self-regulating, self-organization functionalities involving internal modulation ability of self-controlled release rate. We built a capsule community containing two types of nanocapsules that exhibit dynamic behavior by chemical information exchange between capsules ( Fig. 1 ). One type is TiO 2 /polydopamine hybrid capsules (TiO 2 –pH@PDA) with encapsulated initiating cargo (pH buffer glycine) as a core which release can be triggered by visible light, and nanostructured hybrid shell containing TiO 2 and polydopamine. The other, receptor-like type, is SiO 2 /polyelectrolytes composite capsules, which are pH-responsive and contain active agents ( e.g. , benzotriazole, BTA) as a core and composite shell formed by SiO 2 and multilayered polyelectrolyte layers (SiO 2 –BTA@PEs). The communicating behavior and chemical mechanisms of the assemblies in a 3D environment have been investigated. The knowledge gained from these artificial capsule networks may lead to the design of synthetic systems that can perform complex tasks in a biomimetic fashion. It will also provide the route to create platforms and devices with self-recognition and self-regulating functionalities without continuous external impact. Fig. 1 Schematic representation of the self-controlled artificial nanocapsule networks. Stable reverse microemulsion templates were first prepared by mixing cyclohexane, deionized water, and non-ionic surfactant polyoxyethylene nonylphenylether (CO-520). The as-formed microemulsion was ultrasonicated for 15 min to obtain uniform and monodisperse droplets, followed by growing inorganic shells (TiO 2 /SiO 2 ) on the templates. Subsequently, the formation of biomimetic nanocapsules was present by cargo loading and exterior shell building through dopamine polymerisation and polyelectrolyte layer-by-layer (LBL) assembly, respectively. Then, the mixed solution containing TiO 2 –pH@PDA and SiO 2 –BTA@PEs was illuminated by visible light for 400 min using a 100 W Xe lamp under continuous stirring. The pH change of the solution was tested at the given interval. The fluorescence intensity at the emission maximum of BTA was plotted as a function of time to obtain a final communication-controlled release profile.",
"discussion": "Results and discussion Two types of inorganic nanospheres (TiO 2 and SiO 2 ) were synthesized using reverse microemulsions as templates to obtain nanocapsules with large interior space and developed pore structures. Also, the robust inorganic nanospheres allow the high loading of cargo inside the capsule. DLS analysis indicates that the as-synthesized TiO 2 and SiO 2 nanospheres are monodisperse with average hydrodynamic diameters of 20 ± 5 nm and 18 ± 5 nm, respectively. After exterior shell formation, the final capsules with cargo have average hydrodynamic diameters of 30 ± 5 nm and 25 ± 5 nm, respectively (Fig. S1 † ). With polydopamine decoration, TiO 2 nanospheres have a broader size distribution, caused by polydopamine strong adhesion to TiO 2 . However, the SiO 2 nanospheres with multilayered polyelectrolytes have a narrow size distribution, owing to the electrostatic repulsion between poly (sodium 4-styrenesulfonate) (PSS) used as final polyelectrolyte layer. Transmission electron microscopy (TEM) analysis was performed to observe inorganic nanospheres’ morphology and structure before and after encapsulation ( Fig. 2 ). From the images, it can be noted that both TiO 2 –pH@PDA and SiO 2 –BTA@PEs capsules have clear core–shell structure after encapsulation. Fig. 2 TEM image of a typical (a) TiO 2 ; (b) TiO 2 –pH@PDA capsules; (c) N 2 sorption isotherms of TiO 2 and SiO 2 (77.3 K): (i) filled circles, adsorption experiments; (ii) unfilled circle, desorption experiments. (d) TEM images of SiO 2 ; and (e) SiO 2 –BTA@PEs capsules; (f) the enlarged view of a single SiO 2 –BTA@PEs capsule. The nitrogen adsorption–desorption isotherms were measured to characterize the pore structures of as-synthesized TiO 2 and SiO 2 nanospheres ( Fig. 2c ). According to the IUPAC classification, 32 both isotherms are classic type IV showing a hysteresis loop characteristic to mesoporous materials. For the TiO 2 nanospheres, the curve exhibits a hysteresis type 2 loop at the relative pressures between 0.6 and 0.8. It is well known that various size of the cavities causes this type of hysteresis loop. This result indicated that TiO 2 nanospheres have massive disordered mesopores structure. The massive mesopore structure has lots of ink-bottle shapes with narrow necks and broader bodies providing huge cavities for cargo encapsulation. The SiO 2 nanospheres have a narrower hysteresis loop and almost parallel hysteresis branches. It is confirmed that the SiO 2 nanospheres have a highly homogeneous interconnected 3D mesopore structure with large, well-ordered mesopores. The proper interior structure of inorganic nanospheres benefits the maximum loading of different cargos. Thermogravimetric analysis (TGA) demonstrated the maximum encapsulation capacity for both nanospheres (Fig. S2 † ). The final TiO 2 –pH@PDA and SiO 2 –BTA@PEs capsules possess the maximum loading of cargos for 32% and 68%, respectively. The detailed structural information of both capsules is illustrated in Table 1 for comparison. Structures and properties of TiO 2 –pH@PDA and SiO 2 –BTA@PEs capsules Sample Nanosphere size (nm) Capsule size (nm) Surface area (cm 2 g −1 ) Pore volume (cm 3 g −1 ) Encapsulation capacity (%) TiO 2 –pH@PDA capsules 20 ± 5 30 ± 5 97.07 0.15 32 SiO 2 –BTA@PEs capsules 18 ± 5 25 ± 5 161.47 0.42 68 To further confirm the formation of hybrid nanocapsules, the chemical composition of the initial inorganic nanospheres and final encapsulated nanocapsules was characterized by attenuated total reflection-Fourier transform infrared (ATR-FTIR) spectroscopy ( Fig. 3 ). A series of weak absorption peaks at about 3500 cm −1 can be attributed to the stretching vibration of the –OH bond. The peak at 1630 cm −1 is assigned to the bonding modes of Ti–OH. The typical absorption peaks at 630 and 1380 cm −1 are corresponding to the stretching vibration of Ti–O. Compared with the TiO 2 , the TiO 2 –pH@PDA capsules show extra absorption peaks at 1060, 1250, 1502, and 1640 cm −1 which can be assigned to the bending δ (C–H), the indole ring CNC stretching, ν ring (C \n \n\n<svg xmlns=\"http://www.w3.org/2000/svg\" version=\"1.0\" width=\"13.200000pt\" height=\"16.000000pt\" viewBox=\"0 0 13.200000 16.000000\" preserveAspectRatio=\"xMidYMid meet\"><metadata>\nCreated by potrace 1.16, written by Peter Selinger 2001-2019\n</metadata><g transform=\"translate(1.000000,15.000000) scale(0.017500,-0.017500)\" fill=\"currentColor\" stroke=\"none\"><path d=\"M0 440 l0 -40 320 0 320 0 0 40 0 40 -320 0 -320 0 0 -40z M0 280 l0 -40 320 0 320 0 0 40 0 40 -320 0 -320 0 0 -40z\"/></g></svg>\n\n N) stretching, and ν ring (C C) stretching modes ( Fig. 3a ). The spectra profile of SiO 2 –BTA@PEs capsules also provides series of characteristic peaks. The absorption peaks at 475, 820, and 1098 cm −1 can be attributed to Si–O–Si bending vibration, symmetric stretching of Si–O–Si, and asymmetric vibration of Si–O. After encapsulation, a small peak at 750 cm −1 is found, which corresponds to in-plane bending vibrations of C–H in the BTA benzene ring. Also, the peaks at 1470 and 1590 cm −1 can be assigned to the symmetric distortion and asymmetric stretching vibrations of –NH 3+ . The peaks of hydrogen bonding caused by C–H vibration can be found at about 2840–2990 cm −1 . Also, two weak peaks at 3150 and 3400 cm −1 correspond to the –N 2 group ( Fig. 3b ). The ATR-FTIR analysis indicated that the polydopamine and multilayered PEs successfully decorated the surface of TiO 2 and SiO 2 , respectively. Fig. 3 ATR-FTIR spectra of (a) TiO 2 (black), TiO 2 –pH@PDA capsules (red); (b) SiO 2 (black), SiO 2 –BTA@PEs capsules (purple). In order to study the release kinetics of a single capsule, we carried out photodegradation measurement and BTA release curve test on individual TiO 2 –pH@PDA capsules and SiO 2 –BTA@PEs capsules first. The detailed experimental procedures and methods can be found in ESI. † For the light-responsive TiO 2 –pH@PDA capsules, the stimuli-release performance was tested by pH change of the solution and photocatalytic activity through degrading of Rhodamine B (RhB) under visible light irradiation. Typically, pristine TiO 2 can respond to UV irradiation but shows no photoactivity under visible light irradiation. Hence, we employed PDA as a surface modifier to obtain TiO 2 –pH@PDA hybrid shell sensitivity to visible light. The improved photoactivity of TiO 2 –pH@PDA capsules was confirmed by the complete degradation of RhB (wavelength = 560 nm) after 180 min of visible light irradiation (Fig. S3 † ). The release of pH buffer from TiO 2 –pH@PDA was tested under visible light irradiation at continuous stirring (Fig. S4 † ). The pH value increased gradually and reached equilibrium after 220 min of irradiation. However, the suspension pH was kept almost unchanged without visible light irradiation. Therefore, PDA modification can effectively suppress spontaneous leakage from TiO 2 mesopores and provide effective light-response in the visible light range. The layer-by-layer (LBL) technique is a well-studied method to fabricate microcapsules. The release profile of BTA from SiO 2 –BTA@PEs capsules was measured at different pH (Fig. S5 † ). The variation of BTA adsorption peaks (274 nm) versus time is shown in Fig. S6. † The multi-layered polyelectrolyte shells composed of weak polyelectrolytes (PAH and PSS) are responsive to the pH of the environment. The leakage of BTA was restrained by multilayered polyelectrolytes at pH = 7. When pH increases to 10, the multi-layered polyelectrolytes will swell to increase the permeability, BTA molecules demonstrate a gradual release process. About 35% of BTA was released in the first 20 min. After revealing the stimuli-responsive behaviour of single capsules alone, we focused on the internal communication between different capsules in a solution. The nanosized capsules containing different chemical signalling and cargos were mixed in an aqueous solution under stirring. The pH buffer (pH = 10) encapsulated inside the TiO 2 –pH@PDA capsules was introduced as a chemical signal to build the communication bridge between two different types of capsules. After the TiO 2 –pH@PDA was initiated by light exposure, the pH buffer was released as exchanging substances to the solution leading to the pH change. The SiO 2 –BTA@PEs subsequently response to the pH change and finish the communication-controlled BTA release. Fig. 4a shows the typical spectrum of communication-controlled BTA release (TiO 2 –pH@PDA/SiO 2 –BTA@PEs = 5 : 1 mixing ratio). The adsorption peak of BTA has a noticeable increase with the time increased. This indicates that the release of BTA, which is not controlled by light in SiO 2 –BTA@PEs capsules, is controlled by light now in the presence of TiO 2 –pH@PDA capsules. This is owed to TiO 2 –pH@PDA, which can regulate the pH of the closed system acting as a signal bridge. Compared with the direct release mode of BTA (solution pH = 10) from SiO 2 –BTA@PEs capsules, the communication-controlled release has an entirely different release mode ( Fig. 4b ). Fig. 4 Dynamic release behaviour controlled by chemical information exchange between different capsules. (a) UV-vis spectra of BTA released from SiO 2 –BTA@Pes capsules (TiO 2 –pH@PDA/SiO 2 –BTA@PEs = 5 : 1); (b) the release profile of BTA controlled by single SiO 2 –BTA@PEs capsules (red dotted line) and communication-controlled by mixed capsules (red line), and pH change (blue line); (c) the release profile of BTA under different ratio of TiO 2 –pH@PDA to SiO 2 –BTA@Pes capsules; (d) the cumulative release efficiency of BTA under different ratio of TiO 2 –pH@PDA to SiO 2 –BTA@Pes capsules. We revealed three periods of the whole autonomic communication-controlled release process. At the first stage (0–180 min), a slow-release profile is caused by the slight pH change and a low release of BTA was observed. Subsequently, more pH buffer was released from the TiO 2 –pH@PDA under continuous visible light irradiation at 180–280 min period leading to the faster release of BTA at the second stage. The maximum release efficiency (89.6%) was achieved at the end of release process (280–400 min). It is worth noting that the communication system shows a “jet lag” between pH change and BTA release. In principle, the multilayered PSS/PAH shell is open at pH = 10. However, the BTA release rate started to increase at about pH = 9.5. Typically, BTA is an amphoteric compound and benzotriazole species can be transferred through protonic equilibria depending on the solution pH. Under a basic environment (pH = 10), the neutral benzotriazole becomes deprotonated with local OH– consumption leading to the initial delay phenomenon by decreasing free OH– concentration. Before the pH change to 10 (first 200 min), the release efficiency of BTA is about 18%. This evidences that when SiO 2 –BTA@PEs capsules received chemical signals sent by TiO 2 –pH@PDA, they can either release cargos or not respond depending on OH– concentration. The communication behavior can also be affected by the weight ratio of TiO 2 –pH@PDA/SiO 2 –BTA@PEs mixture ( Fig. 4c and d ). The study of communication release behaviour under different ratios was illustrated in Fig. S7. † When the ratio is 1 : 1, there is almost no communication observed between capsules. The quantity of pH buffer released from TiO 2 –pH@PDA capsules is not enough to change the solution pH to 10 because all OH– is consumed by the deprotonated benzotriazole. When the ratio increases to 3 : 1, more pH buffer diffuses into the external solution. The shell of SiO 2 –BTA@PEs capsules was opened after 220 min of irradiation, the release efficiency is gradually increased for about 60 min and then reached a final equilibrium of 50%. Although SiO 2 –BTA@PEs capsules were opened, there is not enough encapsulated pH buffer to reach protonic equilibrium. The existed OH– is continuously consumed by BTA, leading to the pH decrease. So, the multi-layered polyelectrolytes shells are closing again after 50% BTA release. The neutral benzotriazole concentration in the solution also reaches saturation level suppressing the continuous release of BTA from SiO 2 –BTA@PEs capsules and creating stopping feedback response. The communication behavior between capsules stops due to the lack of chemical signals, and the final incomplete release of BTA is achieved. The maximum release efficiency is obtained at the 5 : 1 or 10 : 1 TiO 2 –pH@PDA/SiO 2 –BTA@PEs weight ratio. Sufficient amount of pH buffer triggers intensive release of BTA with positive feedback. 33 ( Fig. 5 ). The abundant pH buffer supply keeps the pH stable during BTA protonation process. The polyelectrolytes shell is open and BTA is continuously released and deprotonated, leading the completed BTA release. By this way, we achieved biomimetic reproduction of the effective chemical signal cooperations between biological cells. Fig. 5 Schematic representation of the communication mechanism of self-controlled artificial nanocapsule networks. The different weight ratios of TiO 2 –pH@PDA/SiO 2 –BTA@PEs capsules create different self-communication strategy. The lower ratio leads to incomplete release of BTA with a negative feedback loop. The higher ratios trigger complete release of BTA with positive feedback loop. The different nanocapsules containing pH buffer and BTA were mixed in an aqueous solution to make complete autonomic signalling system. Hence, considering the dynamic stimuli-response-communication-decide-response behavior, the assemblies of different biomimetic nanocapsules can exhibit self-regulating, self-organization functionalities involving internal modulation of self-controlled release efficiency."
} | 5,600 |
26941941 | PMC4761787 | pmc | 8,261 | {
"abstract": "Abstract The foundational concepts behind the persistence of ecological communities have been based on two ecological properties: dynamical stability and feasibility. The former is typically regarded as the capacity of a community to return to an original equilibrium state after a perturbation in species abundances and is usually linked to the strength of interspecific interactions. The latter is the capacity to sustain positive abundances on all its constituent species and is linked to both interspecific interactions and species demographic characteristics. Over the last 40 years, theoretical research in ecology has emphasized the search for conditions leading to the dynamical stability of ecological communities, while the conditions leading to feasibility have been overlooked. However, thus far, we have no evidence of whether species interactions are more conditioned by the community's need to be stable or feasible. Here, we introduce novel quantitative methods and use empirical data to investigate the consequences of species interactions on the dynamical stability and feasibility of mutualistic communities. First, we demonstrate that the more nested the species interactions in a community are, the lower the mutualistic strength that the community can tolerate without losing dynamical stability. Second, we show that high feasibility in a community can be reached either with high mutualistic strength or with highly nested species interactions. Third, we find that during the assembly process of a seasonal pollinator community located at The Zackenberg Research Station (northeastern Greenland), a high feasibility is reached through the nested species interactions established between newcomer and resident species. Our findings imply that nested mutualistic communities promote feasibility over stability, which may suggest that the former can be key for community persistence.",
"introduction": "Introduction How can ecological communities sustain a large number of species? This is a major question that has greatly intrigued ecologists since the 1920s (Elton 1927 , 1958 ; Odum 1953 ; MacArthur 1955 ; Margalef 1968 ). Two ecological properties have been considered the foundational concepts behind the persistence of ecological communities: dynamical stability and feasibility (MacArthur 1955 ; Gardner and Ashby 1970 ; Vandermeer 1970 , 1975 ; May 1972 ; Roberts 1974 ; De Angelis 1975 ; Goh 1979 ; Yodzis 1980 ; Svirezhev and Logofet 1983 ; Logofet 1993 ). Dynamical stability (hereafter stability) asks whether a community will return to an assumed equilibrium state after a perturbation in species abundances, and it is linked to the strength of interspecific interactions (Svirezhev and Logofet 1983 ; Logofet 1993 ). Feasibility corresponds to the range of tolerated combinations of species demographic characteristics (intrinsic growth rates or carrying capacities) under which all species can have positive abundances (Vandermeer 1970 , 1975 ; Svirezhev and Logofet 1983 ; Logofet 1993 ; Bastolla et al. 2009 ; Nattrass et al. 2012 ; Rohr et al. 2014 ; Saavedra et al. 2014 ). Importantly, the conditions leading to the stability of a community do not automatically imply its feasibility and vice versa (Vandermeer 1970 , 1975 ; Roberts 1974 ; Svirezhev and Logofet 1983 ; Stone 1988 ; Logofet 1993 ; Rohr et al. 2014 ). Over the last 40 years, theoretical research in ecology has emphasized the search for conditions leading to the stability of ecological communities (May 1972 ; De Angelis 1975 ; Goh 1979 ; Yodzis 1980 ; Svirezhev and Logofet 1983 ; Logofet 1993 ; Staniczenko et al. 2013 ), while the conditions leading to feasibility have received considerably less attention (Vandermeer 1970 , 1975 ; Roberts 1974 ; Svirezhev and Logofet 1983 ; Logofet 1993 ; Hofbauer and Sigmund 1998 ; Nattrass et al. 2012 ; Rohr et al. 2014 ). However, theoretical and empirical studies have shown that the sequence of community assembly cannot be understood without feasibility conditions (Drake 1991 ; Law and Morton 1996 ; Weatherby et al. 2002 ; Saavedra et al. 2009 ). Yet, the extent to which species interactions are more conditioned by the community's need to be stable or feasible is still unclear. This is important in order to better understand the link between community structure and dynamics, especially as global environmental change is accelerating the rate at which species are removed and introduced into new habitats (Sala et al. 2000 ; Tylianakis et al. 2008 ). To answer the above question, we introduce general quantitative methods to investigate the role of stability and feasibility in shaping mutualistic communities. Because stability and feasibility are linked and conditioned by species interactions, we study the general association between stability and feasibility in mutualistic communities and how this association is modulated by species interaction networks. We then move to an empirical case by studying how the association between stability and feasibility acts on the assembly of a seasonal Arctic pollinator community located at The Zackenberg Research Station, northeastern Greenland (748300N, 218000W). Finally, we discuss the implications of our findings.",
"discussion": "Discussion The above findings have a series of interesting implications. First, the fact that nestedness tunes a trade‐off between feasibility and stability may imply that different ecosystem services in mutualistic systems are not in the same direction (Loreau 2010 ; Turnbull et al. 2013 ). This means that it is not guaranteed that one component of community dynamics could always be used as a proxy for another component. While previous studies have emphasized the high level of nestedness in mutualistic communities, less attention has been given to why observed nestedness is not even higher. Our results on the trade‐off between stability and feasibility may explain why there might be a limit to nestedness: A further increase of an already high feasibility can be counterbalanced by a strong decrease in stability. Second, the finding that feasibility is increased via nested—as opposed as through an increase in mutualistic strength—in the empirical community may be explained by dynamical and biologic constraints. The dynamical constraints may be imposed by the theoretical observation that high mutualistic strengths can push the community to shift from a weak to a strong mutualistic regime, which can easily take the community to rather unpredictable dynamics (Bastolla et al. 2009 ; Saavedra et al. 2013 ; Rohr et al. 2014 ). The biologic constraints may originate from the empirical observation that mutualisms among free living species are of low specificity, which is compatible with the combination of coevolutionary convergence and complementarity (Thompson 2005 ). In both cases, communities, especially under short‐term dynamics, may have a higher opportunity to increase feasibility by changing the organization of their interactions rather than by increasing the overall mutualistic strength. Third, the finding that feasibility is being promoted over stability may confirm that under short‐term dynamics, the community may not need to be highly dynamically stable in order for species to coexist. For instance, other studies have suggested that asynchronous dynamics, reducing the amplification of perturbations, or reducing the variability of the total abundance may have more biologic relevance for the community than the capacity to return to an equilibrium point (Loreau 2010 ). Importantly, these findings reveal that feasibility is an important condition for species coexistence even under short‐term dynamics and requires further exploration. Finally, it is noteworthy that over more than 40 years, many studies in theoretical ecology have been focused on the dynamical stability of ecological communities, in particular on local asymptotic stability. Indeed, one of the long‐standing questions in ecology has been whether large ecological communities will be more locally stable (May 1972 ). However, empirically and theoretically, there has been no evidence demonstrating that dynamical stability should be the most important ecological property leading to community persistence. In fact, our results show that dynamical stability might not be as relevant as feasibility for species coexistence in seasonal communities. This calls for a stronger research program on the factors modulating feasibility and alternative stability conditions in species interaction networks, as they can hold the key for a general theory of community persistence."
} | 2,161 |
37076886 | PMC10114483 | pmc | 8,262 | {
"abstract": "Background To realize the full potential of softwood-based forest biorefineries, the bottlenecks of enzymatic saccharification of softwood need to be better understood. Here, we investigated the potential of lytic polysaccharide monooxygenases (LPMO9s) in softwood saccharification. Norway spruce was steam-pretreated at three different severities, leading to varying hemicellulose retention, lignin condensation, and cellulose ultrastructure. Hydrolyzability of the three substrates was assessed after pretreatment and after an additional knife-milling step, comparing the efficiency of cellulolytic Celluclast + Novozym 188 and LPMO-containing Cellic CTec2 cocktails. The role of Thermoascus aurantiacus Ta LPMO9 in saccharification was assessed through time-course analysis of sugar release and accumulation of oxidized sugars, as well as wide-angle X-ray scattering analysis of cellulose ultrastructural changes. Results Glucose yield was 6% ( w / w ) with the mildest pretreatment (steam pretreatment at 210 °C without catalyst) and 66% ( w / w ) with the harshest (steam pretreatment at 210 °C with 3%( w / w ) SO 2 ) when using Celluclast + Novozym 188. Surprisingly, the yield was lower with all substrates when Cellic CTec2 was used. Therefore, the conditions for optimal LPMO activity were tested and it was found that enough O 2 was present over the headspace and that the reducing power of the lignin of all three substrates was sufficient for the LPMOs in Cellic CTec2 to be active. Supplementation of Celluclast + Novozym 188 with Ta LPMO9 increased the conversion of glucan by 1.6-fold and xylan by 1.5-fold, which was evident primarily in the later stages of saccharification (24–72 h). Improved glucan conversion could be explained by drastically reduced cellulose crystallinity of spruce substrates upon Ta LPMO9 supplementation. Conclusion Our study demonstrated that LPMO addition to hydrolytic enzymes improves the release of glucose and xylose from steam-pretreated softwood substrates. Furthermore, softwood lignin provides enough reducing power for LPMOs, irrespective of pretreatment severity. These results provided new insights into the potential role of LPMOs in saccharification of industrially relevant softwood substrates. Supplementary Information The online version contains supplementary material available at 10.1186/s13068-023-02316-0.",
"conclusion": "Conclusions In the present study, we investigated the role of LPMO9 in the saccharification of differentially pretreated softwood substrates. Results suggest that the requirements of LPMOs for O 2 as co-substrate and reductant are met, respectively, by aeration over a relatively small headspace and the presence of lignin in the softwood substrates. Although Ta LPMO9 was active on all three softwood substrates, the highest Glc4gemGlc release was found in the material subjected to the harshest pretreatment (STEX 210°C/SO2 ), along with a more pronounced drop in cellulose crystallinity. Taken together, these results indicate that the enzyme benefitted from easier accessibility to cellulose created by the harsher pretreatment. Time-course analysis of Ta LPMO9-supplemented reactions showed that the time frame of Glc4gemGlc accumulation was substrate-dependent. The beneficial impact of LPMOs manifests itself only in the later stages of hydrolysis, likely because processive cellulose hydrolases need some time to exploit the newly created access points. The findings in this study provide new insights on the role and practical applicability of LPMOs in the saccharification of softwood substrates. Accordingly, the use of LPMOs and steam pretreatment speed up the fractionation of spruce and improve monosaccharide yields.",
"discussion": "Results and discussion The overall aim of this study was to investigate the role of LPMO9 in the saccharification of softwood substrates. Spruce material was first steam-pretreated under three various conditions to yield solids with or without hemicellulose retention. The three STEX-pretreated spruce materials were compositionally analyzed, and the hydrolyzability evaluated using the cellulolytic enzyme cocktail Celluclast + Novozym188. The materials were additionally milled to potentially increase saccharification, and then hydrolyzed with Celluclast + Novozym188 or the LPMO-containing Cellic CTec2. After verifying that the chosen reaction conditions support LPMO9 activity, Celluclast + Novozym188 was supplemented with Ta LPMO9 and the sugar release, accumulation of oxidized sugars, and cellulose ultrastructure were investigated for all three softwood substrates. Composition and hydrolyzability of differentially steam-pretreated spruce Steam pretreatment was previously performed under three different conditions [ 13 ], resulting in two materials retaining hemicellulose (STEX 210°C/auto and STEX 210°C/HAc ) and one devoid of hemicellulose (STEX 210°C/SO2 ). Loss of arabinose from arabinoglucuronoxylan and galactose from galactoglucomannan was observed in all three materials after pretreatment. The overall hemicellulose content for STEX 210°C/auto , STEX 210°C/HAc , and STEX 210°C/SO2 was 6%, 3%, and 0% ( w / w ) dry matter, respectively (Fig. 1 A). Fig. 1 Materials composition and comparison of sugar release from steam-pretreated spruce, with or without an additional knife-milling step. A Materials composition expressed as percentage dry mass ( w / w ). B – D Release of B glucose, C xylose, and D mannose was measured after 48 h of hydrolysis. Please note that glucose can be released from both cellulose and galactoglucomannan. Data represent mean values ± standard deviation of triplicate experiments. Where error bars are not represented, they are below 0.5% The hydrolyzability of steam-pretreated spruce was assessed after incubation with Celluclast + Novozym 188 for 48 h (Fig. 1 B–D). The hydrolysis efficiency of cellulose (measured as glucose) release from STEX 210°C/auto , STEX 210°C/HAc , and STEX 210°C/SO2 resulted in 6%, 11%, and 66% ( w / w ) glucan, respectively (Fig. 1 B). To potentially increase enzyme accessibility, the materials were additionally milled after STEX and prior to enzymatic hydrolysis. As compared to the base case, where the materials were not milled, these yields represent increases in glucose release of 1.8-, 1.8-, and 1.0-fold for STEX 210°C/auto , STEX 210°C/HAc , and STEX 210°C/SO2 , respectively (Fig. 1 B). A similar trend was observed for xylose and mannose, whose release improved by 1.6- and 1.2-fold with STEX 210°C/auto and by 2.0- and 1.4-fold with STEX 210°C/HAc following milling (Fig. 1 C, D). Increased hydrolysis efficiency after milling [ 39 ] has been linked to a reduction in particle size and, consequently, larger surface area [ 17 ]. However, due to the high energy demand, milling has an adverse impact on the process economy. Further, as most of the energy required dissipates into heat [ 16 , 17 ], moisture content in the material can decrease. Hornification reactions may follow from the drying that to a varying degree follow milling. Hornification in turn may induce the cellulose ultrastructure to collapse [ 7 , 40 ], further reducing cellulose accessibility to enzymes, as well as decreasing hydrolysis yields and rates [ 7 ]. In the present study, the impact of milling on hydrolysis yields was assessed on STEX 210°C/auto , STEX 210°C/HAc , and STEX 210°C/SO2 . Surprisingly, hydrolysis yields were lower with LPMO-containing Cellic CTec2 than Celluclast + Novozym 188 (Fig. 1 ). This was especially surprising as the LPMOs in Cellic CTec2 were expected to result in increased cellulose conversion yields. Such result could be explained by hydrolysis conditions being unsuitable for LPMOs. To ensure that LPMO can be active under the presented reaction conditions, various reductant and aeration conditions were tested in the next step. Because milling negatively affects the process economy, the following experiments were conducted using non-milled steam-pretreated material. Effect of reductant addition and oxygen availability on saccharification of steam-pretreated spruce by LPMO The lower saccharification efficiency of Cellic CTec2 compared to Celluclast + Novozym 188, prompted the optimization of reaction conditions to ensure LPMO activity. To assess LPMO activity, we measured the release of C4-oxidized cellobiose, known as Glc4gemGlc [ 41 , 42 ]. Glc4gemGlc concentration was significantly higher (approximately 87%) in reactions run with Cellic CTec2 than Celluclast + Novozym 188 (Additional file 1 : Fig. S1). This result confirmed LPMO activity in Cellic CTec2, but its near absent from Celluclast + Novozym 188. To check if LPMO activity could be improved, reactions were supplemented with ascorbic acid as a reducing agent and aeration was increased by ensuring a larger headspace volume. Lignin has been shown to be a natural LPMO reductant, but the type and chemical state of lignin likely impacts its reducing power [ 18 , 41 , 43 ]. Different headspace volumes (10%, 60%, and 80%) were tested to augment O 2 transfer and, hence, the supply of O 2 as co-substrate for LPMO. In the presence of reductants (and transition metals), improved oxygenation can supply the enzyme with H 2 O 2 , another potential co-substrate [ 29 ]. The impact of reductant addition and increased aeration was assessed based on glucose, xylose, mannose, and Glc4gemGlc release (Fig. 2 and Additional file 1 : Fig. S2). The results have been statistically analyzed using the Wilcoxon rank-sum test. Anaerobic conditions were not included into the study due to the need of H 2 O 2 supplementation while we wanted to perform the enzymatic hydrolysis with minimal chemical additions. Fig. 2 Influence of aeration and reductant supplementation on enzymatic hydrolysis yields using Cellic CTec2. Enzymatic hydrolysis of A – C STEX 210°C/auto , D – F STEX 210°C/HAc , and G STEX 210°C/SO2 with or without 10 mM ascorbic acid. Different final reaction weights (1.8, 10, and 20 g) corresponding to different headspace volumes (10%, 80%, and 60%, respectively) of the total reaction volume were assessed. Release of A , D , G glucose; B , E xylose; and C , F mannose after 48 h of hydrolysis. Data represent mean values ± standard deviation of triplicate experiments. Please note that the scales have been adjusted for clarity. Where error bars are not represented, they are below 0.5% When testing different headspace volumes, the oxidized glucose dimer Glc4gemGlc could be detected in almost all reactions (Additional file 1 : Fig. S2), but without any improvement in sugar yields (Fig. 2 ). This finding indicated that the O 2 available in the headspace of the smaller reaction tested (1.8 g and 10% headspace) was sufficient for LPMOs to catalyze their reaction. Statistical analysis did not show any significant difference when the different reactions weights were compared between themselves. Interestingly, supplementation with ascorbic acid caused generally lower saccharification yields compared to lignin as sole reductant (Fig. 2 ). These results suggested that the lignin present in all three materials provided sufficient reducing power for LPMOs. In fact, ascorbic acid might have enhanced production of H 2 O 2 from O 2 inside redox reactions [ 27 ], which can cause auto-inactivation of LPMOs and also harm other enzymes in the cocktail [ 29 , 38 , 41 ]. It was not possible to use the Glc4gemGlc assay to analyze LPMO activity following ascorbic acid addition, likely due to interference from unknown compounds formed in the reaction (data not shown). Even though it was not possible to detect the Glc4gemGlc, we believe that the LPMOs have been activated during the enzymatic hydrolysis due to the use of similar reaction conditions employed in the LPMOs activity test (Additional file 1 : Fig. S1). Similar conclusions were reached by Hansen et al. [ 45 ] in a study on Norway spruce previously steam-pretreated using 2-naphthol as a carbocation scavenger. This and the current work, indicate that lignin has sufficient reducing power to activate LPMO even in materials pretreated under the harshest conditions (e.g., STEX 210°C/SO2 ). Such finding is surprising as the changes induced to the lignin during pretreatment, especially the loss of reactive bonds (e.g., beta-O-4) by condensation reactions, can be profound, as discussed in our previous study [ 13 ]. Role of Ta LPMO9A in the hydrolysis of mildly pretreated spruce Next, the effect of Ta LPMO9A on the saccharification of steam-pretreated spruce (STEX 210°C/auto , STEX 210°C/HAc , and STEX 210°C/SO2 ) by Celluclast + Novozym 188 was investigated. As in the previous experiment, the release of glucose, xylose and mannose and Glc4gemGlc analysis (Fig. 3 and Fig. 4 ) were used to evaluate saccharification efficiencies and Ta LPMO9A activity, respectively. In addition, a time course of the release was plotted to gain insight on the role of LPMO during different phases of enzymatic hydrolysis. Fig. 3 Time-course analysis of sugars released during enzymatic hydrolysis of steam-pretreated spruce with or without Ta LPMO9A supplementation. Enzymatic hydrolysis of A – C STEX 210°C/auto , D – F STEX 210°C/HAc , and G STEX 210°C/SO2 with or without Ta LPMO9A. Release of A , D , G glucose; B , E xylose; and C , F mannose during 72 h of hydrolysis. Data represent mean values ± standard deviation of triplicate measurements. Please note that scales were adjusted for clarity. Where error bars are not represented, they are below 0.5% Fig. 4 HPAEC-PAD chromatograms of the C4-oxidized glucose dimer Glc4gemGlc in time-course reactions with Celluclast + Novozym 188 supplemented with Ta LPMO9A. Chromatograms for A STEX 210°C/auto , B STEX 210°C/HAc , and C STEX 210°C/SO2 ( C ). The reactions were run in triplicates. Given the elevated similarity between all triplicate chromatograms, only one of them is presented in the figure Ta LPMO9A supplementation achieved consistently higher glucose release compared to Celluclast + Novozym 188 alone, with 1.6-, 1.4-, and 1.1-fold greater values for STEX 210°C/auto , STEX 210°C/HAc , and STEX 210°C/SO2 , respectively (Fig. 3 ). Using STEX 210°C/auto as substrate, Glc4gemGlc started to accumulate within 4–8 h, rose until 24 h (Fig. 4 A), and decreased thereafter. A similar effect was observed previously [ 41 ], and was ascribed to abiotic (on-column) degradation of Glc4gemGlc at the high pH encountered during analysis. When STEX 210°C/HAc was the substrate, Glc4gemGlc accumulation started again within 4–8 h, but then increased until 48 h (Fig. 4 B). The delayed onset of Glc4gemGlc detection does not necessarily indicate inactive LPMOs. The enzyme likely started working on the cellulose, creating new oxidized ends in the glucan chain, but Glc4gemGlc could not yet be revealed by high-performance anion-exchange chromatography with pulsed amperometric detection (HPAEC-PAD). Finally, using STEX 210°C/SO2 as substrate (Fig. 4 C), Glc4gemGlc started to accumulate already within 0–2 h and increased until 24 h. A comparison of the sugar release time course (Fig. 3 ) with Glc4gemGlc development (Fig. 4 ), gives additional insights into the role of Ta LPMO9A in softwood saccharification. It seems that the beneficial impact of Ta LPMO9A supplementation on glucose release only became apparent in the later part of hydrolysis. In fact, the initial rates of hydrolysis were similar for non-supplemented and Ta LPMO9A-supplemented reactions (Fig. 4 ). LPMOs are believed to aid cellulose saccharification by creating nicks in the glucan chain on the more ordered (crystalline) parts of the cellulose microfibril. These nicks provide new binding sites for processive hydrolytic cellulases, which can then start degrading the cellulose by surface erosion . Due to the greater energy required to hydrolyze crystalline cellulose [ 46 ], the rate is significantly lower compared to that of less ordered or para-crystalline cellulose [ 7 ]. This might explain why the positive effect of Ta LPMO9A becomes apparent only in the later stages of hydrolysis, as the slower processive enzymes make use of the nicks created at the beginning of the reaction. As reported in ‘‘ Ultrastructural characterization of cellulose before and after enzymatic hydrolysis ’’ Section, crystallinity (i.e., ratio of highly ordered to less ordered cellulose) increased with increasing pretreatment severity (from STEX 210°C/auto to STEX 210°C/SO2 ). The higher crystallinity of STEX 210°C/SO2 might have resulted in a slower hydrolysis rate, possibly explaining the lesser effect of Ta LPMO9A compared to mildly pretreated STEX 210°C/auto and STEX 210°C/HAc . Additionally, Ta LPMO9A supplementation led to a clear improvement in xylose release (Fig. 3 B–E) in mildly pretreated STEX 210°C/auto and STEX 210°C/HAc (1.5- and 1.3-fold, respectively). In a previous study, we showed that the residual hemicellulose is highly recalcitrant towards hydrolysis due to its low amounts and the presence of lignin–carbohydrate complexes [ 13 ]. Separately, we reported that Thermothielavioides terrestris LPMO9 could be active on arabinoglucuronoxylan from spruce [ 32 ]. Collectively, these findings suggest that Ta LPMO9A could promote the hydrolysis of the xylan backbone. Specifically, Ta LPMO9A might act in concert with enzyme cocktails, increasing xylose release and facilitating cellulose hydrolysis by removing hemicellulose. This, in turn, increases enzyme accessibility. Ultrastructural characterization of cellulose before and after enzymatic hydrolysis To further elucidate the role of Ta LPMO9A in saccharification of STEX 210°C/auto, STEX 210°C/Hac , and STEX 210°C/SO2 , changes to cellulose ultrastructure were analyzed by WAXS before and after enzymatic hydrolysis using Celluclast + Novozym 188, with or without Ta LPMO9A addition. The presence of crystalline and para-crystalline cellulose domains led to a distinct scattering pattern (Additional file 1 : Figs. S3, S4), which was used to determine relative crystallinity as well as the size of cellulose crystallites. The width of the crystallite determined from the 200 crystalline plane is often related to the width of cellulose microfibrils [ 47 ]. As shown in a previous study [ 13 ], increased pretreatment severity resulted in greater crystallite width (Fig. 5 A). This can be explained by the removal of less ordered cellulose and hemicellulose, with consequent aggregation of cellulose microfibrils. Fig. 5 Crystallite size ( A ) and crystallinity index ( B ) of steam-pretreated spruce samples before and after enzymatic hydrolysis, with or without Ta LPMO9A, determined by WAXS analysis Here, enzymatic hydrolysis did not affect crystallite width but, instead, drastically reduced cellulose crystallinity (Fig. 5 ). Considering that hydrolysis of crystalline cellulose proceeds as surface erosion by processive cellulases [ 46 ], a decrease in crystallite width was expected. A possible explanation is that, by the end point of hydrolysis (Fig. 5 A), fiber areas that were accessible to the enzymes were degraded completely, but the more crystalline domains were unaltered and, hence, did not decrease in size. Next, cellulose crystallinity before and after enzymatic hydrolysis was analyzed (Fig. 5 B). No statistically significant changes in crystallinity were detected during hydrolysis of STEX 210°C/auto . However, in the other two materials, the impact of the enzymatic hydrolysis on the cellulose crystallinity was pronounced. In the case of STEX 210°C/HAc with Celluclast + Novozym 188, the crystallinity index decreased from 51 to 12%, becoming negative when Ta LPMO9A was supplemented. Similar results were obtained with STEX 210°C/SO2 . A negative value of the crystallinity index indicates that the relative amount of crystalline regions dropped below that of less ordered domains (see Eq. 1 ). Collectively, WAXS data indicate that more severe pretreatment facilitates cellulose accessibility to enzymes (both hydrolytic and oxidative). More interestingly, the scattering data are a direct proof that supplementation with Ta LPMO9A promotes a decrease in cellulose crystallinity, most likely owing to the enzyme’s direct (i.e., oxidative degradation) and indirect (i.e., creating new access points for processive cellulose hydrolases) impact. In support of this mechanistic hypothesis, the largest drop in crystallinity was observed with material subjected to the harshest pretreatment conditions (STEX 210°C/SO2 ), which coincided also with the highest Glc4gemGlc release (Fig. 4 )."
} | 5,176 |
36605741 | PMC9769094 | pmc | 8,263 | {
"abstract": "Effective protection of soil fungi from predators is crucial for their survival in the niche. Thus, fungi have developed efficient defence strategies. We discovered that soil beneficial Mortierella fungi employ a potent cytotoxin (necroxime) against fungivorous nematodes. Interestingly, this anthelminthic agent is produced by bacterial endosymbionts ( Candidatus Mycoavidus necroximicus) residing within the fungus. Analysis of the symbiont's genome indicated a rich biosynthetic potential, yet nothing has been known about additional metabolites and their potential synergistic functions. Here we report that two distinct Mortierella endosymbionts produce a novel cyclic lipodepsipeptide (symbiosin), that is clearly of bacterial origin, but has striking similarities to various fungal specialized metabolites. The structure and absolute configuration of symbiosin were fully elucidated. By comparative genomics of symbiosin-positive strains and in silico analyses of the deduced non-ribosomal synthetases, we assigned the (sym) biosynthetic gene cluster and proposed an assembly line model. Bioassays revealed that symbiosin is not only an antibiotic, in particular against mycobacteria, but also exhibits marked synergistic effects with necroxime in anti-nematode tests. By functional analyses and substitution experiments we found that symbiosin is a potent biosurfactant and that this particular property confers a boost in the anthelmintic action, similar to formulations of therapeutics in human medicine. Our findings illustrate that “combination therapies” against parasites already exist in ecological contexts, which may inspire the development of biocontrol agents and therapeutics.",
"conclusion": "Conclusion In competitive environments such as soil, elaborate protection strategies are necessary for inhabitants to assert spatial claims against predators and competitors. Particularly effective are mixtures of metabolites, which exhibit synergistically acting defensive activities. 54–60 In this study, we report the discovery of the endobacterially produced metabolite symbiosin (3) and demonstrate how it is utilized in fungal host protection to enhance the anthelminthic effect of necroxime (2), illustrating how “combination therapies” are applied in an ecological context. Our data show that the endosymbiont of M. verticillata NRRL 6337 not only produces the toxin necroxime (2), but also synthesises a cyclic lipodepsipeptide (3), which significantly enhances the anthelmintic activity of the toxin 2. Our experiments demonstrate that the protection of the fungal host, even against fungivorous predators, works at physiological concentrations of the two compounds measured in fungal cultures. The production of secondary metabolites by endofungal bacteria instead of the host fungus has previously been demonstrated for other fungal symbionts. One example is the compound rhizoxin, a virulence factor employed by the plant-pathogenic fungus Rhizopus microsporus . 32 The discovery of endosymbionts ( Mycetohabitans rhizoxinica syn. Burkholderia rhizoxinica ) as the true producers of this toxic metabolite was unforeseen but illustrates the benefits of an endosymbiont for fungi. The toxin is not only the major virulence factor, but additionally plays a role in fungal protection. 61 Notably, also other Burkholderia -derived metabolites, such as the toxin rhizonin or the antibiotic icosalide, were originally thought to be fungal metabolites and later proven to be of bacterial origin. 62–64 Although functions of these and other endobacterial metabolites include symbiosis promoting activities, 25,65–67 host reproduction control 68 and pathogenicity causing traits, 32,63,69 research on their role in host protection strategies against predators was only conducted recently. 33,61 Specifically the combination of protective metabolites in M. verticillata NRRL 6337 plays an extraordinary role, as they are the first metabolites from one endofungal bacterium that synergistically provide host protection. Examples of effective symbiotic defense strategies are widespread among the different kingdoms, including bacteria, fungi, plants and animals. 25,32,54,60 They play important roles in protecting plants, 55 guarding fungal gardens in termite or leaf-cutter ant communities, 58,59 providing an antimicrobial shield for egg clutches of solitary wasps 54 and beetles, 60 or protection against grazing of marine sponges. 57 In these ecological contexts especially the combination of different complementary substances provides enhanced protection against a range of potential threats. The synergistic activity of combined bioactive molecules is a known effect in pharmaceutical research and especially useful for the treatment of certain difficult to treat infections or viral infections. 70–72 It was shown that not only the therapeutic selectivity and efficacy are improved by the combination of synergistically acting drugs, 73,74 but also the side effects of a medication can be reduced, if the concentration of therapeutics can be lowered. 75 Similar combined effects are also likely to occur in ecological settings. While synthetic compounds have been screened for synergistic biocontrol agents against nematodes, 76–79 to date little is known about synergistic protection against parasites such as nematodes in the soil. Several studies identified plant-derived compound mixtures to have anthelmintic activities with synergistic effects potentially protecting plants from nematodes, 80–82 yet synergistic effects of nematocidal small molecule protectants from bacteria or fungi were previously unknown. This study illustrates the first case of anthelmintic protection of a fungal soil habitant using the combination of a bacterium-derived toxin and biosurfactant. In particular the usage of biosurfactants as activity-enhancing or drug-delivery promoting substances was discussed in recent studies as effective improvements of pharmaceutical formulations. 83–85 Our study demonstrates that the principle of combinatorial therapies, common in medical contexts, is also already used in natural contexts, and that further studies of the boosting effects of biosurfactants might be useful in the future developments of combinatorial biocontrol strategies.",
"introduction": "Introduction Symbiotic associations are widespread among all kingdoms, shaping not only the lifestyle of the involved partners, but also the surrounding environment and ecological systems. 1–4 Omnipresent in all of the world's habitats, in marine environments, flora and soil biotopes, symbioses can influence the diversity and composition of species in an ecological community and thus play a central role in the development and maintenance of an ecological system. 4–8 In mutualistic associations, partnerships that are beneficial to all symbionts, different organisms live together and combine their individual skills to promote assertiveness or supply nutrients to the alliance. 3,9,10 While one partner may provide the food supply, 11,12 the other partner may possess the genomic abilities to biosynthesize a selection of natural products, such as communication molecules, UV-protectants or antibiotics. 13–17 A particularly important role is played by the diverse defense molecules provided by symbionts against competing bacteria, fungi or even higher organisms that can protect the host from predators. Many studies have shown that symbioses are valuable sources of natural products with pharmaceutically relevant activities such as antibiotics or anti-cancer compounds. 16,18–22 The molecules not only provide a rich source of new, efficient drug candidates, but can teach us strategies which can be adapted in agriculture or medicine. Among the different symbiotic lifestyles are associations between fungi and bacteria, with endosymbiotic interactions being the most intimate examples. 23–26 In these partnerships the bacteria live inside the fungal hyphae and benefit from a steady environment and nutritional support. Meanwhile, the fungus can be influenced in its reproduction, 27–29 growth behavior, 30 energy dynamics, 31 and by host-supporting secondary metabolites biosynthesized by the bacterium. 32,33 Symbioses between fungi and endobacteria are most abundant in the fungal phylum Mucoromycota, harboring Burkholderia -related bacteria. 26,34 Beside the well-studied symbiosis between Rhizopus microsporus and Mycetohabitans rhizoxinica , where the endosymbiont produces the highly cytotoxic macrolide rhizoxin, 32 a high prevalence of endosymbiotic bacteria among the Mortierellomycotinan fungi was reported. 34 Previous studies regarding the existence of endosymbionts in different Mortierella strains revealed the presence of bacteria in 37% of the tested species. 34 Among the known endosymbiont harboring and soil beneficial Mortierella strains is Mortierella verticillata NRRL 6337. We recently discovered that the antihyperlipidaemic, but also highly cytotoxic necroximes C and D (1 and 2; syn. CJ12.950 and CJ13.357) 35 ( Fig. 1A ), formerly believed to be fungal metabolites, are not produced by the fungus but its endosymbiotic bacteria. 33 Moreover, the necroximes exhibit anthelmintic activity and function as potent protectants against nematodal micropredators. 33 Fig. 1 Natural products of Candidatus Mycoavidus endosymbionts inhabiting M. verticillata strains. (A) Structure of necroxime C (1) and D (2), produced by the endosymbiont of M. verticillata NRRL 6337 ( Ca. Mycoavidus necroximicus). (B) Symbiotic (left) and cured (right) cultures of M. verticillata NRRL 6337 and SF9855. (C) Fluorescence microscopy pictures and the corresponding brightfield pictures (of M. verticillata SF9855 with its endosymbiont (left) and as a cured strain without endosymbionts (right)). (D) Metabolic profiles of the two symbiosin producer strains and their corresponding cured (cur.) strains. (E) Phylogenetic relationships of symbiosin producer strains, additional endosymbiotic bacteria from M. verticillata screened in this work ( Candidatus Mycoavidus spp.), and other fungi. Genome analysis of the necroxime-producing endobacterium Candidatus Mycoavidus necroximicus revealed 18 additional biosynthetic gene clusters for secondary metabolites, for which no corresponding metabolites have been identified. 33 This high metabolic potential is unusual among Mycoavidus endosymbionts, 33 and other symbiont metabolites that may have additional or synergistic protective effects for the host fungus are unknown. Here we show that obligate endosymbiotic bacteria of two M. verticillata strains produce a novel cyclic lipodepsipeptide, symbiosin (3). We report that 3 not only acts as an antimycobacterial agent, but also boosts the toxic effect of necroxime D against nematodes and thereby enhances the protective effect of the bacterial metabolites. Our findings illustrate that “combination therapies”, known from human medicine, also occur in an ecological context as a strategy to shield fungi from predators.",
"discussion": "Results and discussion Discovery of a novel cyclic depsipeptide from a fungal-bacterial symbiosis To investigate the biosynthetic potential of bacterial endosymbionts of M. verticillata , we monitored the metabolic profiles of seven fungal strains for which we have verified the presence of Mycoavidus bacteria by PCR (16S rDNA). 33 We varied culture conditions and media, and analyzed the culture extracts by high-performance liquid chromatography (HPLC). In order to assign specialized metabolites to the endosymbionts, we compared symbiotic M. verticillata strains with symbiont-free (cured) M. verticillata strains ( Fig. 1B and C ). In the culture broths of two strains, M. verticillata NRRL 6337 and M. verticillata SF9855, we detected a previously unknown compound (3), named symbiosin ( Fig. 1D and S1, ESI † ). Using high-resolution electrospray ionization mass spectrometry (HRESI/MS) we assigned a mass of 962.5 Da to 3 and deduced its chemical formula of C 49 H 70 N 8 O 12 (calcd m / z 963.5186 [M + H] + and found m / z 963.5184 [M + H] + ). The MS/MS fragmentation pattern indicated a peptide backbone. Retention times, exact masses and MS/MS fragmentation patterns of the metabolites detected in the symbiotic M. verticillata NRRL 6337 and M. verticillata SF9855 cultures proved to be identical. Interestingly, a phylogenetic analysis of amplified 16S rDNA of the endosymbionts shows that the symbionts associated with a symbiosin-positive phenotype are more closely related to each other than to the other fungal endosymbionts of M. verticillata ( Fig. 1E , adapted from previous findings). 33 Notably, the new compound could not be detected in cultures of the sterile fungal strains lacking the endosymbiotic Ca. Mycoavidus strains ( Fig. 1D ), suggesting that either symbiosin is produced by the endobacteria or the presence of the endosymbionts triggers symbiosin production in the fungal host. To characterize the new metabolite, we subjected the ethyl acetate extract of a five week-old holobiont culture (4 L) to size-exclusion chromatography with Sephadex LH-20, followed by preparative HPLC, yielding 8.8 mg of pure 3. By a combination of 1D- and 2D-NMR measurements, LC-HRESI/MS, hydrolysis and LC-HRESI/MS/MS fragmentation, we elucidated the structure of 3 ( Fig. 2A ). The number of proton and carbon signals measured in 1 H and 13 C NMR experiments supports the chemical formula deduced from HRESI/MS data. 13 C NMR and DEPT135 measurements identified 13 quaternary carbon atoms, 16 methines, 18 methylenes, and two methyl groups. Additional signals in the 1 H spectrum indicated the presence of seven primary amide protons. 1 H– 1 H COSY spectra in combination with HMBC couplings revealed the amino acid sequence of Gln, Thr, β-Ala, Trp, Ser, and Tyr with an ester bond between the carbonyl-group of Tyr and the hydroxy-group of Thr. Furthermore, we found that 3-hydroxy-myristic acid is attached to the N-terminus of Gln ( Fig. 2A and Table S10, ESI † ). By means of Marfey's method we determined the absolute configurations of the amino acids, d -Gln, l -Thr, d -Trp, l -Ser and d -Tyr. Mosher esterification followed by HPLC analysis revealed the 3 R -configuration of the hydroxy fatty acid ( Fig. 2B, C and S4, ESI † ). Taken together, symbiosin is a previously unknown compound belonging to the family of cyclic lipodepsipeptides. Fig. 2 Structure elucidation of symbiosin. (A) Structure of symbiosin (3) with key 1 H– 1 H COSY and 1 H– 13 C HMBC couplings from 2D-NMR experiments. (B) Absolute configuration of 3 and used analysis methods. (C) Chromatographic profiles of Mosher ester analysis for configuration elucidation of the hydroxy myristic acid residue (myr.) in symbiosin (sym). Bacteria-produced symbiosin resembles fungal metabolites Interestingly, symbiosin is structurally similar to the known natural products colisporifungin (4), ophiotine (5), verruculin (6), and aselacin A (7) ( Fig. 3A and B ). 36–39 It is remarkable that all of these compounds (4–7) have been isolated from fungi of the phylum Ascomycota, which are not described to harbor endosymbionts. Except for the terminal amino acids that are involved in lactone ring formation, the peptide backbones of lipopeptides 3–7 are almost identical. The main differences between these lipopeptides are notable in the fatty acid side chains. In contrast to 4–7, 3 has a β-hydroxy fatty acid attached to the extracyclic glutamine residue ( Fig. 3B ). The presence of a β-hydroxy fatty acid is a hallmark of lipopeptides from Gram-negative bacteria, 40 thus implicating endobacteria as the source of 3. Validations of the biosynthetic assembly line of 4–7 are not possible, as no genomic data of the fungi are available. Fig. 3 Structural similarities between symbiosin and related fungal lipopeptides. (A) Structures of compounds similar to symbiosin (3) (colisporifungin (4), ophiotine (5), verruculin (6), and aselacin A (7)) produced by Ascomycota. Structural differences are colour-coded. (B) Comparison of fatty acid and amino acid sequences of 3 and similar compounds. Genetic origin and model of symbiosin biosynthesis To support the assumption that bacteria are the producers of 3, and to rule out a fungal biosynthesis, we searched for the gene cluster coding for the symbiosin assembly line. First, we used fungal antiSMASH to search for a possible biosynthetic gene cluster in the fungal genome of M . verticillata NRRL 6337. 41 Although genes encoding NRPSs could be identified, the deduced assembly lines do not fit the structure of 3 (Fig. S5 † ). Thus, we turned to the endobacterial genomes. We reasoned that the discovery of the candidate gene clusters would be facilitated by comparison of the endobacterial genomes of both symbiosin-positive symbioses. Therefore, we isolated and sequenced the genomic DNA of the Candidatus Mycoavidus sp. of M. verticillata SF9855 (GenBank: CP102085) in a similar way as previously performed for Ca. M. necroximicus. 33 The availability of two related genome sequences proved to be helpful in the reassessment of the genome of Ca. M. necroximicus (GenBank: CP076444) and the assembly of related contigs by means of Sanger sequencing allowed us to rectify the genome sequence. The average nucleotide identity of both genomes is 96.91%, meaning that they can be considered the same species. 42 Mining of the endosymbiont genome sequences revealed several putative NRPS gene clusters ( Fig. 4A ). Among the deduced NRPS-type assembly lines, we identified one in both genomes that is the best candidate for the biosynthesis of 3 ( Fig. 4B and S7, ESI † ). Fig. 4 Secondary metabolites encoded in bacterial genomes and the model of symbiosin (3) biosynthesis. (A) Comparison between the genomes of Ca. M. sp. SF9855 (GenBank: CP102085) and Ca. M. necroximicus (GenBank: CP076444). (B) Symbiosin biosynthesis gene cluster in the genomes of Ca. M. sp. SF9855 and Ca. M. necroximicus; hypo: hypothetical protein. (C) Proposed assembly line for symbiosin. FA: fatty acid; C: condensation domain (S: starter; dual: condensation/epimerization); A: adenylation domain; T: thiolation domain; E: epimerization domain; TI: TIGR01720 domain, domain of unknown function; TE: thioesterase domain. The deduced sym NRPS consists of seven modules. In silico prediction of the adenylation (A) domain specificities 43 indicated that the first six modules would produce a heptapeptide composed of Gln, Thr, β-Ala, Trp, Ser, and Tyr, which is in full agreement with the hexapeptide backbone of 3 (Tables S3–S6, ESI † ). The A domains of these modules have a similarity between 64.5% and 82.8% in both strains (Table S4, ESI † ). The presence of a seventh NRPS module is, however, surprising. Scrutinizing the amino acid sequence of this terminal module indicated several variations in conserved core motifs (Table S4 and Fig. S8, ESI † ) and the complete loss of a flavodoxin-like A subdomain with catalytically important core residues in one of the deduced NRPS sequences (Fig. S8 and S9, ESI † ). 44,45 The identity between these extra modules is high at 96.36%, whereas the identities to all other domains are between 38.7% and 44.7% (Table S4, ESI † ). Thus, we concluded that module seven does not incorporate any additional amino acids. There has been precedence that such catalytically non-functional NRPS modules are skipped. 46,47 A particularly valuable indicator for the identity of the sym NRPS is the β-Ala specificity of module 3. We determined the Stachelhaus-code sequence and active site residues proposed for β-Ala specific domains (Fig. S10, ESI † ). 48,49 Specifically, the position of the aspartate residue, which is usually conserved in A domains in the A4 motif of α-amino acids (FDxS), differs in all described β-Ala specific A domains (Fig. S10, ESI † ). 44,49 This deviation is plausible because the negatively charged carboxy group of this aspartate residue interacts with the amino group of the incorporated amino acid, 44 and the spatial arrangements of the α- and β-amino groups clearly differ in the active site. In addition to the A domain specificities of the other six modules, the condensation (C) domains of the deduced modules fit the experimentally determined structure and absolute configuration of 3 ( Fig. 4C ). Specifically, the C Starter domain would load ( R )-3-hydroxy-myristic acid, as these domains are known to introduce fatty acids onto the C-terminus of initiating NRPs. 50 The dual condensation/epimerization C domains (C Dual ) in modules 2 and 5 are responsible for the epimerization of the prior amino acid, resulting in d -Gln and d -Trp. 51 A third C Dual domain is found in module 4, which would act on β-Ala. Since β-Ala has no stereocenter, no epimerization can take place on this amino acid. In the case of d -Tyr, an additional epimerization domain changes the configuration of the introduced l -amino acid building block into the d -isomer. The remaining L C L -domains in modules 3 and 6 are in accordance with the determined configurations of l -Ser and l -Thr. Although it is impossible to rigorously verify the gene cluster assignment by functional gene analysis in the as-yet unculturable symbionts, the genomic and bioinformatic analyses provide strong evidence for the identity of the sym gene cluster encoded in the bacterial genome. Symbiosin is an antimycobacterial agent To identify potential biological functions of the symbiont-derived lipopeptide, we subjected 3 to a panel of whole-cell bioassays using representative bacterial and fungal strains, as well as cancer cell lines. No cytotoxicity was observed on HeLa cells or HUVEC cells and only a moderate antiproliferative effect on K-562 (37.5 μM) was observed (Table S7, ESI † ). In an initial antimicrobial assay 3 showed moderate activity against several bacterial strains, including Bacillus subtilis 6633B1, Staphylococcus aureus 511B3 and vancomycin-resistant Enterococcus faecalis 1528R10, and was found to be particularly active against Mycobacterium vaccae (Table S8, ESI † ). Thus, we tested a range of mycobacteria and determined MIC values of 3 against M. vaccae (6.49 μM), Mycobacterium smegmatis (6.49 μM), and Mycobacterium aurum (12.98 μM). Interestingly, anti-mycobacterial activities, which have been evaluated for 5 and 6, were only reported for 6, 39 which shares tyrosine with 3 as the macrocyclic ring-forming amino acid. Synergistic effect of bacterial metabolites protects the fungal host from nematodes Since the structurally related 5 has moderate nematocidal activities, 37 and because Ca. M. necroximicus has been found to protect the host from nematode attacks, 33 we next tested the anthelmintic activity of 3. Surprisingly, 3 alone showed no effect on the model nematode Caenorhabditis elegans (concentrations up to 100 μg mL −1 ). To evaluate a potential synergistic effect with the necroximes, we evaluated the combined activity of necroxime (2) on C. elegans in the presence of three different concentrations of 3 (0.2 μg mL −1 , 2 μg mL −1 and 20 μg mL −1 ; notably, 2.2 μg mL −1 corresponds to the amount of 3 isolated from fungal holobiont cultures). Therefore, we incubated the nematodes and their bacterial food source in liquid media with increasing concentrations of 2, while keeping a steady concentration of 3. Subsequently, we measured the OD 600 of the bacterial suspension, which is an indirect measurement for nematode viability. 52 The IC 50 of necroxime alone was determined to be 10.78 μg mL −1 (11.4 μg mL −1 , 33 95% confidence interval 9.83–13.00 μg mL −1 ; Fig. 5A ). If necroxime (2) was combined with 3, the anthelmintic activity increased with IC 50 values of 6.22 μg mL −1 , 5.79 μg mL −1 and 4.13 μg mL −1 necroxime, if 0.2 μg mL −1 , 2 μg mL −1 or 20 μg mL −1 of 3 was present, respectively ( Fig. 5A, B and S11, ESI † ). Fig. 5 Biological activities. (A) Synergistic anthelmintic effect of necroxime D (2) and symbiosin (3). Nematode viability measurements in the presence of 3, different concentrations of 2 and of 2 in combination with 0.2 μg mL −1 , 2 μg mL −1 or 20 μg mL −1 3. (B) Differences in IC 50 measured for necroxime alone (IC 50 : 10.78 μg mL −1 ; 23.54 μM, 95% CI 20.67–26.79 μM) and in combination with 3 in concentrations of 0.2 μg mL −1 (IC 50 : 6.22 μg mL −1 ; 13.58 μM, 95% CI 11.80–15.57 μM), 2 μg mL −1 (IC 50 : 5.79 μg mL −1 ; 12.64 μM, 95% CI 11.07–14.38 μM) and 20 μg mL −1 (IC 50 : 4.13 μg mL −1 ; 9.01 μM, 95% CI 6.76–11.70 μM). CI, confidence interval; *, p < 0.05; ***, p < 0.001. (C) Biosurfactant activity of 3. 1% Tween (positive control); 1 mM 3. (D) Activity of 3 alone (20 μg mL −1 ) and synergistic activity of 2 in different concentrations in combination with 3 against the fungivorous nematode Aphelenchus avenae . The numbers of harvested nematodes are relative to the numbers of nematodes harvested from cultures without 2 and 3. *, p < 0.05; **, p < 0.01; ***, p < 0.001; ****, p < 0.0001. Considering that 3 alone has no effect on the nematodes, we wondered how this synergistic effect is achieved. The structure of 3, consisting of a lipophilic fatty acid residue and partially hydrophilic amino acids, suggested that it acts as a biosurfactant, which might influence the permeability of substances into the nematodes. We verified the predicted biosurfactant activity of 3 by means of drop collapse assays ( Fig. 5C ). To test whether the synergistic effect of 3 is based on its tenside activity, we exchanged 3 with another lipodepsipeptide biosurfactant. Therefore, we selected surfactin (8), as it is a strong biosurfactant with a similar size (1036.3 Da) and a cyclic lipodepsipeptide with a structure similar to that of 3. 53 Using a steady concentration of 2.0 μg mL −1 8 and increasing concentrations of 2, we determined an IC 50 of 6.13 μg mL −1 for the combination of 2 and 2.0 μg mL −1 8 (5.92 μM), which is comparable to the results of 2 and 2.0 μg mL −1 3 (6.01 μM; Fig. 5B and S11, ESI † ). Thus, the biosurfactant activity is a plausible reason for the synergistic activity of 3. Based on the synergistic anthelmintic effects demonstrated with the model nematode C. elegans , it appeared feasible that the increased anthelmintic effect could also play a role in the fungal protection against fungivorous nematodes, especially as 2 was previously shown to be active against the fungivorous nematode Aphelenchus avenae . 33 Therefore, we next tested if the presence of 3 affects the nematodes or enhances the protective effect of 2 against A. avenae . We compared the number of harvested nematodes from nematode-fungus co-cultures of a necroxime- and symbiosin-negative M. verticillata strain that was treated with 3 or varying amounts of 2, to cultures that were treated with a combination of 2 and 3. This experimental setup enabled us to quantify the relative reduction of fungivorous nematodes in cultures containing 2 , but even more clearly it showed a significant reduction of nematodes in cultures containing a combination of 2 and 3. Already low concentrations of 2 (5.7 μg mL −1 ) were sufficient in the combination with physiological amounts of 3 (2.0 μg mL −1 ) to almost completely eradicate the presence of nematodes ( Fig. 5D ), which is comparable with reported numbers for M. verticillata NRRL 6337 wild-type cultures. 33 In contrast, experiments with 3 alone did not show any effect on nematode numbers, even when tested with concentrations up to 20 μg mL −1 ( Fig. 5D ). These results show that 2 and 3 synergistically provide protection to the fungus against fungivorous nematodes."
} | 6,954 |
25227415 | PMC4165979 | pmc | 8,264 | {
"abstract": "Biotic interactions can improve agricultural productivity without costly and environmentally challenging inputs. Micromonospora strains have recently been reported as natural endophytes of legume nodules but their significance for plant development and productivity has not yet been established. The aim of this study was to determine the diversity and function of Micromonospora isolated from Medicago sativa root nodules. Micromonospora- like strains from field alfalfa nodules were characterized by BOX-PCR fingerprinting and 16S rRNA gene sequencing. The ecological role of the interaction of the 15 selected representative Micromonospora strains was tested in M. sativa . Nodulation, plant growth and nutrition parameters were analyzed. Alfalfa nodules naturally contain abundant and highly diverse populations of Micromonospora , both at the intra- and at interspecific level. Selected Micromonospora isolates significantly increase the nodulation of alfalfa by Ensifer meliloti 1021 and also the efficiency of the plant for nitrogen nutrition. Moreover, they promote aerial growth, the shoot-to-root ratio, and raise the level of essential nutrients. Our results indicate that Micromonospora acts as a Rhizobia Helper Bacteria (RHB) agent and has probiotic effects, promoting plant growth and increasing nutrition efficiency. Its ecological role, biotechnological potential and advantages as a plant probiotic bacterium (PPB) are also discussed.",
"conclusion": "Conclusion In this study 66 Micromonospora strains were isolated, characterized using BOX-PCR and sequencing of 16S rRNA genes and selected some of them for studying their interaction with alfalfa. Our results, together with those from other authors, indicate that Micromonospora are ubiquitous in legume root nodules, presenting a very high genetic diversity. Most of them exhibit in vitro a great ability to degrade organic polymers as well as presenting a direct mechanism for plant growth promotion (IAA production). We have shown that Micromonospora could play an important ecological role in interaction with the host plant by enhancing aerial growth and nutrient contents, being an increase of N uptake by the plant a general phenomenon in the Micromonospora -alfalfa interaction. It remains to be elucidated whether these positive effects also occur in other plant species. Micromonospora engaged in tripartite interactions with E. meliloti 1021 and alfalfa increase nodulation, and some of their strains can also significantly promote the growth and nutrition of N 2 -fixing plants. Contrary to most of plant growth-promoting bacteria, beneficial effects of Micromonospora do not rely on induction of plant root growth. All the above data suggests that, in general, Micromonospora can be considered as excellent PPB, although a correct selection of strain is of capital importance because of the detrimental effect that some Micromonospora may have for plant growth (i.e. strain AL4 in non-nodulated plants; Table 3 ). Aditionally, Micromonospora is a sporulating bacterium so that it can endure in soil and harsh environments. Thus, some of their strains seem to be excellent candidates for the production of bioinoculants, which would make the use of environmentally unfriendly chemical fertilizers less intensive in a broad range of agroecosystems.",
"discussion": "Discussion The number and diversity of Micromonospora strains recovered from alfalfa nodules strongly suggest that this actinobacteria is commonly associated with the symbiotic organ of legumes. Besides other microorganisms, almost all nodules selected had a population of one or more Micromonospora strains. Moreover, for each isolation experiment two sterile, non-crushed nodule was rolled over YMA agar and incubated under the same conditions as the homogenized samples in order to assess the effectivity of the sterilization procedure. No colonies appeared on any YMA plate indicating that sterilization was effective. BOX–PCR fingerprinting has been shown to be a useful tool to discriminate highly related strains and has been applied to study the genetic diversity of different bacterial taxa, including Micromonospora 30 31 . A high degree of genetic variation was observed among the 66 isolates, when analysed by BOX–PCR fingerprinting, indicating that they represented different bacterial genotypes. It should also be noted that none of them were clones of one another, supporting the idea of the existence of high genetic diversity among Micromonospora strains in legume root nodules. Our results are coherent with data from Lupinus angustifolius and Pisum sativum root nodules 13 15 . The BOX grouping provided a useful background for determining the taxonomic relationship of the strains isolated since these groups served to select strains for 16S rRNA gene sequencing. Even though several strains have more than 99% similarity with described species, others had 16S rRNA gene sequence similarities below 99%, indicating that they were not related to any of the already known species of Micromonospora and probably represent new ones. This case has been observed previously by Trujillo and co-workers, who described two new Micromonospora species: M. lupini and M. saelicesensis , whose 16S rRNA genes were highly similar to already described species 12 . Our results (BOX–PCR fingerprinting and 16S rRNA gene sequences) also suggest that the diversity of Micromonospora was independent of the location where they were isolated since in several BOX-PCR groups there are strains from more than one of the five locations sampled ( Figure 1 ). The high diversity and ubiquity of this actinobacteria inside legume root nodules suggest that its presence is not fortuitous, but that Micromonospora might have an important ecological role in nature. To discern ecological roles of Micromonospora in interaction with alfalfa, we evaluated in a mesocosm experiment their effects on plant growth and nutrition, both in nodulated and non-nodulated alfalfa plants. Multivariate statistics showed that E. meliloti 1021-nodulated plants tended, of course, to have higher values of aerial dry matter and nutrient content than the non-nodulated ones ( Figure 3a ). But we also found a significant effect of the inoculation with Micromonospora as well as a significant interaction between both E. meliloti 1021 and Micromonospora inoculations, indicating different behaviour of the Micromonospora strains according to the nodulation status of the plant ( Table S4 ). In non-nodulated plants, twelve out of the fifteen Micromonospora strains produced significant multivariate differences with respect to the uninoculated control ( Table S5 ; Figure 3b ), while in nodulated plants only the treatments co-inoculated with the strains ALFb5 and ALFpr18c differed significantly from the Micromonospora -free control treatment ( Table S5 ; Figure 3c ). Several actinobacteria, including strains of Micromonospora sp., had been shown to promote both shoot and root growth and nodulation in alfalfa as well as in the actinorhizal plant species Ochetophila (Discaria) \n trinervis when co-inoculated with the corresponding nitrogen-fixing micro-symbiont, Ensifer or Frankia 32 33 34 . Contrary to our results, these authors found that actinobacteria alone exerted no effect on plant growth. It should be emphasized that, unlike us, Solans and co-workers grew the plants in soilless, gnotobiotic conditions 32 33 34 . Three main empirical facts from our greenhouse experiment must be highlighted. The first one is that Micromonospora does not induce larger root systems. The most common effect of PPBs on plant is the formation of larger root systems, which allow exploring a greater volume of soil for water and nutrients 35 . Root biomasses (Rdw) in Micromonospora -inoculated treatments were similar to or lower than in control plants ( Table 3 ; Figure 3a,b ). The second fact is that most Micromonospora strains increased the N shoot content in non-nodulated plants ( Table 3 ; Figure 3b ). It has been reported in recent years the existence of putative N 2 -fixing Micromonospora 13 36 . Therefore, in a previous work we conducted an exhaustive experimental study centred on two of the representative strains here assayed, namely ALFb5 and ALFpr18c, in order to discern if they could fix N 2 either as free-living diazotrophs or in symbiosis with alfalfa plants 37 . Neither of the strains grew in nitrogen-free media or reduced acetylene under micro-aerobic conditions. Incorporation of 15 N into the microbial biomass or alfalfa tissues was not detected. Also, attempts to amplify putative nif H genes in these strains were unsuccessful. Besides, we tracked the presence of structural genes for N 2 -fixation in two other Micromonospora strains that have their genome sequenced 38 39 , finding no evidence for them 37 . These results seem to rule out N 2 fixation by Micromonospora as source of nitrogen for plants, focusing the explanation for higher N shoot contents on enhanced nutrient uptake efficiency and/or more plant-available nitrogen in soil. Although there are few published studies on the impact of PPB on nutrient uptake systems, concomitant improvement of mineral nutrition (including N, P and K) and increase of root surface area has been described in several plant species 35 . With regards to N nutrition, it has been hypothesized that PPB could directly stimulate nitrate transport systems in plants 40 , but recent genetic studies on Arabidopsis thaliana indicate that while there are two NO 3 − transporter genes ( NRT2.5 and NRT2.6 ) that are strongly upregulated in response to inoculation with the PPB Phyllobacterium brassicacearum strain STM196, plant growth promotion is not linked to changes in NO 3 − uptake rate or NO 3 − distribution between roots and shoots 41 . However, most actinobacteria are saprophytes able to produce a wide range of extracellular hydrolytic enzymes 2 42 43 44 . All the strains we studied synthesize hydrolytic enzymes able to cleave complex nitrogen-containing polymeric substrates, such as caseinase and gelatinase ( Table 2 ), strongly suggesting that Micromonospora can favour plant nutrition by enhancing nitrogen mineralization in soils. Nonetheless, further research is needed to fully explain the rationale for improved nitrogen nutrition in plants inoculated with Micromonospora . Moreover, all the fifteen Micromonospora showed neutral and alkaline phosphatase activities ( Table 2 ), which can enhance the mineralization of organic phosphate in neutral or alkaline soils 45 like the one used in our greenhouse experiment (pH 7.47; Table S1 ), thus making soil P more available to plants as suggested by higher shoot P content in some Micromonospora -inoculated treatments than in the controls ( Table 3 ; Figure 3b, c ). In the cohort of plants nodulated by E. meliloti 1021 only two strains of Micromonospora (ALFb5 and ALFpr18c) produced statistically significant multivariate differences with respect to the Micromonospora -free control group ( Table S5 ; Figure 3c ). The success of the interaction between a PPB strain and the plant relies on a set of adaptation mechanisms by both partners, among which the phytochemical profile of the root exudates plays a fundamental role in the bacterial colonization of the root as well as in the regulation of PPB plant beneficial properties 46 . The composition of root exudates has been shown to differ in legumes depending on their nodulation status 47 48 49 , so that the biochemical environment in the rhizosphere of E. meliloti 1021-nodulated alfalfa plants might be less advantageous for Micromonospora compared with that of non-nodulated plants. Considering the soil pH, legumes are known to acidify the rhizosphere because of the release of protons following excess uptake of cations over anions during N 2 fixation 50 51 52 . Only six out of the 15 Micromonospora strains tested in this study grew vigorously in vitro at pH 6.5 ( Table 2 ) and none at lower pH values (4.5 or 5.5). Indeed, strains ALFb5 and ALFpr18c are among those able to grow at acidic pH (6.5) while the strain ALFr5, a strain that only excelled in the cohort of non-nodulated plants, does not ( Table 2 ; Figure 3b,c ). Therefore, a more acidic rhizosphere in the N 2 -fixing plants may partially explain differential effects of certain Micromonospora strains on growth of nodulated and non-nodulated plants. Nonetheless, given the great influence that the nodulation has on N acquisition capacity and growth of legumes, it is plausible that the effect of most of the Micromonospora is not marked enough to be statistically significant in symbiotically N 2 -fixing alfalfa plants. And third, although in plants nodulated by E. meliloti 1021 only Micromonospora ALFb5 and ALFpr18c were found to have significant, globally positive effects on plant growth and nutrition ( Table 3 ), it was observed a trend towards the improvement in nodulation intensity (number of nodules) with the presence of Micromonospora ( Figure 3c ; Figure S2c ). Furthermore, all the fifteen Micromonospora strains could be re-isolated from nodules of random plants of each co-inoculated treatment, suggesting that none of the Micromonospora strains here assayed had incompatibility with E. meliloti 1021. Plant growth-promoting bacteria can increase nodulation in legumes through different mechanisms, including the production or degradation of phytohormones involved in nodule initiation and organogenesis 53 , or by affecting the interaction between plant and rhizobia 54 55 56 . Auxins are involved in the initiation and normal development of both determinate 57 and indeterminate nodules, like Medicago root nodules 58 . IAA production has been associated with the induction of increased nodule numbers in Medicago truncatula plants inoculated with an E. meliloti strain that overproduces IAA 59 and also with nodule-like structures even in non-leguminous plants 60 61 62 . Moreover, rhizobial cellulases have been shown to be crucial for legume nodulation 63 . The ability of Micromonospora strains to produce cellulases could thus explain the increase in the number of nodules observed in co-inoculated plants compared to the control plants only inoculated with E. meliloti 1021. However, IAA production by Micromonospora may not be directly related in our study to an increase in nodulation despite of the literature. Conclusion In this study 66 Micromonospora strains were isolated, characterized using BOX-PCR and sequencing of 16S rRNA genes and selected some of them for studying their interaction with alfalfa. Our results, together with those from other authors, indicate that Micromonospora are ubiquitous in legume root nodules, presenting a very high genetic diversity. Most of them exhibit in vitro a great ability to degrade organic polymers as well as presenting a direct mechanism for plant growth promotion (IAA production). We have shown that Micromonospora could play an important ecological role in interaction with the host plant by enhancing aerial growth and nutrient contents, being an increase of N uptake by the plant a general phenomenon in the Micromonospora -alfalfa interaction. It remains to be elucidated whether these positive effects also occur in other plant species. Micromonospora engaged in tripartite interactions with E. meliloti 1021 and alfalfa increase nodulation, and some of their strains can also significantly promote the growth and nutrition of N 2 -fixing plants. Contrary to most of plant growth-promoting bacteria, beneficial effects of Micromonospora do not rely on induction of plant root growth. All the above data suggests that, in general, Micromonospora can be considered as excellent PPB, although a correct selection of strain is of capital importance because of the detrimental effect that some Micromonospora may have for plant growth (i.e. strain AL4 in non-nodulated plants; Table 3 ). Aditionally, Micromonospora is a sporulating bacterium so that it can endure in soil and harsh environments. Thus, some of their strains seem to be excellent candidates for the production of bioinoculants, which would make the use of environmentally unfriendly chemical fertilizers less intensive in a broad range of agroecosystems."
} | 4,111 |
31796765 | PMC6890687 | pmc | 8,265 | {
"abstract": "While natural communities are assembled by both ecological and evolutionary processes, ecological assembly processes have been studied much more and are rarely compared with evolutionary assembly processes. We address these disparities here by comparing community food webs assembled by simulating introductions of species from regional pools of species and from speciation events. Compared to introductions of trophically dissimilar species assumed to be more typical of invasions, introducing species trophically similar to native species assumed to be more typical of sympatric or parapatric speciation events caused fewer extinctions and assembled more empirically realistic networks by introducing more persistent species with higher trophic generality, vulnerability, and enduring similarity to native species. Such events also increased niche overlap and the persistence of both native and introduced species. Contrary to much competition theory, these findings suggest that evolutionary and other processes that more tightly pack ecological niches contribute more to ecosystem structure and function than previously thought.",
"introduction": "Introduction Historically, prediction and management in ecology have been thought to be constrained mostly by demographic and ecological processes, with evolution playing a much more limited role or happening on much longer time scales 1 . In light of recent empirical studies, it is now widely recognized that evolution and ecology (and their interplay) affect the response of communities to environmental changes, and that the two processes may happen on similar timescales 2 – 4 . This is especially true when environmental changes occur at large scale or have high amplitude (e.g., current global changes), as selective pressures can then act efficiently to alter natural selection processes and/or species’ coevolution. Empirical examples suggesting the importance of evolution on ecological timescales are however usually focused on either one species responding to changes in the abiotic environment 5 , 6 or on one interaction 7 , 8 . These studies strongly suggest evolution alters the fate of the studied population, which links the individual gene/organism level with population structure but leaves unanswered how such effects might propagate to higher levels of organization such as communities or ecosystems (Fig. 1 ). For example, such propagation may occur in more complex trophic modules when local evolutionary adaptation modifies community structure and ecosystem functioning by altering competitive interactions and enabling predators to more strongly reduce prey abundance 9 and when different intraspecific genetic variants of key species alter the structure and functioning of an ecosystem 10 , 11 . Analyses of complex food webs suggest that strong selection on fish’s life history due to fishing may cause evolutionary processes to decrease catch and destabilize the functioning of fishery ecosystems by selecting for small body size and early maturation of fished populations 12 . These observations link evolutionary processes to communities and ecosystem function 13 (Fig. 1 ). Figure 1 Evolution on ecological time scales influence populations and link contemporary evolution with community structure across levels of organization. Black arrows indicate more studied influences between adjacent organizational levels. For example, population genetics focuses on links above the dashed line between genes and phenotypes. The green arrow indicates the less studied links focused on here between evolutionary changes such as those in phenotype and more removed higher-level properties, such as community composition and food webs that may be less sensitive to evolutionary changes at lower levels. In general, such linkage between organizational levels may critically depend on the degree to which mutation and/or speciation introduces organisms into ecological systems that are functionally distinct from organisms that migrate into the system. Migration due to changes in the environment e.g., climate change 14 , stochastic dispersal processes, or human facilitation can rescue populations that might otherwise go extinct or add new functional groups to the community 15 , 16 . Both of these effects can alter other species’ densities or interactions. Evolution can also cause these effects by altering interactions, e.g. evolution of diet 17 , 18 , while also altering intraspecific phenotypic and genetic diversity 10 , 19 . We focus here on forms of evolution that may alter a population’s phenotype such as sympatric and parapatric speciation within local communities or ecosystems, and we focus on these local systems in terms of their trophic interactions, food webs, and the degree to which evolution and migration may alter the structure and robustness of food webs. The role of evolution in community assembly, such as new species being introduced via local speciation, has been explored more recently and less extensively than the role of invasions, i.e., introductions from other communities within a region 20 – 22 . Community assembly processes may vary from being almost completely dominated by introductions from nearby source pools over the short time scales to being largely dominated by speciation after long periods of time in isolated environments very far from source pools 23 . Several 16 , 24 – 28 but not all 29 eco-evolutionary theories of community assembly describe dynamics where new phenotypes are introduced and potentially establish within the community and alter the community’s network structure and dynamics depending on selective pressures formalized as ordinary differential equations. One type of community evolution model relies on a large number of biologically undefined phenotypic traits free of trade-offs 24 , 25 , 30 . Clearer interpretations emerge from another type of model that relies on one or a few biologically defined traits such as body size 28 , 31 – 35 which can produce persistent and empirically realistic networks 31 , 34 and elucidate how coevolution of species may affect community stability 34 , 36 , 37 . Fundamental similarities among these trait-based studies of community coevolution include the introduction of new phenotypes into a community and subsequent selection based on the population dynamics that emerge from ecological interactions structured according to traits of each phenotype or “species” within the community. While the introductions and dynamics of species in evolutionary and ecological community assembly models are similar, they differ with respect to the variability of species that are introduced 15 , 27 . Introductions in community evolution models typically mimic speciation or mutation events derived from ancestral species within the community by slightly modifying that ancestor’s traits 31 , 34 , 38 . This simulates phylogenetic conservation of niches 39 by typically introducing species whose function is similar to at least one species within the community in terms of life history and interactions. New functional groups may arise 31 , but most likely by progressive selection events operating on relatively similar species. By contrast, community assembly models typically introduce species whose traits such as preferences for prey and vulnerabilities to predators are relatively uncorrelated with traits of species already within the network. Such ecological assembly often adds more variability to the community including new functional groups more immediately than does evolutionary assembly. In both cases, phenotypes more or less similar to those already in the community are introduced and then subjected to selection determined by species interactions structured according correspondence of species’ traits. We use this framework of functional similarities and differences 15 , 40 to focus on how similarity of introduced species to those residing in the food web affects the robustness and structure of the assembled network 41 . We account for evolutionary processes responsible for phenotypic variation by assuming that niche conservatism 39 causes “speciation” events to introduce species whose niches are more similar to at least one of the species of the network while “invasion” events introduce species whose niches are typically more different from species within the network. We focus on two main questions: (1) Are food webs assembled through invasions more or less persistent compared to food webs assembled by speciation events? (2) Is the structure of “invasion” networks more or less similar than “speciation” networks to empirically observed networks? Based on competitive exclusion principles that assert species are less likely to persist the more they share niches with resident species 15 , 40 – 42 , one expects that species introduced via speciation events cause more extinctions and are therefore less persistent than introductions of invaders who share less of their niche space with resident species. However, invasions may cause a larger disturbance than speciation events due to the greater functional difference between invaders and resident species. Therefore, one may expect that “invasion” webs should be less persistent. Our results support this latter prediction and indicate that the mechanism can be more finely understood by accounting for distribution of vulnerability and generality of the newly introduced type. Our results also show that “speciation” networks more closely resemble empirical food webs. These findings have important implications for invasion ecology and co-evolution within complex communities 22 . We conducted our simulations in three steps. First, we generated three sets of realistically structured networks 43 with 35 species with low, medium, and high levels of connectance (fraction of all possible links realized) along with three corresponding sets of species whose traits such as trophic niche width enabled them to be introduced and trophically linked to species in each web with minimal methodologically enforced changes to their connectance (Fig. 2 ). Allowing introductions of species from different connectance classes would systematically change connectance levels during introduction sequences. We avoid such changes to better represent an ecological realistic range of connectance, which may reflect different habitats or regional biotas with systematically different biotas. For example, soil food webs appear to have unusually high connectance 44 and contain species unable to survive in lakes with lesser connectance 45 or in above ground communities found to have much lower connectance 46 . We avoid distinguishing connectance classes further because results at each connectance level are qualitatively identical and quantitatively very similar (Fig. S1 ). Our second step used allometric trophic network models 47 – 49 to simulate species’ population dynamics within each network for 2000 time steps, which generated dynamically persistent food webs with 30 species that are remarkably similar to empirical food webs 50 . Our third step uniformly randomly chose 30 species from a set of species created in the first step and sequentially introduced one from this set every 200 time steps into each of 100 persistent webs from each of the three levels of connectance. The simulations were then stopped 2000 time steps after the last introduction. Extinctions were measured as the number of species that went below our threshold of 10 −30 during the simulations. Figure 2 We use the stochastic “niche model” 43 with inputs of species richness ( S ) and directed connectance ( C = # of links/ S 2 ) 45 to generate food webs. Each i th of S species shown as a triangle is assigned a randomly drawn ‘niche value’ ( n i ) from the interval (1,0). Each i th species is then constrained to consume all prey species within a range of beta-distributed values ( r i ) whose mean is C and whose randomly chosen center ( c i ) is less than the consumer’s niche value. If an introduced node shown as a black triangle differs greatly from other nodes (e.g., gray triangle) in the network, we assume this node should be considered as ( a ) an ‘invasion’ event whereas ( b ) if an introduced node is largely similar to another node already in the network the new node represents a ‘speciation’ event. Given the centrality of niche overlap to eco-evolutionary theories of community assembly 41 , 51 , 52 , we calculated the niche similarity of the introduced species upon introduction to its most trophically similar species already in the web 45 . This “maximum similarity” of an introduced species is the number of predators and prey shared in common divided by the pair’s total number of predators and prey 45 , 53 , 54 . We assume introducing trophically dissimilar species represents longer range dispersal events of species arriving to the community from a regional species pool while introducing trophically similar species represent sympatric or parapatric speciation events hereafter referred to as ‘speciation’ 15 . Overall characterization of each introduction sequence as either invasion or speciation dominated was based on the average maximum similarity of all 30 introduction events for each web, hereafter called “mean maximum similarity.” See Methods for additional details.",
"discussion": "Discussion Overall, a number of differences were observed in assembly dynamics and final food web structure along the mean maximum similarity (invasion to speciation) gradient. In particular, webs dominated by speciation events had a higher proportion of successful introductions and fewer extinctions, which resulted in webs that tended to have more species, higher connectance and higher trophic levels. Compared to the low generality and vulnerability of invading species, species introduced by speciation had higher generality and vulnerability which suggests that speciation introduces more interactions than invasions. Despite the possibility for increased interaction density to increase differences among species, such density creates more trophic overlap and helps explain the high trophic similarity of speciated species of speciation dominated webs. The structure of speciation-dominated webs more closely resembled those of a set of empirical food webs. This suggests that evolutionary dynamics may play a larger role in the structure and function of food webs than previously thought but appear opposite to expectations based on competitive exclusion 15 , 40 , 41 , 58 . Such expectations include niche partitioning where competition between species that share niches cause extinctions while species that avoided such overlap by partitioning niches would allow more species to persist 15 , 40 , 41 , 58 . However, these expectations may be due to incorrect interpretation of Lotka-Volterra models that simplify indirect effects between consumers sharing resources by translating them as direct interactions between consumers 59 . We avoid such problems here by “working with models that better encapsulate the mechanisms of interactions among multiple species” as suggested by others 59 . Absent potentially overstated difficulties of sharing niches 41 , 42 , our findings may result from niches that enable species to persist being already occupied in our native webs which restricts successful introductions to be those that join residents of already occupied niches. Conversely, introduced species with different niches than residents are more likely to occupy niches that are dynamically unable to sustain species. Others have suggested that competition can cause convergent evolution to lead to sympatric speciation 40 for which there is substantial field evidence 40 , 58 . Another possibility is that the relatively small non- or less overlapping parts of similar species’ niches provide sufficient resources for their sustenance as in pollination networks 60 . High generality could help maintain this sustenance by enabling species to shift their feeding towards less shared resource species whose identity may vary during dramatic changes that may accompany repeated introductions. This key role of small parts of generalists’ niches with a few strong links and many weak links is consistent with the stabilizing effect of low mean interaction strength of species with many interactions 61 . While future studies need to better explore these and other potential explanations, our findings add to other findings indicating that sharing much of one’s niche with other species, as one may expect in cases of sympatric and parapatric speciation, does not appear to strongly prevent species from coexisting and dynamically persisting 15 , 40 , 58 , 62 . Invasions in our simulations dramatically affect the food web in several ways similar to invasions in nature by causing secondary extinctions that severely reduce the diversity of the native community 63 , 64 and by greatly simplifying and reducing the number of trophic levels in the community’s overall structure. A prominent example is the invasion by Burmese pythons of the Florida Everglades (USA) which substantially altered the abundance distribution of its prey 65 . Another is the invasion of Australia by cane toads whose toxicity reduced the abundances and reorganised the communities of the toads’ predators 66 . These observations anecdotally suggest that invasions by species substantially different than native species can greatly alter the local network structure and diversity in nature qualitatively similar to those in the present study. Beyond such effects on communities, invasions often fail by not persisting within the system as seen where invading species remain at low populations for extended periods 67 or need several attempts to become an effective invader 68 . Our results combined with the frequent implicit assumption that phylogenetic distance is strongly related to ecological similarity in general 69 and to network similarity in particular 70 , 71 suggests that invasion is easier for species that are phylogenetically close to one of the local species in agreement with recent data 62 , 72 used in invasive species prevention schemes 73 . Our results showing that networks assembled through speciation events are more similar to empirical networks agrees with other community evolution models that showed that evolutionary dynamics allow realistic network structures to emerge 30 , 31 , 34 , 74 , 75 . However, our evolutionary rules differ from these former models by relying on interaction similarity of phenotypes that emerge from evolutionary and ecological processes such as speciation and long range dispersal rather than relying on more explicitly modeled dispersal 16 or evolutionary dynamics of traits such as body size 31 , 75 , foraging traits 34 or competition based on interaction distributions 74 . Still, as in our study, these evolutionary events introduce new species whose niches substantially overlap with established species but whose traits such as body size, predators, and prey may slightly differ. While the rules generating new species differ, the basic co-evolutionary structure of these studies share the fundamental components of introduced trait variation and selection based on interaction with all species in the community. The bioenergetic basis of these studies ensure that such variation is retained over multiple generations and, as such, is effectively inherited. Disparate approaches to studying these fundamental components of evolutionary dynamics all lead to at least somewhat empirically realistic structures. This consistency suggests that a key aspect this process, i.e., a large degree of niche similarity between new and native species or morphotypes, is important to assembling realistic networks. When the network is assembled by speciation events, community robustness increases as evidenced by fewer secondary extinctions 76 which leads to more complex trophic structures. While this stabilizing effect of evolution may not be expected generally in diverse communities, it appears more likely in trophic networks 77 . Evolution in food webs leading to stable and complex structures has been observed in many different models, relying on very different sets of rules 30 , 31 , 36 , 78 . Introducing species into bipartite networks of plants and pollinators found more conventionally expected results where increased niche overlap among pollinators led to less persistence of native species 79 . However, this only occurred when there was an extraordinary difference in the invader’s niche beyond niche overlap; the cost-free ability to feed twice the rate of native pollinators. This ability caused the invaders to extirpate all natives whose niches were a subset of the invader’s niche and greatly reduce the abundance of other natives whose niches only partly overlapped the invader’s niche who survived by shifting their feeding to plants not consumed by the invader. Higher trophic levels (e.g., carnivores) may reduce such dramatic effects by preventing competition within lower levels from extirpating species 80 , 81 and weakening the strong interactions that destabilize complex food webs 77 , 82 , 83 . Our analysis advances this latter idea by suggesting that such balance more specifically concerns maintaining high levels of vulnerability and generality between new species and their progenitors. Our finding regarding “speciation” and “invasion” networks may be tested against natural systems greatly contrasting in their openness to migration. For very closed networks (e.g., very isolated islands with no humans that introduce invasive species), species invasion may be quite rare, so that the network may mostly represent the effects of species local adaptation or coevolution. Our “speciation” results may best match such closed networks in e.g., showing largely stable and complex structures. On the other hand, we expect very open networks (e.g., a continental ecosystem at the crossroad of many migratory paths or close to many human populations) may have much more frequent invasions causing extinction cascades as well as having a lower complexity with evolution playing a secondary role compared to invasions. We emphasize here our qualitive results largely because they are relatively insensitive to the uncertain boundary between similarity extremes that may distinguish invasion from speciation (Fig. 4 ). However, we also note that feedbacks between these two extremes may strongly influence evolution by changing the community structure and thereby altering how coevolution propagates through the network 84 . New predators may alter defense strategies among native prey as has been seen where mussels evolved thicker shells in response to the invasion by the Asian crab 85 . Similarly, predators may evolve in response to an invasive prey as predators have to invasive cane toads 86 . Our use of average similarity of 30 different invaders and the spectrum from invasion- to speciation-dominated webs helps inform such interactions between invasion and speciation by suggesting how webs generated by more interactions between these two types of species introductions are intermediate between webs mostly generated by one type of introduction. This interplay of invasion and evolutionary processes may help predict future states of ecological networks e.g., whether communities become closed or otherwise resistant to future invasions. Simulated invasions often do not cause such community closure both due to cyclic behaviours 87 and because drawing rates at random allow for infinite combinations. On the other hand, evolutionary dynamics may be trapped in local maxima of fitness which prevents introduced species or morphotypes from differing greatly from resident populations 40 . This greatly restricts the sampling of the possible parameter space by evolutionary processes. For these reasons, it has been suggested that allowing for a mix of invasions and evolution may be key to understanding community closure 22 . Our approach may be easily adapted to tackle this type of question by focusing how assembly changes over time rather than more simply comparing average outcomes between beginning and end states. More broadly, our results add to other findings suggesting that, whether assembled by ecological or evolutionary mechanisms, surprisingly large numbers of species are able to coexist despite large amounts of niche overlap 15 , 40 , 58 , in our case both in terms of resource and consumer species, the latter of which has received little previous attention. For example, Morlon et al . 52 found that species in a region’s 50 lake communities share prey much more often than expected if community composition resulted from randomly sampling all species from all 50 lakes within the region. Such findings taken together suggest that there are strong ecological and evolutionary mechanisms forcing species to fit within a relatively restricted architecture of trophic niches as described by theory such as that formalized by the trophic niche model (Fig. 2 ), which has much higher niche overlap than more randomly structure niche architectures 43 , 88 – 91 . Further research into these and other mechanisms that increase interspecific functional similarity and its consequences may greatly elucidate processes and patterns within complex natural ecosystems."
} | 6,356 |
38014336 | PMC10680560 | pmc | 8,268 | {
"abstract": "Microbial metabolism sustains life on Earth. Sequencing surveys of communities in hosts, oceans, and soils have revealed ubiquitous patterns linking the microbes present, the genes they possess, and local environmental conditions. One prominent explanation for these patterns is environmental filtering: local conditions select strains with particular traits. However, filtering assumes ecological interactions do not influence patterns, despite the fact that interactions can and do play an important role in structuring communities. Here, we demonstrate the insufficiency of the environmental filtering hypothesis for explaining global patterns in topsoil microbiomes. Using denitrification as a model system, we find that the abundances of two characteristic genotypes trade-off with pH; nar gene abundances increase while nap abundances decrease with declining pH. Contradicting the filtering hypothesis, we show that strains possessing the Nar genotype are enriched in low pH conditions but fail to grow alone. Instead, the dominance of Nar genotypes at low pH arises from an ecological interaction with Nap genotypes that alleviates nitrite toxicity. Our study provides a roadmap for dissecting how global associations between environmental variables and gene abundances arise from environmentally modulated community interactions.",
"introduction": "Introduction The metabolic activity of microbiomes in natural environments from soils to oceans drives the global cycling of carbon, nitrogen, and other elements essential to life on Earth (?). This activity arises from the collective action of metabolic pathways carried by diverse, interacting microbes in communities. Despite this complexity, global sequencing surveys of natural communities have revealed ubiquitous patterns relating the abundance of the genes that make up these pathways to local environmental variables. For example, bacterial metabolic capabilities vary with nutrient levels in marine systems ( 1 ), and gene content changes with soil pH ( 2 – 4 ) and host diet ( 5 – 7 ). Therefore, variation in the local environment modulates microbiome gene content and metabolic activity upon which the biosphere relies ( 3 , 8 , 9 ). Understanding the origins of environmentally-mediated variation in microbiome gene content is a necessity to understand how human activity impacts global nutrient cycles. How do environmentally dependent patterns in gene content reproducibly emerge in microbial communities across the planet? One common hypothesis is environmental filtering: strains exhibiting traits that are adapted to their local environment are able to colonize that environment even in the absence of any other strains being present ( 10 ). In this case, the dominance of a strain in an environmental condition is not dependent on biotic interactions, and depends only on the traits of the environmentally selected strains. In some cases, associations between environmental variables and certain genotypes make intuitive sense with the environmental filtering hypothesis. For example, the relative abundance of acid-tolerant Acidobacteria in soil increases as the pH declines ( 9 , 11 , 12 ). Similarly, the prevalence of anaerobic nitrate respiration and anoxygenic photoautotrophy increase with decreasing concentrations of oxygen in the global ocean microbiome ( 1 ). In these cases, we presume that specific metabolic strategies are enriched in niches where those traits facilitate a competitive advantage. However, we also know that microbial communities are complex systems with strong interactions between constituents ( 13 – 16 ), resulting in feedback, counter-intuitive inhibitory effects, evolutionary consequences, and predation, all of which conspire to determine abundances ( 17 – 21 ). Thus environmental factors can modulate interactions and drive impacts on community diversity, gene content, and metabolic activity. Despite this, it remains unclear what role these interactions play in giving rise to ubiquitous large-scale patterns observed in communities. From this vantage, a problem arises: how can simple patterns in environmentally mediated gene content emerge given the apparent complexity of community interactions? One possibility consistent with the environmental filtering hypothesis is that interactions are weak, and changes in gene content reflect an adaptation of individual genotypes to local environmental conditions. A second possibility is that the patterns in gene content emerge from ecological interactions whose strength and specificity are modulated by local environmental variables. In this case, the emergence of reproducible patterns requires that the interactions exhibit regularity across different locations with similar environmental conditions. However, given the complexity of interactions in microbiomes, imagining how this regularity could manifest is a challenge. Thus, two points are clear. First, patterns relating environmental variation to community composition are ubiquitous in many microbiomes. Second, it is clear from laboratory studies that interactions are important in structuring microbial communities and these interactions can depend on the environmental context. Less direct evidence also strongly suggests that interactions are important in more complex communities in the wild ( 22 ). What remains unclear is how interactions in very complex communities can give rise to the ubiquitous regular patterns we observe in natural consortia on a large scale. Here, we develop and demonstrate a new approach to answering this question. Here, we present evidence that global patterns in the gene content of microbial communities emerge from environmentally dependent interactions between members of the microbiome. We use bacterial denitrification as a model process to understand the mechanisms underlying global patterns in gene content. We quantify patterns in the abundances of genes associated with denitrification in the global topsoil microbiome ( 3 ), finding that soil pH strongly associates with variation in denitrification reductase gene content in soils. Using enrichments and quantitative phenotyping of isolates, we demonstrate the insufficiency of the environmental filtering hypothesis to explain these global patterns. Specifically, we show that strains possessing the denitrification gene Nar (Nar + ), encoding a cytoplasmic nitrate reductase, are globally enriched in low pH soils, and that strains possessing this gene dominate low pH enrichment cultures as well. Counterintuitively, Nar + strains generically accumulate nitrite, which is toxic in acidic conditions, giving rise to reduced growth when cultured alone. We show that a strong interaction between Nar + strains and strains possessing the alternative nitrate reductase Nap reduces this toxicity, allowing Nar + to dominate at low pH. We argue that these interactions yield reproducible genomic patterns on a global scale as a result of the conserved physiological properties of the genotypes involved ( 23 ) and the interactions these physiological traits support. We support this claim by showing that denitrifying populations in soils are dominated by strains that are taxonomically similar to the enriched isolates used in our study. Further, a soil microcosm experiment reveals that just a handful of taxa out of the thousands present in soils respond to the presence of nitrate, suggesting that interactions between a small number of strains may be relevant even in the ecological context of a soil microbiome. Our study provides a unified way of understanding ubiquitous patterns in microbiomes as emerging from environmentally-modulated interactions between members of the community.",
"discussion": "Discussion Surveys of microbiomes in contexts from hosts to oceans have elucidated environmentally-mediated patterns in the composition of these complex communities. Understanding these patterns, and ultimately their implications for the function and persistence of microbial consortia is a long-standing problem in ecology. We have presented a data-driven and empirically validated approach to understanding how changing environmental conditions drive changing community composition at the level of gene content. Our results reveal that environmentally mediated patterns on a global scale can emerge from interactions between members of a community. We arrived at this conclusion by first dissecting which environmental factors correlate with variation in gene abundances in the global topsoil microbiome, finding pH to be strongly correlated with the composition of the denitrification pathway. Our statistical approach motivated an experiment varying pH and assaying the structure of resulting denitrifying communities in the laboratory. Our enrichments recapitulated the observed pattern, allowing us to dissect interactions quantitatively. We showed that the dominance of the Nar + genotype in acidic conditions relies on the presence of a Nap+ genotype to reduce pH-dependent nitrite toxicity. The central finding of our study is that patterns in the relative abundances of genes on a global scale can emerge from ecological interactions between members of the community. The finding points to the key idea that patterns can emerge from interactions in communities as a whole, and not necessarily from specific traits conferring an advantage to individual taxa in certain environmental conditions ( 10 ). While filtering clearly plays a role in defining these patterns, for example at neutral pH in our study, interactions are likely essential to understanding the existence of patterns on a global scale. Limitations and extensions Our analysis of gene content in the global topsoil microbiome revealed a pattern of gene abundances across a gradient of pH from very acidic soils (pH ~4) to alkaline (pH ~8). Over this range, we observed continuous variation in the abundances of the narG and napA genes. However, our monoculture and coculture experiments were restricted to pH 6.0 and 7.3. As a result, it remains unclear whether or not other interactions, or even filtering, might play a role in defining the full extent of the global pattern we observe. For example, it is common to find obligate nitrate reducers possessing narG that do not reduce nitrite ( 23 ). Additionally, we note that in enrichments at pH < 5.5, we do not find taxa of the Nap genotype, only Nar ( Supplementary Information ). Therefore, it may be that additional ecological and physiological processes are important in more acidic conditions. Interactions involving such genotypes and environments could be essential for fully understanding the global pattern we observed. Interrogating interactions in more acidic conditions could be useful in uncovering additional processes. Next, while we provide strong evidence of the generality of the observed interaction, we do not show direct evidence of the toxicity-alleviation interaction between PD Nar + and RH Nap + in other strains. Thus, this interaction may not be present between all pairs of Nar + and Nap + genotypes at every pH. We show via simulation, however, that such interactions are likely to arise from ecological interactions that give rise to complementary metabolic phenotypes ( Figs. 5 \n S13 ). Finally, our statistical analysis of variation in pathway magnitude and composition focused on variations in the composition of the denitrification pathway across the globe. As a result, our study shows that interactions are important for determining which genotypes participate in the process of denitrification. However, the magnitude of the pathway, i.e., approximately the fraction of reads in the metagenome mapping to denitrification enzymes, decreased with increasing C/N ratio in soils ( ρ < 0, Fig. S1 ). Denitrification is known to compete with dissimilatory nitrate reduction to ammonia (DNRA) in anaerobic environments ( 31 ), with low C/N favoring denitrification over DNRA for reasons of redox balance ( 51 ). Our findings are consistent since rising C/N inhibits denitrification globally. In this sense, it is likely that an environmental filtering framework does indeed explain which process, denitrification versus DNRA, dominates in a given environmental context. It will be an important avenue for future work to use the methods developed here to generate concrete hypotheses for the role of filtering and interactions in pattern formation. Testing these will be important for understanding the fate of nutrients in the environment, as DNRA results in the production of bioavailable ammonia, and denitrification drives nitrogen loss to N 2 . Trait conservation and trade-offs A key finding of our study is the idea that consistent patterns in gene content on a global scale can reflect conserved genotype-phenotype relationships across diverse bacteria. This result is in agreement with our previous work showing that denitrification genotypes confer similar phenotypes across different taxa ( 23 ). The present study suggests that this conservation of traits can lead to the conservation of interactions between genotypes and thus patterns observable via metagenomic sequencing over many environments. In our system, the interaction between Nar + and Nap + genotypes arises due to specialization by these two genotypes into the first and second steps of the denitrification cascade. This specialization suggests the possibility that there is a physiological trade-off between nitrate and nitrite reduction rates. Previous studies have implicated such a trade-off and suggested that it might emerge from competition for intracellular resources ( 33 , 52 ). We investigated this further using simulations that account for the trade-off between nitrate and nitrite reduction rates and the toxicity of nitrite, which preferentially inhibits nitrate specialists ( Supplementary Information ). These simulations identified two phases of community assembly: (I) with high toxicity (low pH) and concave trade-off, we observed a regime where two generalist strains that perform both nitrate and nitrite reduction coexist ( Fig. S12B ), and (II) when toxicity is low (neutral pH) and the front is concave, a single generalist (reduces both nitrate and nitrite) dominates ( Fig. S12B ). Thus, we found that the observed experimental phenomena at both pH 6.0 (regime I) and pH 7.3 (regime II) emerge generically from the imposition of empirically observed physiological constraints. Together, these findings suggest the idea that learning conserved physiological constraints on traits will be an important step in understanding ecological processes and patterns on a global scale. The role of enzyme properties in denitrification traits In this work we have used the genes narG and napA as genetic markers of conserved, interacting, phenotypes where strains with narG are nitrate specialists and strains with napA are nitrite specialists. Here we discuss how the enzymes that make up the pathways might mechanistically confer the traits we observe. Consistent with our results here we previously showed that nitrate reducers possessing narG instead of napA exhibit faster nitrate reduction ( 44 ), a finding supported by in vitro studies of these enzymes ( 53 – 57 ). In addition, the Nar nitrate reductase is protected from extracellular pH in the cytoplasm, while the periplasmic Nap reductases are not ( 24 ), possibly explaining faster nitrate reduction at low pH for Nar + strains. A key finding of our study is the alleviation of nitrite toxicity, and thus the nitrite reductases encoded by the genes nirS and nirK are also likely important. Statistically, we observe nirS and narG have the same relationship with pH as do napA and nirK . This observation aligns with the fact that nitrite toxicity alleviation is accomplished by a denitrifier possessing nirK ( Fig. 2 ), suggesting that nitrite specialists possess the NirK nitrite reductase and not the NirS enzyme ( Figs. 1H , 2D ; additionally, every Nar + strain in Fig. 6B also contains NirS except PDM21, and every Nap + strain also contains NirK ). This is also consistent with previous work, which shows NirK enzymes have higher activity than NirS ( 24 , 58 – 71 ), and the presence of nirK rather than nirS predicts a faster nitrite reduction traits ( 44 ). Thus, qualitatively the known properties of enzymes in the pathway accord with the nitrate and nitrite specialization we observe in our isolates. In addition, studies examining the transcriptional and translational dynamics of reductases of denitrifiers to changing pH show that nirS possessing strains have decreased enzyme levels at low pH ( 72 , 73 ) while NirK possessing strains do not ( 74 – 76 ). A search of the BRENDA ( 77 ) database for experiments on purified NirS and NirK enzymes did not show significant differences in pH optima ( 78 – 84 ). Thus, the nitrite reductases likely have differential basal activity and pH alters their transcriptional or translational dynamics. Together, the properties of the denitrification pathway qualitatively agree with the traits we observe for these genotypes. Learning from environmental variation in the wild Our results demonstrate the power of statistical analysis to motivate in vitro experiments that connect strain-level phenomena to macroecological patterns. Our approach used a statistical decomposition of variation in natural communities to reveal patterns in gene abundances. This approach is in contrast to a “mechanism-first” approach recently undertaken by Abreu et al . ( 85 ) that identified the phenomenon of slow growth being favored at warmer temperatures in a simplified laboratory community, before searching for this pattern in the wild. While both approaches are powerful for understanding patterns in nature, a statistical approach allows us to discover patterns and processes without first postulating the environmental variables responsible or the details of the interaction in the laboratory. In this way, our approach allows us to uncover the key environmental drivers of variation in the wild before dissecting the role of interactions in the community. The stress gradient hypothesis The stress gradient hypothesis (SGH) provides a framework for understanding how the sign of ecological interactions depends on environmental conditions ( 86 ). The SGH suggests that interactions should skew positive (facilitation) in the presence of an environmental stressor, and negative (competitive) in the absence of stress. Previous attempts to demonstrate the SGH in the plant and microbial world highlight challenges associated with the ambiguity of the hypothesis ( 87 ). Namely, not all stressors appear to select for facilitative interactions ( 88 ), and in situations where the SGH has been observed ( 89 ), it is unclear whether any conserved underlying mechanisms are responsible. These challenges mean that we do not yet know whether a systematic description of the impact of stress on community interactions is plausible. Our results represent a clear demonstration of the SGH, with nitrite toxicity at low pH serving as the stressor and the alleviation of toxicity by Nap + genotypes serving as the facilitative interaction. More importantly, our approach demonstrates a systematic methodology for elucidating the SGH in communities. Specifically, we have shown that conserved genotype-phenotype relationships can drive patterns of facilitative interactions, leaving a signature of the SGH in covariation between environmental variables and metagenomic data. Further analyses of these datasets could yield insights into the importance of the SGH for other microbial processes. Such analyses could enable a more systematic interrogation of the relevance of the SGH in structuring natural communities. The environmental filtering hypothesis The concept of environmental filtering has a long history, perhaps beginning with the foundational work of von Humbolt and Bonpland ( 90 ), and the belief that traits alone can determine community composition has remained persistent when interpreting data in the modern era. Surveys in marine microbiomes show that functional traits correlate with environmental variables ( 91 ). Similarly, studies of the gut microbiome have interpreted reproducible changes in the abundances of specific taxa in response to diet in terms of environmental filtering ( 92 ). In these very complex ecological contexts, it is challenging to definitively determine when interactions matter and when they do not. Experimental and analytical methods for dissecting interactions in these contexts will be important going forward. More broadly, we hope this work helps establish a link between two conceptual frameworks in community ecology. First, decades of work have focused on the structure of interactions between taxa and the resulting abundance dynamics ( 93 ). In this picture, traits are often assigned randomly, and community function or ecosystem services play a secondary role. Second, quantitative physiology and an appreciation of the importance of metabolic traits in community assembly ( 94 ) have revealed that traits are not random, but rather that strong conservation yields regular structure in the physiology of microbes, yielding robust quantitative principles (e.g., Ref. 95 ). Our study proposes a path forward that bridges these two conceptual frameworks, revealing how conserved traits can give rise to recurrent patterns of interactions in wild communities with implications for ecosystem functions. We hope that understanding the origins of these patterns will drive deeper insights into the assembly and function of complex natural microbiomes."
} | 5,427 |
37770950 | PMC10540321 | pmc | 8,269 | {
"abstract": "Background Plant-beneficial bacterial inoculants are of great interest in agriculture as they have the potential to promote plant growth and health. However, the inoculation of the rhizosphere microbiome often results in a suboptimal or transient colonization, which is due to a variety of factors that influence the fate of the inoculant. To better understand the fate of plant-beneficial inoculants in complex rhizosphere microbiomes, composed by hundreds of genotypes and multifactorial selection mechanisms, controlled studies with high-complexity soil microbiomes are needed. Results We analysed early compositional changes in a taxa-rich natural soil bacterial community under both exponential nutrient-rich and stationary nutrient-limited growth conditions (i.e. growing and stable communities, respectively) following inoculation with the plant-beneficial bacterium Pseudomonas protegens in a bulk soil or a wheat rhizosphere environment. P. protegens successfully established under all conditions tested and was more abundant in the rhizosphere of the stable community. Nutrient availability was a major factor driving microbiome composition and structure as well as the underlying assembly processes. While access to nutrients resulted in communities assembled mainly by homogeneous selection, stochastic processes dominated under the nutrient-deprived conditions. We also observed an increased rhizosphere selection effect under nutrient-limited conditions, resulting in a higher number of amplicon sequence variants (ASVs) whose relative abundance was enriched. The inoculation with P. protegens produced discrete changes, some of which involved other Pseudomonas . Direct competition between Pseudomonas strains partially failed to replicate the observed differences in the microbiome and pointed to a more complex interaction network. Conclusions The results of this study show that nutrient availability is a major driving force of microbiome composition, structure and diversity in both the bulk soil and the wheat rhizosphere and determines the assembly processes that govern early microbiome development. The successful establishment of the inoculant was facilitated by the wheat rhizosphere and produced discrete changes among other members of the microbiome. Direct competition between Pseudomonas strains only partially explained the microbiome changes, indicating that indirect interactions or spatial distribution in the rhizosphere or soil interface may be crucial for the survival of certain bacteria. \n Video Abstract Supplementary Information The online version contains supplementary material available at 10.1186/s40168-023-01660-5.",
"conclusion": "Conclusions The proliferation of plant-beneficial bacterial inoculants in rhizosphere microbiomes remains poorly understood due to the highly complex nature of the rhizosphere and soil environments, and fully reductionist approaches that focus on bacteria-bacteria interactions or synthetic communities only provide limited insight. Nonetheless, the investigation of the interplay of an inoculant with a species-rich microbiome and a key environmental factor such as nutrient availability as undertaken in the present study might broaden our understanding of inoculant establishment strategies. We studied the effect of the plant-beneficial strain Pseudomonas protegens CHA0 when introduced into a soil natural, species-rich bacterial community (NatCom) [ 46 ] established in bulk soil and upon its assembly in the rhizosphere of wheat. We explored two community states, based on nutrient availability, a growing NatCom produced by community dilution and addition of new nutrients and a stable, nutrient limited NatCom in a stationary growth state. Our results are in line with the concept that access to nutrients is a major determinant of microbiome composition and structure, diversity, and assembly processes [ 96 , 97 ]. The P. protegens inoculant was able to establish at a relatively small abundance within the microbiomes in all conditions tested, but this peaked in the wheat rhizosphere under stable conditions, which also reduced the diversity of the rhizosphere microbiome. This supports the concept that in a nutrient-limited environment, the plant rhizosphere can provide a niche available to specific taxa, due to root-exuded nutrients and signalling compounds [ 34 , 36 ], thus supporting proliferation of adapted inoculants. The changes in the microbiomes over time showed that under growing conditions, the environment and the inoculation pattern lead to diverging trajectories. The lack of convergence along sampling times evidenced an early microbiome assembly process [ 23 , 81 ]. Whether microbiome convergence will be attained or if the inoculant will also persist for a longer period remains to be studied. In addition, the changes observed in the growing condition were mainly explained by homogeneous selection as the main deterministic assembly process, independent of environment (bulk soil or wheat rhizosphere) or inoculation pattern, emphasizing nutrient availability as the dominant force driving communities to the observed compositions. Conversely, under stable, nutrient-limited conditions, we found microbiome divergence only in the wheat rhizosphere while observing no differences in the inoculation regime and a reduced speed of microbiome drift in the stable bulk soil. Furthermore, the lack of a common driving force in this condition fits a more complex scenario in which various processes may explain the observed microbiome composition. Notably, the rhizosphere effect might be amplified, being the only source of available nutrients in the stable community, where other processes, such as specialized scavenging mechanisms, might be more prevalent [ 82 ]. Indeed, the rhizosphere environment produced changes in certain ASVs, mainly from the Actinobacteria, Bacteroidota, Firmicutes and Proteobacteria phyla, which were mostly enriched in stable conditions. Association inference of taxa showed that communities were dominated by positive interactions while modularity remained constant across samples. Nonetheless, the presence of different keystone taxa depending on the inoculation pattern suggests altered community networks. In fact, the introduction of P. protegens CHA0 changed a discrete number of ASVs mostly affecting the rhizosphere of both growing and stable conditions. This implies that the introduction of the inoculant does not radically alter the composition of the rhizosphere or bulk soil microbiome, as also observed by other authors using different Pseudomonas inoculants [ 26 , 27 ]. The observed changes involve nevertheless a reduced prevalence of specific Pseudomonas ASVs, commonly associated with plant hosts [ 92 , 93 ], making them likely competitors in the rhizosphere environment. Direct competition of the inoculant with two Pseudomonas strains isolated from the NatCom revealed that competition in the bulk soil or the rhizosphere environment only partially explains the observations at the microbiome level. This suggests that other factors may be at play, including a more complex network of interactions across the microbiome members or a different spatial distribution across roots or the soil interface [ 19 , 23 ], which may play an important role in defining the plant-rhizosphere microbiome. The approach followed in the present study, using a species-rich natural soil bacterial community in a structurally complex environment, allowed us to demonstrate that the niche created by the wheat rhizosphere allows the proliferation of P. protegens CHA0, which also competes efficiently with closely related bacteria.",
"discussion": "Results and discussion Pseudomonas protegens proliferates best in stable resident communities under accessibility to the rhizosphere niche We assessed the proliferation performance of the plant-beneficial inoculant Pseudomonas protegens CHA0 in response to plant roots (wheat) when exposed to a natural soil bacterial community (NatCom) in a growing versus a stable state. Simultaneously, we analysed the relative abundance of the different bacterial classes in the growing and stable NatComs. Overall, samples of the growing NatCom were dominated by Gammaproteobacteria (Fig. 1 A; Supplementary Table S 2 ), followed by Alphaproteobacteria, Bacteroidia, Actinobacteria and Bacilli. Conversely, the stable NatCom samples were dominated by Bacilli, followed by Alphaproteobacteria, Gammaproteobacteria, Bacteroidia and Planctomycetes. These bacterial classes are commonly found in soils worldwide [ 75 , 76 ] and were also previously detected in the original soil NatCom used in this work [ 46 ]. However, the unusual higher abundance of Gammaproteobacteria in the growing community compared to other soils [ 75 , 76 ] might indicate specific changes resulting from the initial high availability of nutrients (i.e. growing conditions), causing a dominance of fast-growing bacteria. These differences are better highlighted in a comparison of normalized relative class abundances between samples from the growing against the stable conditions (Fig. 1 B, Supplementary Fig. S 1 ). Notably, Acidimicrobiia, Bacilli, Planctomycetes, Polyangia and Thermoleophilia were significantly more abundant in the stable condition, while Actinobacteria and Gammaproteobacteria were more abundant in the growing community state. No differences in the relative abundance of Alphaproteobacteria and Bdellovibrionia were observed for most of the comparisons. Interestingly, bacterial classes whose relative abundances increased in stable, nutrient-limited conditions showed a decrease in the wheat rhizosphere (Fig. 1 B, Supplementary Fig. S 1 ). Although root exudates are rich in organic compounds [ 34 ], they also contain signalling molecules that could inhibit the growth of specific taxa, including Bacilli [ 16 ]. The opposite effect was observed for Gammaproteobacteria, whose abundances in the wheat rhizosphere increased under stable conditions compared to the corresponding bulk soil sample (Fig. 1 B). In contrast, under growing conditions, most bacterial classes did not significantly differ in abundance in the wheat rhizosphere compared to the bulk soil. The exceptions were Acidimicrobiia and Actinobacteria, whose abundances increased (Fig. 1 B). This might be due to a specific exploitation of root exudates. The differences observed between both community states (i.e. growing and stable) can likely be attributed to the initial access to nutrient niches under the growing condition, which allows the rapid growth of part of the population, resulting in the observed class differences. Fig. 1 Composition and diversity of NatComs exposed to wheat roots and the inoculation of Pseudomonas protegens . A Relative abundance per replicate of the top 100 taxa at the bacterial class level in the growing or stable soil NatCom and of P. protegens CHA0 ASV (green bars, left) across sampling times (dpi, days post inoculation). B Differences in the CSS-normalized relative abundance of the three main classes Alphaproteobacteria, Bacilli and Gammaproteobacteria (see Supplementary Fig. 1 for all classes) or C P. protegens CHA0 ASV across samples. Samples taken at different sampling times are merged. For Gammaproteobacteria, P. protegens ASV was removed to avoid an artificial inflation of the class relative abundance. Significance based on Kruskal-Wallis rank-sum test with LSD post hoc analysis and p -value corrected by fdr. Different letters indicate significant differences between groups at p -value < 0.05. D Comparison of Shannon diversity in samples from different community states (growing or stable), environments (bulk soil or wheat rhizosphere (Riz.)) or inoculated with P. protegens CHA0 (represented by green dots) or not inoculated (empty dots). Significance based on Wilcoxon rank-sum test. Not significant (n.s.): p -value > 0.05. E Spearman correlation between the Shannon diversity index and sampling time in growing (green) or stable (yellow) community states. Correlation coefficient ( R ) and p -value ( p ) are indicated with the colour according to the growing state. Curves represent the general additive model (GAM) fit (average, line) and the 95% confidence interval (shadow) Inoculation with P. protegens CHA0 did not significantly affect the relative abundances of the dominant bacterial classes. Notable exceptions were lower abundances of Actinobacteria, Alphaproteobacteria and Planctomycetes classes in the P. protegens -inoculated wheat rhizosphere under stable conditions (Fig. 1 B, Supplementary Fig. S 1 ). After removing P. protegens ASV counts from the Gammaproteobacteria calculations to avoid an artificial inflation, an increase in the abundance of this class was observed in the uninoculated wheat rhizosphere of stable conditions (Fig. 1 B), suggesting a rhizosphere-specific effect. The fact that the largest changes in response to the plant-beneficial inoculant occurred within the wheat rhizosphere under stable, nutrient-limited conditions is probably the result of the selective effect of secreted root exudates. These contain nutrient-rich compounds that create a specific new niche within the otherwise niche-limited bulk soil, which is accessible to specific root-targeting microbiota [ 11 , 48 ]. Indeed, the highest abundance of the ASV matching P. protegens CHA0 was detected in the rhizosphere of the stable NatCom (Fig. 1 AC, average relative abundance across sampling times of 14.11%), which was significantly higher than in the bulk soil condition of the stable community (average relative abundance of 2.01%) and compared to the growing conditions (average relative abundance of 1.79% in the rhizosphere or 1.63% in bulk soil). The inoculant was therefore able to efficiently reach the rhizosphere microbiome and persisted for at least 9 dpi (Fig. 1 C, Supplementary Table S 2 ). Pseudomonas protegens proliferation alters the rhizosphere diversity under nutrient-limiting conditions We next evaluated whether the diversity of the rhizosphere microbiome was influenced by the inoculant or by the nutrient availability. Growing NatComs exhibited significantly higher Shannon alpha diversity than stable ones (Fig. 1 D), likely as a result of an initial higher availability in nutrients. Within the growing NatCom, the wheat rhizosphere significantly increased all diversity indexes (Shannon diversity, observed ASVs and Faith’s phylogenetic diversity), while the stable NatCom only showed increased Shannon diversity (Supplementary Fig. S 2 , Supplementary Table S 3 ). This may point to a rhizosphere enrichment effect and can be attributed to the secretion of specific nutrients by the wheat roots [ 77 ]. Inoculation of P. protegens CHA0 did not result in significant differences in diversity, except in the rhizosphere of the stable condition, which showed a reduced Shannon diversity (Fig. 1 D). However, the number of observed ASVs and the phylogenetic diversity remained constant here (Supplementary Fig. S 2 ). This reduction in the Shannon diversity in the wheat rhizosphere of the stable NatCom is due to the population of P. protegens CHA0 (average relative abundance across timepoints of 14.11%, Fig. 1 A, C), which proliferated in the niche that otherwise other NatCom bacteria would colonize. There was mostly no significant correlation of sampling time with Shannon diversity (Spearman correlation, p -value > 0.05; Fig. 1 E), except for a positive correlation in the wheat rhizosphere under growing conditions ( R = 0.8, p -value = 0.0019). This may be explained by the contribution of root exudates in addition to the nutrients contained in the soil extract. Importantly, the inoculation with P. protegens CHA0 also produced a discrete increase in wheat growth compared to non-inoculated conditions (Supplementary Fig. S 3 ). Mean weight values of fresh and dry shoots in the growing or stable conditions were higher in the inoculated samples, as were shoot lengths and root weights. The beneficial effect on plant growth of P. protegens CHA0 is in line with previous research reporting the strain’s growth-enhancing effects on different plant species [ 32 , 78 ] and is related to its capacity to solubilize nutrients [ 79 ] and to synthesize phytohormones [ 80 ]. Community succession is influenced by nutrient availability, wheat roots and Pseudomonas protegens proliferation Community succession was evaluated based on Bray-Curtis dissimilarities. Samples from both community states showed different compositional succession over a period of 9 days (Fig. 2 AB) and were clearly different based on hierarchical clustering (Supplementary Fig. S 4 ). In the case of the growing NatCom, the inoculated Pseudomonas had an impact on both the bulk soil and wheat rhizosphere communities, leading to measurable diverging trajectories (PERMANOVA p -value = 0.0369). For the stable NatCom, all bulk soil samples clustered together, regardless of the sampling time or the inoculation pattern (Fig. 2 B, Supplementary Fig. S 4 ), whereas the corresponding wheat rhizosphere communities diverged, depending on inoculation with P. protegens CHA0. This shows that they are undergoing successional changes, coherent with the assembly of the root microbiome [ 81 ], the proliferation of P. protegens CHA0, or a combination of both. The finding that no differences in the trajectories of the bulk soil communities of stable conditions were detected in the absence or presence of the inoculant (Fig. 2 B, Supplementary Table S 4 ) is likely due to the inability of the community to grow given the nutrient-deprived environment. Interreplicate variability measured as the distance of each replicate to the centroid showed a moderate increase in variability with sampling time in the growing NatCom (Fig. 2 C), probably caused by a still-evolving community, while this was only observed in the wheat rhizosphere of the stable condition, which could indicate that part of the stable NatCom population can grow using the root exudates as a carbon source [ 34 , 36 ]. In addition, sampling time positively correlated with Bray-Curtis dissimilarities in growing NatCom samples, except in the inoculated wheat rhizosphere (Fig. 2 D), while in the stable NatCom, a significant correlation with time was only observed in the rhizosphere of wheat. Fig. 2 NatComs succession trajectories when exposed to wheat roots and the proliferation of Pseudomonas protegens . Nonmetric multidimensional scaling (NMDS) based on Bray-Curtis dissimilarities of the growing A or stable B NatCom samples, using a k = 2, and kernel density estimate of the replicate distribution per NMDS axis. Replicates (small dots) are connected to the centroid (big dot) by coloured lines according to samples and sampling times (dpi, days post inoculation). Numbers indicate sampling timepoint. C Inter-replicate consistency measured as the distance of each replicate to the centroid. Bars (±standard error) represent average values. Dots represent individual replicate values. D Spearman correlation between pairwise Bray-Curtis dissimilarities across samples and sampling time in growing (violet) or stable (blue) community states. Correlation coefficient ( R ) and p -value ( p ) are indicated. Curves represent the general additive model (GAM) fit (average, line) and the 95% confidence interval (shadow) These results indicate that nutrient availability must have been a major determinant for the different community trajectories between growing and stable NatComs. The successional changes that followed responded to the presence of wheat roots or the inoculation with P. protegens CHA0, except in the bulk conditions of the stable NatCom, where the slower rate of compositional changes across sampling times and no effect of the inoculant point to changes related to other phenomena such as competition for growth-limiting nutrients [ 82 ]. Assembly processes are governed by nutrient availability but do not impact overall association networks Processes driving the assembly of the communities were assessed based on entire-community null models. The community state alone (i.e. growing or stable) influenced the deviation from the null models (βNTI, Fig. 3 A), with growing state being the most deviated. While the growing NatCom was dominated by deterministic processes, mainly homogeneous selection representing the 90%, the assembly of the stable NatCom was driven by stochastic processes (63% of undominated processes and 35% of homogenizing dispersal, Supplementary Table S 5 ). Environment (bulk soil or wheat rhizosphere) was a significant factor in the model deviation (Fig. 3 B), but inoculation was not, except in the case of the stable community rhizosphere (Fig. 3 C). The attributed assembly processes were similar, but the fraction of undominated processes increased up to 74% in the stable community rhizosphere (Fig. 3 B). Inoculation of P. protegens CHA0 reduced the proportion of homogeneous selection in favour of undominated processes in the stable community, both for the bulk soil and the wheat rhizosphere (Fig. 3 C, Supplementary Table S 5 ). Fig. 3 Community assembly processes of NatComs and ecological association upon establishment of Pseudomonas protegens . Community assembly processes dominating the bacterial communities in response to the different community states, growing or stable A ; different environments, bulk soil or wheat rhizosphere B ; or the inoculation with P. protegens CHA0 C were calculated based on deviation from null community models. Differences in beta-nearest taxon index (βNTI), together with Raup-Crick-based Bray-Curtis served to determine the community assembly process that dominated the samples (bar plots). Per tested condition, all sampling times were merged. The threshold for the |βNTI| = 2 is highlighted as horizontal red lines. Differences in βNTI are based on Wilcoxon rank-sum test. Not significant (n.s.): p -value > 0.05. Ecological association inference is based on sparse inverse covariance estimation among the top 150 ASVs (dots, coloured according to the bacterial class assignation) from growing or stable community states, in bulk soil or the wheat rhizosphere in response to the inoculation with P. protegens CHA0 D or not inoculated E . Samples from the sampling times 5, 7 and 9 days were combined. Positive and negative interactions are coloured in red or green, respectively. Hub centrality scores of each ASV (dots) in the networks are based on the scaled Kleinberg’s hub centrality score. ASVs with a hub score ≥ 0.7 were considered keystone taxa and coloured according to the bacterial class assignation The dominance of homogeneous selection within the growing NatCom suggests that the addition of nutrients drives the community succession [ 63 , 83 ], which was expected as the replicability of the NatCom is based upon this concept [ 46 ]. However, in the absence of a dominating force (i.e. nutrients), the stable NatCom drifted apart mainly due to stochastic processes of turnover and the absence of net growth [ 84 ]. Surprisingly, the proportion of undominated stochastic processes in the wheat rhizosphere of stable conditions was even higher than in the bulk soil and could be related to the observed interreplicate variability (Fig. 3 B). This increased stochasticity is possibly linked to the early stages of rhizosphere microbiome formation, where root exudates serve as carbon sources for the bacteria [ 10 ], which may not be homogenously distributed in the vicinity of the roots. In addition, signalling molecules may enhance or impair the growth of certain taxa [ 16 , 17 ], as would the plant immune responses [ 18 ]. Competition between the members of the community by multiple mechanisms and different levels of spatial exclusion [ 85 ] likely contribute as well to a complex mixture of processes that control the rhizosphere microbiome assembly. The inference of taxa associations across treatments showed that communities overall were dominated by positive interactions, which were approximately three times more abundant than negative interactions (Fig. 3 D, E). This finding contrasts with a current assumption that competition would be the dominant type of interaction between bacterial species [ 86 ]. However, positive correlations have been found to dominate in the rhizosphere microbiome [ 23 ]. Network modularity was almost independent of the community state, environment or the inoculation with P. protegens CHA0, ranging from 0.7049 in the bulk soil of not inoculated stable NatCom to 0.8071 in the bulk soil of the inoculated stable NatCom (Fig. 3 D, E). This is in agreement with the modularity scores observed in other studies focusing on early rhizosphere microbiome and bulk soil assemblages [ 23 ]. Interesting differences, however, were found among the attributed keystone taxa in the different conditions and treatments. This attribution is based on the hub centrality score and indicated, for example Singulisphaera as a keystone taxon in all networks from samples inoculated with P. protegens CHA0, regardless of the community state or the environment. In non-inoculated samples, Pirellula had a dominating role except for the wheat rhizosphere of the growing community where other taxa such as Agromyces emerged as keystone. Both Singulisphaera and Pirellula are largely understudied members of the Planctomycetes class, usually ubiquitous in moderately acidophilic or mesophilic terrestrial habitats [ 87 ]. This difference may not be a direct effect of inoculation with P. protegens CHA0 but rather resulting from an indirect process affecting the interaction network of taxa. The reason is that the abundances of the two Planctomycetes members were not different in the presence or absence of the inoculant (see below) and might suggest that both genera exhibit similar niche exploitation in our microcosm conditions. The assembly of the NatCom-derived wheat rhizosphere microbiome selects specific taxa Differential abundance analyses showed that the wheat rhizosphere environment produced a significant change in ASVs belonging to the Actinobacteriota, Bacteroidota, Firmicutes, Proteobacteria and Verrucomicrobiota phyla (Supplementary Fig. S 5 ). The number of ASVs that were significantly enriched in the wheat rhizosphere was roughly two times higher in the stable NatCom compared to the growing condition, with 33 ASVs enriched in the wheat rhizosphere compared to 12 ASVs specifically enriched in the bulk soil inoculated with P. protegens CHA0 (and 29 compared to 15 ASVs in the non-inoculated systems, respectively). These changes reflect what would be expected for a community assemblage selecting a specialized rhizosphere microbiome [ 11 , 88 ]. The enrichment of specific taxa in the wheat rhizosphere is also consistent with the previously observed increase in diversity, in both stable and growing conditions (Fig. 1 D) and the divergence of rhizosphere communities from their bulk soil counterparts (Fig. 2 A). Taxa specifically enriched in the wheat rhizosphere belong to known plant-associated genera, in particular Flavobacterium , Paenibacillus , the Rhizobium group, Enterobacter and Pseudomonas [ 89 , 90 ], which are also found associated with the wheat rhizosphere [ 29 , 88 ]. The specific enrichment of ASVs in the stable bulk soil (12 and 5 ASVs in the inoculated and non-inoculated conditions, respectively) might correspond to bacteria that are negatively affected by the plant (e.g. by allelopathic signalling molecules [ 16 ] or by plant immune responses [ 18 ]) or to bacteria that can secure limiting nutrients through different scavenging mechanisms, such as through the use of siderophores [ 82 ]. No specific bulk soil enrichment of ASVs was observed for the growing community condition. Pseudomonas ASVs became differentially enriched depending on community states (Supplementary Fig. S 5 ). Pseudomonas ASV1173 was enriched in the rhizosphere of the growing conditions, whereas Pseudomonas ASV1142 was enriched in the rhizosphere of the nutrient-limited conditions. The enrichment of both Pseudomonas occurred independently of the inoculation pattern. This suggests an additional component of selection of potential competitors to CHA0 depending on the state of the resident community. The establishment of Pseudomonas protegens alters the relative abundance of other NatCom Pseudomonas ASVs We further explored differential changes in the relative abundance of ASVs in the growing or stable NatComs in response to the inoculation with P. protegens CHA0. Overall, the number of changing ASV relative abundances was limited to a few taxa (Fig. 4 A, Supplementary Table S 6 ), consistent with previous studies in which the introduction of plant-beneficial inoculants resulted in small and transient changes in the overall rhizosphere microbiome (at least at this level of taxa resolution) [ 27 , 91 ]. Most changes occurred in the wheat rhizosphere (Fig. 4 A). Under nutrient-limited conditions, we observed a strain-specific selection by root exudates and/or the presence of the inoculant (e.g. Paenibacillus , Flavobacterium and Pantoea , Fig. S 4 ). In the growing conditions, changes affected Pseudomonas ASVs different from the inoculant, i.e. ASV1168 and ASV1173, suggesting direct or indirect competition with the inoculant. The Pseudomonas genus is a highly diverse bacterial taxa, usually found in soils or associated with plants [ 92 , 93 ], in which kin competition has been previously reported [ 22 , 94 ]. Fig. 4 The establishment of Pseudomonas protegens causes discrete changes in growing or stable NatComs. A Differential abundance analyses comparing samples from the growing or stable community states, in the bulk soil or the wheat rhizosphere, inoculated with P. protegens CHA0 with those uninoculated at the initial or final sampling times (dpi, days post inoculation). The wheat rhizosphere (Rhiz.) was not sampled at the initial sampling time (1 dpi) due to limited biomass. Dots represent ASVs, with sizes according to adjusted (adj.) p -value and coloured according to their phylum assignation. The numbers of differentially abundant ASVs are indicated above (left, more abundant in the inoculated condition; centre and grey, not significantly different between conditions; right, more abundant in the uninoculated condition). ASVs with a |log 2 FoldChange| > 2.5 and a adj. p -value < 0.01 were considered significant. B Average relative abundance of the top five Pseudomonas ASVs at the initial (1 dpi) or final (9 dpi) sampling time. ASV1148 (green) corresponds to P. protegens CHA0. C CSS-normalized relative abundances of individual Pseudomonas ASVs across community states, environments, and inoculation patterns. Bars (±standard error) represent the average. Coloured dots represent individual replicates. Significance is based on Kruskal-Wallis rank-sum test with LSD post hoc analysis and p -value corrected by fdr. Different letters indicate significant differences between groups at p -value < 0.05. The type strain closest to the ASV is indicated below the ASV name (% sequence identity) A detailed exploration of the top five most abundant Pseudomonas ASVs (Fig. 4 BC) revealed that in growing conditions, ASV1142 (assigned to Pseudomonas putida , Supplementary Table S 7 ) dominates the Pseudomonas fraction of the communities, with up to ca . 30% of the relative abundance of the total microbiome at the first sampling time (Fig. 4 B). However, under stable, nutrient-limited conditions, ASV1142 became scarce irrespective of P. protegens CHA0 inoculation, suggesting a nutrient-based growth limitation rather than competition with the inoculant (Fig. 4 B). In contrast, the abundance of ASV1142 significantly increased in the wheat rhizosphere compared to stable bulk soil conditions (Fig. 4 B, Supplementary Fig. S 5 ). However, ASV1168, ASV1169 (both assigned to P. turikhanskensis ) and ASV1173 (assigned to Pseudomonas koreensis , Supplementary Table S 7 ) showed a reduced relative abundance when co-inoculated with CHA0 in the wheat rhizosphere of growing conditions (Fig. 4 C), which may indicate competition. In fact, ASV1168, ASV1169, ASV1173 and CHA0 (ASV1148) all belong to different subgroups within the Pseudomonas fluorescens complex of species, largely known for their positive interaction with plants [ 92 , 93 ], which make them likely dwellers of the wheat rhizosphere environment. To verify this further, we conducted competition experiments (in absence of resident NatComs) between the three closely related Pseudomonas strains (Fig. 4 , Supplementary Fig. S 6 ), i.e. P. protegens CHA0 (ASV1148), and Pseudomonas sp. ASV1168 and ASV1173, both in pairs and in triplets. The results showed that CHA0 and ASV1168 are able to coexist and reach similar proportions either in bulk soil or in the wheat rhizosphere (Fig. 5 A, Supplementary Table S 8 ). In contrast, both strains were able to outcompete Pseudomonas sp. ASV1173 in pairs, to a higher extent in the bulk soil than in the wheat rhizosphere (Fig. 5 A). Co- or triple inoculation with P. protegens CHA0 increased the shoot and root biomass of the wheat plants (Fig. 5 B), but the response was higher with the CHA0-ASV1173 inoculation than with the CHA0-ASV1168 pair (Fig. 5 B). This could be due to the displacement of ASV1173, thus increasing the abundance of P. protegens CHA0 (Fig. 5 A) capable of exerting this growth-promoting effect. Fig. 5 Competition of Pseudomonas strains in bulk soil or in the rhizosphere of wheat. A Pairwise (left, 1:1 ratio) or triplewise (right, 1:1:1 ratio) competition of Pseudomonas protegens CHA0, Pseudomonas sp. ASV1169 and Pseudomonas sp. ASV1173 in bulk soil or in the rhizosphere of wheat in growing conditions after 9 days without the NatCom. The competition index (CI) was calculated using the formula: CI = (CFUs of Sx at t 9 days — CFUs of Sx at t 0 days)/(CFUs of Sy at t 9 days — CFUs of Sy at t 0 days), where Sx and Xy denote the two strains being compared. Four technical replicates of four biological replicates (consisting of equal pools from four rhizospheres or bulk soil samples) were performed. The red dotted line indicates a competition where both strains would not be influenced by the presence of one another. Above this threshold, the competitor strain (CHA0 or ASV1168) outcompetes the other strain. Stacked bar plots below show the average percentage of CFUs recovered per tested strain. Significant pairwise differences are calculated according to Wilcoxon rank-sum test. Not significant (n.s.): p -value > 0.05. B Plant growth-promotion measurements in response to the inoculation of the two or three Pseudomonas strains compared in A . Root and shoot fresh weights were calculated within the four pools of four rhizospheres or shoots per replicate. The length of leaves was measured individually and averaged by the pools of four plants shown for the other parameters. Significant differences were calculated using the Kruskal-Wallis rank-sum test with LSD post hoc analysis and p -value corrected by fdr. Different letters indicate significant differences between groups at p -value < 0.05 The results obtained from these competition assays contrast with those obtained in the presence of a resident soil microbiome, where an order of magnitude higher normalized relative abundance of P. protegens CHA0 compared to ASV1168 was achieved in the wheat rhizosphere or bulk soil within the growing NatCom (Fig. 4 C). However, in direct competition, in the absence of the NatCom, the two strains established at similar numbers. Furthermore, ASV1168 showed a significantly reduced abundance when exposed to P. protegens CHA0 in the wheat rhizosphere (Fig. 4 C). Several factors could explain these differences, including the tagging of the strains (although no difference in growth rates compared to their wild types was observed, Supplementary Fig. S 6 ), indirect fine-tuning interactions with other members of the microbiome [ 23 , 29 ], or they might respond to different spatial colonization patterns on the plant roots [ 95 ], highlighting the importance of structurally complex environments, such as the wheat rhizosphere or the soil matrix for the prevalence of certain bacteria."
} | 9,151 |
37896374 | PMC10610547 | pmc | 8,271 | {
"abstract": "Self-polarized energy harvesting materials have seen increasing research interest in recent years owing to their simple fabrication method and versatile application potential. In this study, we systematically investigated self-polarized P(VDF-TrFE)/carbon black (CB) composite thin films synthesized on flexible substrates, with the CB content varying from 0 to 0.6 wt.% in P(VDF-TrFE). The presence of –OH functional groups on carbon black significantly enhances its crystallinity, dipolar orientation, and piezoelectric performance. Multiple characterization techniques were used to investigate the crystalline quality, chemical structure, and morphology of the composite P(VDF-TrFE)/CB films, which indicated no significant changes in these parameters. However, some increase in surface roughness was observed when the CB content increased. With the application of an external force, the piezoelectrically generated voltage was found to systematically increase with higher CB content, reaching a maximum value at 0.6 wt.%, after which the sample exhibited low resistance. The piezoelectric voltage produced by the unpoled 0.6 wt.% CB composite film significantly exceeded the unpoled pure P(VDF-TrFE) film when subjected to the same applied strain. Furthermore, it exhibited exceptional stability in the piezoelectric voltage over time, exceeding the output voltage of the poled pure P(VDF-TrFE) film. Notably, P(VDF_TrFE)/CB composite-based devices can be used in energy harvesting and piezoelectric strain sensing to monitor human motions, which has the potential to positively impact the field of smart wearable devices.",
"conclusion": "5. Conclusions The composite films were investigated using multiple characterization techniques, demonstrating their high material quality across all CB percentages. The piezoelectric voltage generated via the composite films, under similar applied force, increased monotonically with higher CB concentration for both poled and non-poled films, reaching a peak in piezoelectric voltage generation at 0.6 wt.%, beyond which the films exhibited low resistance due to conducting bridges formed by the CB. At a 0.6 wt.% CB composition, we measured the highest peak-to-peak output voltage of 3 V, which is six times higher than that of the unpoled 0 wt.% CB film. The piezoelectric properties of the unpoled composite films also exhibited excellent stability with time, in contrast to the rapid reduction observed for poled films, leading to superior piezoelectric performance for the unpoled 0.6 wt.% CB film compared to the poled 0 wt.% CB film after a week. The superior piezoelectric performance of the unpoled 0.6 wt.% composite films was further enhanced with poling, resulting in high d 33 values of 10.5 and 35 pC/N, respectively, which are among the best reported so far for all carbon-based composite P(VDF-TrFE) films.",
"introduction": "1. Introduction Recent advancements in flexible piezoelectric materials have paved the way for the potential realization of self-powered flexible devices in wearable electronics and other fields [ 1 , 2 , 3 ]. These materials have the ability to efficiently convert various forms of mechanical force into electrical power, eliminating the need for external power sources. Among the materials explored for this purpose, polyvinylidene fluoride (PVDF) and its copolymer P(VDF-TrFE) have stood out for their outstanding properties, including high piezoelectric coefficients, improved crystallinity, enhanced remnant polarization, and superior temperature stability [ 4 , 5 , 6 , 7 ]. P(VDF-TrFE) is classified as a ferroelectric polymer with an inherent β phase structure, which is achieved by changing the TrFE molar ratio with respect to PVDF [ 8 , 9 ]. However, P(VDF-TrFE) has a relatively low piezoelectric coefficient [ 9 , 10 , 11 ]. To enhance the piezoelectric coefficient and optimize energy conversion efficiency, aligning the dipoles within P(VDF-TrFE) films in a specific orientation is crucial. Traditional methods, such as polarization under a high electric field [ 3 , 12 ], electrospinning [ 13 , 14 ], corona poling [ 15 , 16 ], or thermal poling [ 17 ], have been employed to achieve this self-alignment. However, these methods often involve complex processing steps, are time-consuming, and may lead to dielectric breakdown within the film when a high electric field is applied, leading to reduced yield and constraints on practical implementation. Recently, researchers have explored the fabrication of self-poled polymer/composite films using various techniques, such as Langmuir–Blodgett deposition [ 18 ], casting [ 19 , 20 , 21 ], and the incorporation of nanofillers such as ZnO, Yb3+, BTO, CNT, MXene, and graphene oxide [ 22 , 23 , 24 , 25 , 26 , 27 , 28 , 29 ]. In these approaches, nanofillers play a crucial role as nucleating agents, inducing the alignment of polymer molecular chains toward the nanofiller surface. This alignment is facilitated by the strong interaction between the negatively charged fluorine atoms and the positively charged hydrogen atoms originating from the hydrophilic tail group (hydroxyl group) on the nanofiller surface. As a result, the seed layer aligned perpendicular to the substrate, and subsequent layers were arranged in an upward fashion. This alignment not only simplifies the fabrication process but also enhances versatility, scalability, and cost-effectiveness while reducing energy consumption. Among carbon-based nanofillers, carbon black is considered a promising candidate owing to its excellent electrical properties, high chemical and thermal stability, abundant availability, affordability, and a straightforward synthesis process [ 30 ]. The presence of –OH groups on the surface of carbon black facilitates the uniform arrangement of P(VDF_TrFE) molecular backbones. Previous studies have demonstrated significant enhancements in open-circuit voltage and harvested power density when CB is incorporated into polymer films. Wu et al. [ 31 ] introduced CB into poly(vinylidene fluoride-co-hexafluoropropylene) (PVDF-HFP) and observed a remarkable enhancement in the open-circuit voltage and harvested power density of the composite films after poling, with improvements of 104% and 364%, respectively, compared to pristine PVDF-HFP. Similarly, Alamusi et al. [ 32 ] fabricated P(VDF-TrFE)/CB composite films, achieving an open-circuit voltage of 10.07 V for the 0.8 wt.% poled CB composite film, which exceeded the 5.63 V obtained via the pure poled P(VDF-TrFE) film by a factor of 1.7. However, these studies mainly focused on relatively thick films prepared through post-poling treatments, which can be labor-intensive. Furthermore, comprehensive characterization and long-term performance evaluation were often lacking, which are critical aspects for practical applications. In this study, we systematically incorporated carbon black into P(VDF-TrFE) at varying compositions (0–1 wt.%) to prepare self-poled P(VDF-TrFE)/CB composite films using a simple spin-coating process. Our investigation incorporated an in-depth analysis of the influence of CB composition on the P(VDF-TrFE) matrix and the formation of the β phase, contributing to high piezoelectric output. To accomplish this, we utilized a range of characterization techniques, including optical microscopy, atomic force microscopy (AFM), X-ray diffraction (XRD), and Fourier-transform infrared (FTIR) spectroscopy. Furthermore, we measured the piezoelectric performance and piezoelectric coefficient (d 33 ) of the resulting CB composite films under poled and unpoled conditions. The unpoled 0.6 wt.% CB composite film exhibited exceptional performance, superior to that of the unpoled composite films, and maintained its piezoelectric performance consistently over time. These findings significantly exceeded the performance of the poled films and were confirmed in a 7-day study.",
"discussion": "3. Results and Discussion Material Characterization Following synthesis, the surface morphology of the resulting P(VDF-TrFE)/CB composite films was examined. Optical microscope images were captured using Olympus BX41M-LED at 50× magnification. In addition, atomic force microscopy (AFM, Veeco Dimension 3100) operated in tapping mode was employed to provide higher-resolution images of the composite films, and the AFM images were subsequently processed using the dedicated AFM software. To determine the effect of blending CB to the P(VDF-TrFE) polymer matrix and the crystallinity of the P(VDF-TrFE)/CB composite films, X-ray diffraction (XRD) measurements (Rigaku Smart Lab system) were made on the composite films (with CB varying from 0 to 1 wt.%) using Cu Kα radiation (wavelength 15.406 nm) in the 2θ range from 5° to 90° with a step size of 0.5°. Furthermore, FTIR spectra were measured (model no: Thermo Scientific Nicolet380) in the range of 4000–400 cm −1 with 64 scans at a 4 cm −1 resolution. The percolation threshold (transition point from insulator to conductor) and conductivity measurements were conducted using a Source Measure Unit (SMU, B2902A, Keysight, Santa Rosa, CA, USA). To measure the piezoelectric voltage output of the composites, they were excited by an external shaker (LDS V201, Brüel & Kjær, London, UK), and the voltage waveforms were recorded using a digital storage oscilloscope (DSO 5102P, Hantek, Qingdao, China). Figure 2 a shows the XRD spectrum of raw carbon black and P(VDF-TrFE) powder with a molar ratio of 55:45. Raw P(VDF-TrFE) powder exhibits a prominent β-phase peak (110/200) at 2θ = 19.1°, which experimentally confirms the presence of the TrFE unit in more than 20%, directly crystallized into the β-phase within the polymer [ 33 , 34 ]. Additionally, a broader peak at 40.8° corresponds to the diffraction plane (111/201), further validating the presence of the β-phase in the polymer matrix. Notably, no significant peak was observed at 2θ = 18.27°, corresponding to the (100) crystal planes of the α-phase of the P(VDF-TrFE) polymer [ 35 ]. In the XRD results for raw CB powder, the (002) diffraction peak appears at 24.5°, along with a broader and weaker (001) peak at 43°, consistent with previous reports [ 36 , 37 ]. Figure 2 b shows the XRD diffraction peaks of the 55/45 copolymer films crystallized at different temperatures (T cr ) to determine the optimum crystallization temperature for these films. Typically, crystallization begins when the film is subjected to a temperature above the Curie temperature (T c = 60 °C for 55:45 mol ratio). We found that at a crystallization temperature (T cr ) of 60 °C for P(VDF-TrFE), the maximum diffraction angle appeared at 2θ = 19.8°, which is close to 20.12°, corresponding to the Bragg diffraction of (110)/(200) of the β-phase. However, the diffraction peak intensity of the composites annealed at 60 °C is relatively small compared to peak intensities at other crystallization temperatures and existed within the amorphous region. As the crystallization temperature (T cr ) increased, the amorphous region gradually disappeared, and the peak intensities became narrower and sharper. Meanwhile, the diffraction angle (2θ) shifted to a lower value, changing from 20.12° to 19.3° (with an increase in d-spacing from 4.46 to 4.59 Å) as displayed in Figure 2 b. This shift was due to the ferro-to-paraelectric transition occurring during the crystallization phase, which changed the trans-planar conformation (TT) to the trans-gauge conformation (TG). Consequently, it shifted the prominent β-phase peak from 2θ = 20.12° to lower values and increased peak intensities as the crystallization temperature (T cr ) increased [ 38 ]. Moreover, the full width at half-maximum (FWHM) value, which is directly related to the degree of crystallinity, gradually improved as the annealing temperature increased from 60 to 140 °C. This observation clearly indicates an enhancement in the electroactive polar β-phase content in the composite films. We investigated the pure P(VDF-TrFE) film under various annealing temperatures (e.g., 60 °C, 80 °C, and 100 °C) and found a similar trend (see Supplementary Figure S1 ). From the XRD results, we can conclude that annealing the composite at 140 °C results in optimal crystalline quality, corresponding to the most stable β-phase in the film. Table 1 summarizes the diffraction peak angles (2θ), interplanar spacing (D), FWHM values for the prominent β-phase peak, and corresponding peak intensities of the 0.6 wt.% CB film annealed at various crystallization temperatures. Figure 2 c displays the XRD pattern of P(VDF-TrFE)/CB composite films, ranging from 0 to 1 wt.%, within the scan range of 10° to 90°. An analysis of the XRD pattern reveals that a well-defined diffraction peak appears at an angle of 19.3°, corresponding to the (110)/(200) planes of the β-phase, which comprises all-trans TT conformation [ 39 ]. In particular, we did not observe any significant peak for CB due to its amorphous nature and the low percentage of CB present in the polymer matrix. This clearly indicates that the addition of CB does not significantly alter the crystalline structure of P(VDF-TrFE). Figure 2 d shows the magnified view of XDR spectra of different composite films in the scan range of 17°–22°, varying from 0 to 1 wt.%. With an increase in CB content from 0 to 0.2, 0.4, 0.6, 0.8 wt.%, and 1 wt.% in the composite, XRD peak intensity varied from 5805 to 6589, 7903, 9266, 7892, and finally to 4986, respectively. Above 0.6 wt.%, a reduction in the diffraction peak intensity and an increase in FWHM were observed, as seen in the inset image of Figure 2 d, indicating that the best crystalline properties were achieved for the 0.6 wt.% composition. However, when the amount of CB percentage exceeds the optimum amount of 0.6 wt.%, the aggregation created by CB reduces the formation of the β-phase in the polymer. The homogeneous dispersion of carbon black particles within the polymer matrix is a major factor in the increased intensity. Similarly, Yaseen et al. [ 40 ] observed the same trend with P(VDF-TrFE)/reduced graphene oxide (rGO) composite films. The observed intensities for all composite films, along with interplanar spacing (D) and FWHM, are tabulated in Table 2 . FTIR characterization of the films was also conducted to confirm the XRD results and study the interaction between the nanoparticle’s surface and P(VDF-TrFE). The formation of the beta phase in P(VDF-TrFE) was identified by examining three important absorbance peaks (850 cm −1 , 1288 cm −1 , and 1400 cm −1 ) in the FTIR spectra. The 1400 cm −1 band corresponds to the -CH 2 wagging vibration, while the 1288 cm −1 and 850 cm −1 bands are attributed to the -CF 2 symmetric stretching, with dipoles parallel to the b-axis [ 41 ]. Figure 3 a illustrates the FTIR spectra of P(VDF-TrFE)/CB thin films ranging from 0 to 1 wt.%, and the prominent peaks for the β-phase are consistent for all CB compositions within the IR detection limit (1400 cm −1 to 400 cm −1 ). This indicates that the crystalline quality remains unaffected by the incorporation of CB nanoparticles. The P(VDF-TrFE) film and other CB composite films show increasing intensity in all observed peaks (1400 cm −1 , 1290 cm −1, and 850 cm −1 ) as the CB composition is increased up to 0.6 wt.% beyond which it remains constant or even decreases somewhat. This increase is attributed to the specific interaction between hydroxyl (–OH) groups found on the surface of carbon nanofillers and CF 2 segments of P(VDF-TrFE). This interaction becomes more pronounced with an increase in CB content, reaching a maximum of 0.6 wt.%. Figure 3 b provides a magnified view of the peaks in the scan range of 750 cm −1 to 900 cm −1 , allowing for the calculation of the percentage of β-phase crystallization F(β) by measuring the absorbance intensity of the β-phase and α-phase using the following formula: [ 42 ]\n (1) F ( β ) = X β X α + X β = A β ( K β / K α ) A α + A β = A β 1.26 A α + A β \nwhere X α and X β are the crystalline mass fractions of the α and β-phases, and A α and A β correspond to absorbance at 764 cm −1 and 850 cm −1 , respectively [ 40 , 43 ]. The values of the absorption coefficients result in K β / K α = 1.26. FTIR measurements were carried out to calculate F(β) for various CB composites, and the results are plotted in the inset of Figure 3 b and listed in Table 2 . As shown in Figure 3 b inset, F(β) is approximately 76% for the pure P(VDF-TrFE) film prepared and crystallized at 140 °C. It increases monotonically until it reaches a maximum of ~97% for the 0.6 wt.% CB composition. However, at higher CB contents, the incorporation of carbon particles has a negative impact on the β-phase formation, leading to a reduction in the percentage of beta crystallinity to 75% [ 44 ]. These results are in good agreement with the XRD results discussed earlier, indicating that the best film quality is obtained for 0.6 wt.% CB. Moreover, the broadened absorbance peak occurring between 3600 and 3400 cm −1 implies the formation of intermolecular hydrogen bonding between -CF2- dipoles and the hydrophilic groups from CB, as well as the remaining oxygen-containing groups from CB [ 40 , 44 ]. The OH stretching is much stronger for the 0.6 wt.% concentration than at the lower concentration (0 wt.%), clearly indicating that more –OH groups of CB form bonds with the most negatively charged fluorine atoms in the P(VDF-TrFE) molecular chains (see Supplementary Figure S2 ). This explains the possible electrostatic interaction between positively charged hydrogen atoms drawn from the hydrophilic tail group (–OH) of carbon black and negatively charged fluorine atoms from P(VDF-TrFE), as seen in Figure 4 . Such bonding is favored due to the large electronegativity differences between the atoms involved. The hydrophilic nature of the –OH groups causes the seed layer to align perpendicularly to the substrate. As shown in the enlarged view in Figure 4 , other induced dipoles resulting from hydrogen intermolecular bonding can interact with each other within subsequent polymer matrices. This interaction occurs between carbon black and –CH 2 dipoles, leading to local alignment during crystallization [ 24 , 25 , 28 , 45 ]. The interaction between –CF 2 –CH 2 – dipoles and carbon black can be confirmed by investigating the –CH 2 symmetric and asymmetric stretching vibrational bands at 3012 cm −1 and 2978 cm −1 in the FTIR spectra, which are not associated with any other vibrational bands. These vibrational bands shifted to lower frequencies as the CB loading increased; meanwhile, the absorbance peak intensity increased with respect to carbon weight percentage [ 20 , 40 ] (see details in Supplementary Figure S2 ). Figure 5 shows optical images of different CB composite films at 50× magnification with a 500 µm scale bar, ranging from 0 to 1 wt.%. In Figure 5 a, the smooth surface of the pure P(VDF-TrFE) film is illustrated, while Figure 5 b–d show the distribution of CB particles within the polymer matrix. These CB particles are clearly noticeable in optical images, consistent with an earlier report [ 46 ]. As shown in Figure 5 b, CB particles diffuse randomly and form tiny agglomerates at lower CB wt.%. As the CB fraction increases, these agglomerates become larger and larger, ultimately creating a conductive path between them. However, at very high CB content (~1%), structured agglomerates are no longer formed. Instead, the polymer matrix becomes saturated, resulting in a continuum of particles, indicated by a uniform dark color, as seen in the inset of Figure 5 d. We also studied the surface morphologies of the composite films using AFM to analyze nanoscale variations in surface roughness caused by CB incorporation. Figure 6 displays AFM images (5 × 2.5 µm) of different films with CB concentrations ranging from 0 to 0.6 wt.%. These images generally reveal uniformly distributed “rice grain”-like crystallites with dimensions in the tens of nanometers. Previous studies have also reported similar rice grain domains for pure P(VDF-TrFE) films annealed at 140 °C [ 47 ]. In contrast, when the films were subjected to temperatures near their melting point (T m = 153 °C), we observed more interconnected nanofiber-like crystallites, characterized by a higher roughness of approximately 36.8 nm for the 0 wt.% CB sample (see Supplementary Figure S3 ). In general, nanofillers tend to increase the surface roughness [ 48 ], as observed in these P(VDF-TrFE)/CB composites annealed at 140 °C. The lowest roughness, approximately 5 nm, was observed for the 0 wt.% CB film. This roughness increased progressively with increasing CB content: 5.2 nm, 8 nm, 21 nm, and finally 37 nm for the CB compositions of 0.2 wt.%, 0.4 wt.%, 0.5 wt.%, and 0.6 wt.%, respectively. A 600 nm line profile across an elevated carbon aggregate “island” region on the 0.5 wt.% CB film is shown in Figure 6 e, which indicates a height and span of ~400 nm for the island."
} | 5,249 |
36005510 | PMC9409691 | pmc | 8,272 | {
"abstract": "The development of harmless substances to replace biocide-based coatings used to prevent or manage marine biofouling and its unwanted consequences is urgent. The formation of biofilms on submerged marine surfaces is one of the first steps in the marine biofouling process, which facilitates the further settlement of macrofoulers. Anti-biofilm properties of a synthetic polyphenolic compound, with previously described anti-settlement activity against macrofoulers, were explored in this work. In solution this new compound was able to prevent biofilm formation and reduce a pre-formed biofilm produced by the marine bacterium, Pseudoalteromonas tunicata . Then, this compound was applied to a marine coating and the formation of P. tunicata biofilms was assessed under hydrodynamic conditions to mimic the marine environment. For this purpose, polyurethane (PU)-based coating formulations containing 1 and 2 wt.% of the compound were prepared based on a prior developed methodology. The most effective formulation in reducing the biofilm cell number, biovolume, and thickness was the PU-based coating containing an aziridine-based crosslinker and 2 wt.% of the compound. To assess the marine ecotoxicity impact of this compound, its potential to disrupt endocrine processes was evaluated through the modulation of two nuclear receptors (NRs), peroxisome proliferator-activated receptor γ (PPARγ), and pregnane X receptor (PXR). Transcriptional activation of the selected NRs upon exposure to the polyphenolic compound (10 µM) was not observed, thus highlighting the eco-friendliness towards the addressed NRs of this new dual-acting anti-macro- and anti-microfouling agent towards the addressed NRs.",
"conclusion": "4. Conclusions GBA26, a synthetic gallic acid derivative that was recently designed following a lead optimization strategy, exhibited highly effective AF activity against the settlement of Mytillus galloprovincialis adhesive larvae, both in solution and after incorporation in a polyurethane (PU) marine coating, with higher potency than the previous analog and the commercial biocide tralopyril [ 12 ]. GBA26 also had a higher LC 50 /EC 50 ratio than the previous analog and tralopyril [ 12 ], thereby highlighting its safer profile against this hard fouler species. As bioactivity against a single hard fouler may not be seen in assays against soft fouling, the impact of GBA26 on biofilm prevention and/or reduction was studied in this work. It was shown that GBA26 was able to prevent the formation of biofilms of Pseudoalteromonas tunicata , and also to promote the reduction in preformed biofilm of this representative marine bacterium. Under a hydrodynamic assay that simulates the marine environment, formulations of a marine PU-based coating containing varying contents of this AF agent showed promising in vitro anti-biofilm performances when tested against biofilms of the P. tunicata microfouler. The new PU-based marine coating containing 2 wt.% GBA26 and the trimethylolpropane triaziridine propionate (TZA) crosslinker provided the best long-term performance. The improved performance of this coating formulation compared to the one containing only a similar amount of GBA26 indicated that the compatibility of the compound in this polymer matrix, and the service life of the generated matrix, were improved due to the crosslinks with TZA. Even though GBA26 was previously shown to have a safer profile than the previous analog and tralopyril on different trophic levels of aquatic organisms (mussel larvae, marine shrimp, and marine diatom [ 12 ]), possible interference of GBA26 on endocrine processes of aquatic organisms was assessed in this work to pursue a systematic ecotoxicological assessment [ 2 ]. All biocidal products require authorization before they can be placed on the market and the active substances contained in that biocidal product must be previously approved. In principle, active substances that fulfill the exclusion criteria will not be approved, which is the case with endocrine disruptors. It was observed in this study that this compound does not modulate the transcription of the selected NRs, at least in vitro and at low concentrations (≤10 µM). Although the present results support a lack of interaction of GBA26 with the nuclear receptors tested, caution should be taken and additional ecotoxicological assays should be performed in the future. These include a larger portfolio of NRs and tested taxonomic groups, together with chronic exposures combined with “omics” analysis. Given the AF activity of GBA26-based PU marine coatings against the selected hard and soft foulers and the ecotoxicological studies performed, the next step will be to up-scale the synthesis of this simple compound in order to test the improved GBA26-coating formulation developed in this work in a real marine environment, so as to evaluate the formulation’s effectiveness on the whole biofouling community. Selecting simple, nature-inspired chemical structures during the first stages of the AF discovery process will be rewarding at this stage as large quantities will be possible to obtain and at a low cost [ 14 ]. The synthesis must also be adapted to obey the Twelve Principles of Green Chemistry, namely the use of eco-friendly solvents, replacement of hazardous reagents, and waste minimization. Furthermore, the starting material for the synthesis of this gallic acid derivative is a natural compound that may be obtained from grape waste extract, allowing for the valorization of winery industries waste. By extracting the starting material from grape waste and transforming it into value-added AF product, both tasks that apply green chemistry methodologies, this new AF compound may pave the way for new affordable and sustainable AF agents.",
"introduction": "1. Introduction The development of innovative and eco-friendly technologies to combat marine biofouling is critical given the associated economic, environmental, and human health consequences [ 1 , 2 ]. Although the addition of biocides to marine coatings has been the most used and effective solution to avoid marine biofouling, the biocides currently used for this purpose are persistent, bioaccumulative, and can be toxic to some non-target marine organisms [ 3 , 4 ]. When a surface is submerged in water, the marine biofouling process is triggered by the accumulation and physical adsorption of organic molecules. This conditioning film creates the perfect environment for the settlement and growth of pioneer bacteria, which leads to the formation of a biofilm matrix. This initial process leads to the so-called secondary colonization, where a biofilm of multicellular species is formed. Tertiary colonization occurs with the capture of particles and organisms, such as the larvae of marine macro-organisms, including macroalgae, sponges, cnidarians, polychaetes, mollusks, barnacles, bryozoans, and tunicates [ 5 ]. In the last five years, several antifouling (AF) compounds capable of inhibiting the settlement of the macrofouler, Mytilus galloprovincialis (mussel) larvae, were synthesized by some of us [ 6 , 7 , 8 , 9 , 10 , 11 ]. After structure-AF activity relationship studies on several gallic acid derivatives, we recently discovered that N -(2-aminoethyl)-3,4,5-trihydroxybenzamide hydrobromide (GBA26) ( Figure 1 ) exhibited an optimized potency against the larvae settlement of this mussel with higher LC 50 /EC 50 than the previous parent compound and the emerging biocide Econea ® (tralopyril) [ 12 ]. This property was also maintained after the incorporation of GBA26 (1 wt.%) in a polyurethane (PU)-based marine coating [ 12 ]. The IUCN/SSC Invasive Species Specialist Group has listed M. galloprovincialis among the 100 “World’s Worst” invaders, highlighting the relevance of this species around the world. On the other hand, mussels can be considered a target and non-target organism [ 13 , 14 ], making these results even more relevant. On the coastal shores, mussels represent a key species operating as ecosystem engineers by providing habitats for other organisms, filtering out sediments and pollutants, and providing food for higher trophic levels. However, as bioactivity against a single hard fouler may not be seen in assays against soft fouling, in this work, the anti-biofilm performance of GBA26 in a solution was studied to assess if this compound is also capable of preventing the formation and/or reduction of biofilms of a common marine microfouler, Pseudoalteromonas tunicata ( Figure 1 ). This organism was applied for the assessment of the in vitro AF performance of surfaces and coatings [ 15 , 16 , 17 , 18 ]. Hence, new PU-based coating formulations containing GBA26 aziridine-based crosslinker (CL) to enable higher service life of coatings [ 19 ] were evaluated for P. tunicata biofilm growth ( Figure 1 ), at defined hydrodynamic settings [ 20 ] mimicking a real marine scenario. Regarding ecotoxicity, GBA26 was previously shown to have lower toxicity than tralopyril against Artemia salina , a marine shrimp. While tralopyril cause 100% mortality to A. salina , GBA26 did not cause more than 10% mortality at the same concentration (50 µM) ( Figure 1 ). To pursue a systematic ecotoxicological evaluation, the OECD 201 test was chosen to determine the impact of GBA26 on the growth of Phaeodactylum tricornutum , a marine diatom among the most common type of phytoplankton. GBA26 exhibited a safer profile than tralopyril and the previous analog ( Figure 1 ) [ 12 ]. However, ecotoxicity could also be expressed by the ability of a compound to disrupt endocrine processes. For example, tributyltin, a booster biocide used in AF paints which was banned since 2008, is widely known to have an endocrine-disrupting action in mollusks [ 21 , 22 , 23 ]. Previous studies demonstrated that TBT, at a low ng/L range, modulates the nuclear receptor (NR) retinoid X receptor (RXR), inducing imposex development in gastropods [ 22 ]. SeaNine211 ® was also found to have endocrine disrupting and reproductive impairing effects [ 24 ]. Early this year the effects of short-term exposure to tralopyril (Econea ® ) on physiological indexes and endocrine function in turbot ( Scophthalmus maximus ) were reported [ 25 ]. This negative effect should be foreseen during the development of new effective eco-friendly AF agents ( Figure 1 ). Since NRs are ligand-activated transcription factors, participating in the regulation of numerous biological processes such as development, metabolism, and reproduction [ 26 ] we, therefore, investigated the ability of GBA26 to modulate the activity of selected NRs. Therefore, we also investigated the ability of GBA26 to modulate the activity of selected NRs.",
"discussion": "2. Results and Discussion 2.1. Anti-Biofilm Performance of GBA26 The in vitro anti-biofilm efficacy of GBA26, in several concentrations, was determined through a biofilm prevention assay (GBA26 mixed with inoculum) and a biofilm reduction assay (pre-formed biofilms exposed to GBA26 in solution) using P . tunicata ( Figure 2 ). Results of the two assays showed that GBA26 can prevent and reduce preformed biofilms of P . tunicata in a concentration-dependent manner. 2.2. PU-Based Coatings Containing GBA26 To assess the in vitro performance of GBA26 as an anti-biofilm agent in surface coatings, this compound was incorporated at different concentrations, approximately 1.0 to 2.0 wt.%, ( Table 1 ) into a representative, commercial, two-component, PU-based marine coating. Previous studies on the incorporation of GBA26 in a PU-based coating system showed that, although GBA26 had maintained AF activity against the settlement of M. galloprovincialis larvae, it was rapidly released from the PU-based marine coating. Therefore, to increase the service life of these coatings, further coating optimization was performed, according to an existing immobilization methodology [ 19 ], which involved the addition of the trimethylolpropane triaziridine propionate (TZA) crosslinker (CL) into the formulation. This aziridine-based crosslinker, widely used in polymeric formulations such as paints, reacts rapidly with nucleophilic functional groups of the compounds (e.g., amines, alcohols), promoting cross-links between different functional additives, and improving their compatibility with the polymeric systems [ 19 ]. The generated GBA26-based coating systems were further assessed in terms of anti-biofilm performance. 2.3. Pseudoalteromonas Tunicata Biofilm Formation under Defined Hydrodynamic Conditions Figure 3 shows the biofilm content in terms of the number of biofilm cells (cells/cm 2 ) of P. tunicata for the three investigated coating formulations. For the 1 wt.% GBA26 PU-based coating, the number of cells increased only from day 14 until day 49. No significant difference was observed between day 7 and day 14, suggesting that a strong anti-biofilm activity was exerted in the first 14 days. The anti-biofilm effect was observed for the 2 wt.% GBA26 PU-based coating in the first 21 days, after which the number of biofilm cells started to increase, although never reaching the number of cells observed for the 1 wt.% GBA26 PU-based coating. A long-lasting effect was observed for the 2 wt.% GBA26 PU-based coating containing the crosslinker (CL), GBA26/PU/CL, where the number of cells only started to increase from day 28 and the number of biofilm cells was lower (around 4 × 10 9 cells/cm 2 ) than the number of cells observed for the 2 wt.% GBA26 PU-based coating without CL at day 49 (around 8 × 10 9 cells/cm 2 ). The structural differences of P. tunicata biofilms, which were developed on the three tested surfaces after a 49-day assay, were assessed through a confocal laser scanning microscope (CLSM); this is similar to what was recently performed by other authors for novel anti-biofilm materials for marine applications [ 27 ]. Three-dimensional (3D) images of visualized stacks are presented in Figure 4 (showing the aerial view of biofilm and including a virtual shadow projection on the right-hand side which represents the biofilm section) and Figure S1 in the Supplementary Material (isosurface rendered 3D visualizations of the same data set used to obtain Figure 4 ). A thicker and denser biofilm, developed on the GBA26 PU-based coating (1 wt.% GBA26/PU), confirms the results obtained from the biofilm cell count ( Figure 3 ). On the other hand, biofilms formed on the top of the GBA26/PU/CL (2 wt.% GBA26) did not cover the entire surface area and only scattered cell aggregates could be observed. Regarding biofilm biovolumes, it was significantly lower for the GBA26/PU/CL surface when compared to 1 and 2 wt.% GBA26 PU-based coatings ( p < 0.05, Figure 5 A). Accordingly, the biofilm thickness was higher for the 1 wt.% GBA26 PU-based coating formulation than for the 2 wt.% GBA26 PU-based coating and the 2 wt.% GBA26 PU-based coating containing the CL, GBA26/PU/CL ( p < 0.05, Figure 5 B). 2.4. In Vitro Transcriptional Activation of HsPPARγ, DrPPARγ, and DrPXR Given that a high number of NRs exists across the metazoans (i.e., 48 NRs in humans, 73 NRs in teleosts [ 28 ]), it is not feasible to test all NR/test compound combinations, to assess possible NR-dependent disruption mechanisms upon exposure to novel contaminants. Since GBA26 is a polyphenolic compound, and polyphenols are described in the literature as activators of both PPARγ [ 29 ] and PXR receptors, the two receptors were selected for the in vitro transactivation assays [ 30 ]. PPARγ is a master regulator of an adipogenesis, and a known target of environmental chemicals, such as the AF biocide TBT and PXR, a modulator of detoxification responses exhibiting a highly plastic ligand-binding pocket [ 31 , 32 , 33 , 34 , 35 ]. Therefore, for the present study, we addressed the ability of GBA26 to modulate PPARγ from H. sapiens and the teleost, Danio rerio and D. rerio PXR. As expected, the control compounds, rosiglitazone and clotrimazole, yielded significant fold-induction values with H. sapiens PPARγ and D. rerio PXR, respectively, when compared to DMSO, confirming the validity of the present assay ( Figure 6 ). Overall, the results suggest that GBA26 does not modulate the transcription of the selected NRs, at least at low concentrations, suggesting that this new AF compound may not interfere with endocrine processes mediated by PPARγ (a known target of endocrine disruptors) and PXR (a highly promiscuous receptor involved in drug metabolism) in the studied species."
} | 4,141 |
31923911 | null | s2 | 8,274 | {
"abstract": "Natural visual systems have inspired scientists and engineers to mimic their intriguing features for the development of advanced photonic devices that can provide better solutions than conventional ones. Among various kinds of natural eyes, researchers have had intensive interest in mammal eyes and compound eyes due to their advantages in optical properties such as focal length tunability, high-resolution imaging, light intensity modulation, wide field of view, high light sensitivity, and efficient light management. A variety of different approaches in the broad field of science and technology have been tried and succeeded to duplicate the functions of natural eyes and develop bioinspired photonic devices for various applications. In this review, we present a comprehensive overview of bioinspired artificial eyes and photonic devices that mimic functions of natural eyes. After we briefly introduce visual systems in nature, we discuss optical components inspired by the mammal eyes, including tunable lenses actuated with different mechanisms, curved image sensors with low aberration, and light intensity modulators. Next, compound eye inspired photonic devices are presented, such as microlenses and micromirror arrays, imaging sensor arrays on curved surfaces, self-written waveguides with microlens arrays, and antireflective nanostructures (ARS). Subsequently, compound eyes with focal length tunability, photosensitivity enhancers, and polarization imaging sensors are described."
} | 374 |
34843460 | PMC8659421 | pmc | 8,275 | {
"abstract": "Finding out the physical structure of neuronal circuits that governs neuronal responses is an important goal for brain research. With fast advances for large-scale recording techniques, identification of a neuronal circuit with multiple neurons and stages or layers becomes possible and highly demanding. Although methods for mapping the connection structure of circuits have been greatly developed in recent years, they are mostly limited to simple scenarios of a few neurons in a pairwise fashion; and dissecting dynamical circuits, particularly mapping out a complete functional circuit that converges to a single neuron, is still a challenging question. Here, we show that a recent method, termed spike-triggered non-negative matrix factorization (STNMF), can address these issues. By simulating different scenarios of spiking neural networks with various connections between neurons and stages, we demonstrate that STNMF is a persuasive method to dissect functional connections within a circuit. Using spiking activities recorded at neurons of the output layer, STNMF can obtain a complete circuit consisting of all cascade computational components of presynaptic neurons, as well as their spiking activities. For simulated simple and complex cells of the primary visual cortex, STNMF allows us to dissect the pathway of visual computation. Taken together, these results suggest that STNMF could provide a useful approach for investigating neuronal systems leveraging recorded functional neuronal activity.",
"introduction": "Introduction One of the cornerstones for developing novel algorithms of neural computation is to utilize different neuronal network structures extracted from experimental data. The connectome, wiring diagrams , becomes an increasingly important topic, especially, for those relatively simple neuronal circuits that are well-studied, such as the retina [ 1 – 6 ]. Based on certain experimental techniques, the wiring diagram of neuronal connections has been identified for simple animal models, including Caenorhabditis elegans [ 7 ], Drosophila [ 8 ], and tadpole larva [ 9 ]. So far, most of these methods can only take a static view of connection strengths for neural circuits by imaging data, and the dynamics of synaptic strengths, which is a unique and essential feature of neural computation, is hardly estimated. The function of neuronal computation has been shown to be highly dynamics in the temporal domain with strong adaptation to stimulus statistics [ 10 , 11 ], nonlinear temporal integration [ 12 , 13 ], trial specific temporal dynamics [ 14 , 15 ]. The question of how to obtain a functional and dynamical neuronal circuit has been studied experimentally [ 16 ] and computationally [ 17 , 18 ] with great efforts in recent years. Spike-triggered non-negative matrix factorization (STNMF) is one of the methods proposed to infer the underlying structural components of the retina based on temporal sequences of spiking activities recorded in ganglion cells [ 17 ]. STNMF takes the advantage of machine learning technique NMF, which has a great capacity to capture local structures of given datasets [ 19 ]. It has been used recently to identify functional units localized in space and time in neuronal activities [ 20 – 26 ]. STNMF takes a step further to analyze the mapping between stimuli and neural responses leveraging neural spikes while leaving out non-responsive stimuli [ 17 , 27 ], with an assist of spare coding, as neurons generally fire with a low rate of spikes [ 28 ]. However, it is not clear whether the STNMF is applicable to dissecting a complete neural circuit with multiple stages or layers all formed by multiple spiking neurons. Here we address this question by comparing the true dynamic connection and strengths in a model and those estimated by STNMF. The model is a spiking neural network mimicking the feedforward connection at multiple stages in early visual systems, including the retinal ganglion cells (RGCs), lateral geniculate nucleus (LGN), and primary visual cortex (V1). We first demonstrated STNMF can reliably infer presynaptic spikes from postsynaptic spikes and obtain presynaptic strengths and dynamics for multiple spiking neurons projecting to a single postsynaptic neuron. Then we showed that when there are more than one postsynaptic neurons, STNMF is able to map out the entire neural circuit by analyzing each individual postsynaptic neuron. With a multiple layer neural network, STNMF can identify each layer in the model. Particularly, STNMF is applicable to the complex stimulus of natural images. Finally, we show that STNMF is applicable to V1-like simple and complex cells of neural networks with mixed cell types. Taken together, our results indicate that STNMF is an effective approach to describe the underlying neural circuits using neural spikes of single cells.",
"discussion": "Discussion In this study, we demonstrated that the STNMF is capable to dissect functional components of spiking neural networks and reconstruct spike trains of presynaptic neurons by analyzing spikes of the output neurons. Within feedforward networks with multiple stages or layers and multiple neurons, applying STNMF to spikes of neurons at the final layer allows us to recover the entire neural network, not only the structural components of neurons and synapses, but neuronal spikes of cascaded layers, which transfer the input stimulus to final output neurons. These results suggest the STNMF is a useful technique for interpreting neural spikes and uncovering the relevant functional and structural components of neuronal circuits. 0.1 The role of presynaptic neurons in postsynaptic neural spikes Here we demonstrated the scenarios where a postsynaptic neuron receives a few presynaptic neurons that all are firing spikes and contribute to the firing of the postsynaptic neuron. It is well known that a neuron has morphology with a dendritic tree receiving nonlinear inputs from presynaptic neurons [ 36 ]. The neuronal morphology varies significantly, depending on the cell type, location, and brain area [ 37 ]. Similarly, the firing rate of cells also varies remarkably depending on the dynamic states of synaptic strength [ 38 ]. A typical cortex cell with thousands of synapses maintains a very low firing rate [ 39 ]. Recent evidence, using advanced experimental techniques recording the activity of single synapses in vivo , shows that single synapses could be active, while the population of synapses is rather spare [ 40 ]. This indicates that, in terms of spikes of a postsynaptic neuron, only a small subset of synapses actively contribute to the somatic firing at one time, while most of the synapses are silent. Experimental observations and theories utilizing this feature suggest complex scenarios of the interaction of spare synaptic firing and dendritic computation at the single-cell level [ 40 ], and spare neural coding at the level of neural circuits [ 41 ]. The STNMF may have an advantage in utilizing these shreds of evidence for understanding the computational principle of neural coding. 0.2 Reconstruction of the dynamics of neuronal networks Recent experimental advances provide tools to reconstruct large-scale neural circuits [ 8 , 9 ] and relatively stereotyped retinal circuits [ 1 , 3 , 4 ]. However, these static connectomic structures can not explain ever-changing neuronal dynamics and reveal valuable functions performed by neurons. Taking the example of direction selectivity in retinal neurons, the structure basis was suggested as the asymmetrical distribution of inhibitory amacrine cells around ganglion cells [ 42 ], however, direction selectivity is rather dynamical and reversible [ 43 ]. Thus it is important to reconstruct the functional dynamics of neural networks. The methods that can analyze network connectivity using neural response are still limited. Granger causality [ 44 ], dynamic causal modeling [ 45 ], and transfer entropy [ 46 ]) are popular methods used for this purpose yet with certain limitations [ 47 – 50 ]. The STNMF, as a relatively new method, provides a different means to systematically investigate functional neural circuits using spikes. Together with other recent studies focusing on dynamical structures of neural networks [ 16 , 18 ], it is possible to incorporate dynamic components, such as synaptic strengths and presynaptic spikes, to reveal detailed functional organization of neural circuits. 0.3 Inferring neural spikes The complexity of dendritic organization in neurons depends on the type of neurons. For some neurons, such as Purkinje cells in the cerebellum, there is a large dendritic tree receiving tens of thousands of presynaptic inputs [ 51 ]. However, some neurons, such as unipolar cells of the cerebellum, have only one dendrite receiving one presynaptic input [ 52 ]. Yet, the underlying computations in both types of neurons are rich [ 53 ]. It is thought that many synapses are silent, perhaps at particular time points, during the spike dynamics of postsynaptic neurons. Thus, it is meaningful to extract the contribution of presynaptic neurons from the viewpoint of postsynaptic neurons. Here we noticed that STNMF can classify spikes of postsynaptic neurons into a set of spikes, where each set is considered as ab overall contribution of presynaptic neurons. The results of STNMF are meaningful, in that it allows us to obtain the dynamic strengths of presynaptic cells, according to whether they deliver the effect on spike dynamics of postsynaptic cells. Therefore, the outcome of STNMF is naturally for inferring spikes to capture the underlying dynamics of neural circuits, rather than static connections between neurons. In this sense, STNMF could provide more information than Granger causality, which tells the direction of information between neurons [ 54 ]. For experimental data where no spike can be exacted, such as graded signals in retinal bipolar cells [ 55 , 56 ], or the coarse version of neural signals, such as neuronal calcium imaging data [ 15 , 26 ], and local field potentials representing a small or large network of neural population [ 57 , 58 ], STNMF could be potentially applied to extract useful information within neural circuits, as long as neural signals are dynamics with meaningful states reflecting neural spikes. Given recent advances in experimental techniques for simultaneously recording multiple brain areas with single cell resolution [ 59 ], these data could yield interesting protocols for utilizing STNMF on the level of large scale neural circuits. 0.4 Multilayered neuronal networks A ubiquitous feature of neural circuits in the brain is that neurons are organized by layers or stages. Although there are dramatic feedback and/or recurrent connections between neurons [ 60 ], the information flow within recurrent neural networks could reinforce neurons to form a prevailing feedforward format of dynamics, utilizing synaptic plasticities [ 61 , 62 ]. One prominent example is the neural trajectory, in which different neurons fire at particular time points so that the overall dynamics of neural populations becomes a trajectory spanning over time, such as songbird neural dynamics [ 63 ], and space, such as memory dynamics of place cells [ 64 ]. Nevertheless, the dynamics of the neural network is controlled by multiple layers and pathways [ 65 , 66 ]. In some neural systems, feedforward networks are more prominent. The typical example is the visual pathway modeled here, starting the retina to LGN and visual cortex. The relatively simple organization of the retinal circuit makes it a perfect system for dissecting the dynamics and computations of the the multilayered neural network [ 67 , 68 ]. Leveraging the feature of macaque retina with less dense distribution and large size of photoreceptors away from the fovea, the STA analysis, using fine-size white noise checker images, can infer photoreceptors of the input layer while analyzing the spikes of ganglion cells of the output layer [ 67 ]. However, such an approach is difficult for analysis of general retinal neural systems, and STA analysis can not detect bipolar cells of the hidden layer [ 17 ]. The STNMF was introduced to consider the restricted two-layer network of bipolar cells and ganglion cells, where there are no spikes in bipolar cells [ 17 , 27 ]. Here we demonstrated that STNMF is applicable to fully spiking neural networks with multiple layers. It is well known that a simple three-layer perceptron with one hidden layer can greatly expand the computational power of artificial neural networks. Similarly, multilayered neural network presents many interesting features, such as synfire chains [ 69 ], of neural activities in neuroscience, resembling some experimental observations, such as songbird neural dynamics [ 63 ]. STNMF could serve as a tool for understanding these dynamics. Much effort has been made to characterize the neuronal receptive fields in LGN and visual cortex [ 70 – 74 ]. However, the computation in the visual pathway is carried out by different layers and stages [ 65 , 75 ], and there is no efficient way to dissect them systematically across multiple layers [ 76 ]. Here we demonstrated that the STNMF is able to identify the receptive fields of neurons in the input layer, even the STNMF was applied to output neurons in the final layer. Such an across-layer analysis of STNMF is a manifestation of nonlinear computation within neuronal networks. Spike response of neurons is an indication of the nonlinear computation using various ion channels in neurons [ 52 ]. Thus, the STNMF, leveraging the advantage of NMF for describing local structures of images, can naturally fit in the neuronal systems with spikes. The ultimate goal of reconstructing neural circuits is to utilize those neural and synaptic components for neural computation. In recent years, detailed neuroscience knowledge strengthens the bottom-up approach of neural network modelling [ 77 ], in which one prominent feature is to utilize neuroscience-revealed network structures to design, rather than hand-craft, possible artificial network architectures [ 78 ]. Here we indicated that the STNMF can detect computational components across layers or stages of cascade neural networks. Recent studies show that NMF variants can be combined with the framework of multilayer architecture [ 79 , 80 ] to learn a hierarchy of attributes between layers. Thus, one future direction is to extend STNMF to infer all the computational components simultaneously in multilayered neural networks. Therefore, further extension of STNMF is likely to be fruitful for understanding the hierarchical architecture of neuronal systems in the brain. 0.5 Limitations A variety of advanced experimental techniques in neuroscience can measure different types of functional neural signals. Spiking signal is one of the many formats. Other continuous signals measured for single cells, such as two-photon calcium imaging, as well as for coarse-scale cell ensembles, such as electroencephalogram and functional magnetic resonance imaging, can not infer spikes directly. Further effort is needed to adapt STNMF to investigate these non-spiking signals. Meaningful neural responses are often defined as peaks of these signals. Recent studies imply there is a close correlation between peaks of a neural signal of two-photon calcium imaging with spikes [ 81 , 82 ]. Thus, extracting peaks as spikes can make STNMF work for neural calcium imaging signals. Systematic studies are deserved for detailed examination of the coarse-scale non-spiking neural signal using STNMF. Although neural circuits are organized by layers across the brain and sensory information flows in a feedforward way, recurrent connections between neurons are also prevailed and useful for dynamic coding [ 83 , 84 ]. We showed that STNMF can work well in networks with weak recurrence and feedback. Future work is needed to extend STNMF to take into account recurrence. However, these structure indices are rather static. Dynamical routing of information in a network is more dramatic, which makes networks be in a regime of feedforward dynamics with recurrent structures [ 61 ]. Recent studies using graph theory suggest that neural network in the brain contains multiple ensembles of local community or module subnetworks [ 85 ]. One possible way is to utilize the coding principle of sparse firing and ensemble firing in a large network to separate the whole network into a set of local networks. One can apply STNMF iteratively and hierarchically through subsets of local networks for disentangling the effect of recurrent and feedforward connections on the information flow."
} | 4,181 |
37485539 | PMC10361621 | pmc | 8,276 | {
"abstract": "The microbiota inhabiting soil plays a significant role in essential life-supporting element cycles. Here, we investigated the occurrence of horizontal gene transfer (HGT) and established the HGT network of carbon metabolic genes in 764 soil-borne microbiota genomes. Our study sheds light on the crucial role of HGT components in microbiological diversification that could have far-reaching implications in understanding how these microbial communities adapt to changing environments, ultimately impacting agricultural practices. In the overall HGT network of carbon metabolic genes in soil-borne microbiota, a total of 6,770 nodes and 3,812 edges are present. Among these nodes, phyla Proteobacteria, Actinobacteriota, Bacteroidota, and Firmicutes are predominant. Regarding specific classes, Actinobacteria, Gammaproteobacteria, Alphaproteobacteria, Bacteroidia, Actinomycetia, Betaproteobacteria, and Clostridia are dominant. The Kyoto Encyclopedia of Genes and Genomes (KEGG) functional assignments of glycosyltransferase (18.5%), glycolysis/gluconeogenesis (8.8%), carbohydrate-related transporter (7.9%), fatty acid biosynthesis (6.5%), benzoate degradation (3.1%) and butanoate metabolism (3.0%) are primarily identified. Glycosyltransferase involved in cell wall biosynthesis, glycosylation, and primary/secondary metabolism (with 363 HGT entries), ranks first overwhelmingly in the list of most frequently identified carbon metabolic HGT enzymes, followed by pimeloyl-ACP methyl ester carboxylesterase, alcohol dehydrogenase, and 3-oxoacyl-ACP reductase. Such HGT events mainly occur in the peripheral functions of the carbon metabolic pathway instead of the core section. The inter-microbe HGT genetic traits in soil-borne microbiota genetic sequences that we recognized, as well as their involvement in the metabolism and regulation processes of carbon organic, suggest a pervasive and substantial effect of HGT on the evolution of microbes.",
"introduction": "1. Introduction Microorganisms are the foundation of the Earth’s biosphere and play an integral and unique role in various ecological processes and functions, where they interact to form complex functional networks ( Zhou et al., 2010 ). Soil is an important component of the global carbon cycle and is critical to climate change mitigation. Almost all life on Earth cannot leave the living soils, which emphasizes the significance of soil-borne microbiota for essential life-supporting processes (e.g., carbon and nitrogen cycling). The soil organic carbon (SOC) reservoir (about 1,500 Gt) is supposed to be greater than the sum of the carbon stocks in the air and global flora (about 560 Gt) ( van Elsas, 2019 ; Jassey et al., 2022 ). Microorganisms have much higher growth rates and carbon turnover rates than plants. The size of the soil organic carbon pool depends to a large extent on microorganisms, as their growth and activity balance the accumulation of organic carbon and its release through the decomposition of plant die-offs. Soil-borne microbes control the kinetics of soil carbon transformation by converting carbon from plants, incorporating carbon resources to increase biomass, and breaking down terrestrial organic compounds. For instance, strong biological methane-oxidizing activities in agricultural soils can lead to the emissions of biogenic CO 2 linked to CH 4 oxidation by a large biodiversity of methanotrophs ( Cappelletti et al., 2016 ). Globally, soil algae absorb about 3.6 Pg of carbon per year, which is 6.4% of annual terrestrial net primary productivity (NPP) and equivalent to 31% of global anthropogenic carbon emissions ( Jassey et al., 2022 ). Soil-borne microflora is not only dynamic temporally but also varied geographically. As a result, fluctuations in the activities and quantity of the communities that constitute such soil-borne microflora are frequently seen ( Fierer, 2017 ). The heterogeneity in soil-borne microbiota composition is primarily caused by the spatial heterogeneity of amounts and concentrations of, for example, nutrition, mineral resources, pH, and moisture in the soil mass ( Fierer, 2017 ; van Elsas, 2019 ), as well as shaped by genetic recombination and gene-specific selection processes ( Crits-Christoph et al., 2020 ). Besides, the carbon resource abundance and diversity in soil have been proven to correlate the ecological certainty during the bio-control of microbe-induced plant disease ( Zhou et al., 2023 ). Horizontal gene transfer (HGT), or the interchange of genetic material across phylogenetic clades, is thought to be an efficient strategy for dispersing reproductive fitness for prokaryotic and eukaryotic cells ( Huang, 2013 ; Daubin and Szöllősi, 2016 ; Li et al., 2022 ). Employing transmitting mobile genetic elements (MGEs) like plasmids, viruses, transposons, and gene transfer agents (GTA, tailed phage-like entities capable of packing and transferring random pieces of the host genome) as well as by direct absorption and assimilation of naked DNA by homologous or unauthorized recombination, new genes are transferred through HGT ( Thomas and Nielsen, 2005 ; Lang and Beatty, 2007 ). As a result, the origins of specific genes within a particular species can vary, and patterns of gene exchange between near and remotely affiliated lineages can be seen on various genomes ( Gogarten et al., 2002 ; Koonin, 2005 ; Kunin et al., 2005 ). HGT is commonly described based on contradicting phylogenetic trees upon genomic comparison ( Delsuc et al., 2005 ; Li et al., 2019 ). In recent decades, the evolution and HGT processes of genetic traits have been investigated in a variety of ecosystems, where multiple factors (e.g., environmental conditions, ancestral genome sizes) might have influenced the frequency of HGT during evolution ( Huang, 2013 ; Daubin and Szöllősi, 2016 ; Li et al., 2022 ). Stable HGT flux induced under selection was also suggested to enhance microbial interplay’s structural stability and thereby maintain microbial communities’ equilibrium ( Fan et al., 2018 ). For instance, it was reported that 9.6% of the genes within a prokaryotic genome were recently acquired on average ( Kloesges et al., 2011 ), while in E. coli 18% of genes were recently acquired via HGT ( Lawrence and Ochman, 2002 ), and in Rickettsiales 25% of core genes were recently transferred ( Hernández-López et al., 2013 ). Even though the terrestrial area is a continually fluctuating and difficult place ( Fierer, 2017 ), factors like limited nutritional and ion intensities that endorse competency, clay deposits that sustain the perseverance of bacteriophages and naked DNA, and the capacity of subsoil microbes to aggregate for genetic exchange suggest that HGT processes could be performing in this globalized context. Consistently, the transmission of antibiotic-resistance genes in the soil via MGEs (e.g., bacteriophage) has been highlighted as a public concern ( Gallo et al., 2019 ; Zhang et al., 2022 ). Besides, soil is stratified geographically and varied physiologically and biologically on various dimensions, which offers a unique habitat for microbiota of diverse background. Many agronomic factors, such as pH, saltiness, warmth, and humidity, influence the organization of the microbiota inside the complicated soil-borne biomass ( Frindte et al., 2019 ). The microscopic society’s shape and function in soils could also be influenced by temporal changes such as meteorological conditions, rhizosphere exudation, as well as other periodical supplies of plant organic matter ( Zheng et al., 2019 ; Chernov and Zhelezova, 2020 ). In natural soil settings, the first investigative research on HGT among microorganisms was conducted in the 1970s ( Weinberg and Stotzky, 1972 ; Graham and Istock, 1978 ). Since then, further investigations have employed fieldwork and microcosm experiments to evaluate ecological HGT ( van Elsas et al., 2006 ). Plant-associated soils and biofilm communities are well-known hotspots for HGT due to the great genetic variety on such a restricted geographical level ( Fan et al., 2018 ). It was found that the genetic factors in HGT are more abundant in the rhizosphere than in bulk soil, therefore HGT may aid in the development of rhizosphere competence ( Vieira et al., 2020 ). Previous bipartite network analyses have provided evidence of genetic exchange, plasmid fusion and fission, exogenetic plasmid transfer, and environmental adaptation of the soil bacteria such as Rhizobium ( Corel et al., 2018 ; Li L. et al., 2020 ; Li X. et al., 2020 ). In addition, plasmid transmission from transplanted P. fluorescens to native gram-negative rhizobacteria in soil has been proven to happen in natural settings ( Balthazar et al., 2021 ). Many terrestrial subterranean microbes have also been found to contain MGEs, and some of them have already shown conjugative activity in a lab environment or even in bulk soil. Solitary ( Feng et al., 2007 ) or multiple ( Brockman et al., 1989 ) plasmids could be present in subterranean isolates, with big plasmids that are more apt to own recombinant functionality seeming to predominate ( Fredrickson et al., 1988 ). Microorganisms buried deeper underground have been shown to harbor extremely large (>150 kb) conjugative plasmids at a higher rate than microorganisms from shallower subsurface soils ( Ogunseitan et al., 1987 ; Fredrickson et al., 1988 ). Degradative genetic determinants are commonly found in subsurface large plasmids, which significantly increase the host’s metabolism adaptability ( Romine et al., 1999 ; Basta et al., 2004 ). Elements similar to recombinases are also present in several accessible plasmids, implying the possibility of incorporation and excision from the soil microbial genome ( Romine et al., 1999 ). Among the sequenced soil subsoil genomes, G. thermodenitrificans was discovered to match 11 and 3% of its genetic makeup with Bacillus sp. and other Firmicutes, respectively. Around 2.7% of the genes in G. thermodenitrificans NG80-2 exclusively occur in distant relatives and may have undergone HGT. This includes two groups of proteins associated with nitrogen use that seem to have originated separately. The addition of such genes and the catabolic plasmid pLW1071 has significantly improved the metabolic adaptability in nutrient-poor conditions ( Feng et al., 2007 ). Furthermore, the introduction of MGEs into synthesized microbial communities (SynComs) was considered an effective way for manipulating the microbial community for applicable purposes (e.g., carbon storage) ( Song et al., 2014 ; Tsoi et al., 2019 ; de Lorenzo, 2022 ). Yet, unlike simple bacterial mixes in a lab, the complexities of HGT mechanisms in wild areas, including the heavily heterogeneous populations, still need to be fully deciphered. In situ critical cellular events, additional chemical triggers for genetic exchange or absorption, and natural factors like cell-MGE interactions are examples of the complexities. In addition, there needs to be more focus on the environmental implications, such as how these HGT activities affect microbial communities’ capacity to adapt to wild areas, and how these HGT mechanisms reflect the features of the particular ecosystem. Though there is a growing interest in elucidating the underlying microbial mechanisms driving soil carbon transformation, stabilization, and release processes, many unknowns remain. Nonetheless, many earlier studies mainly concentrated on HGT episodes of particular lineages and on significant issues such as adaption-associated processes ( Li et al., 2019 ; Li L. et al., 2020 ; Li X. et al., 2020 ; Li et al., 2021 ). Regardless of these efforts, there is so much to be learned in a holistic mode about the quantity and effects of HGT-driven genetic makeup in soil-borne microbiota that can confer a variety of carbon assimilation and dissimilation functions (e.g., central carbon cycles, fatty acid biosynthesis). This poses significant challenges to understanding gene flow in terrestrial microbial communities. As a result, it remains to be seen whether there are inter-microbe HGT(s) that have promoted the development of soil-borne microbiota in a variety of conditions. If these HGT markers exist, how do they work to counteract environmental stressors, and how widely disseminated are these markers throughout the genetic material of soil-borne microbiota? A variety of natural systems, including protein expression ( Alon, 2007 ), biochemical processes ( Pál et al., 2005 ), biomolecule interplay ( Jeong et al., 2001 ), contradicting evolutionary indications ( Huson and Bryant, 2006 ), and ecosystem dynamics ( Rezende et al., 2007 ), are being modeled using network infrastructure ( Proulx et al., 2005 ). In theory, the network is capable of better displaying the patterns of microbial genomic evolution ( Doolittle and Bapteste, 2007 ). HGT networks are a distinctive form of sharing gene networks. They are intended to investigate trends in genetic dispersion brought on by HGT throughout ecological evolution. Under this context, we performed BLASTP-driven screenings and HGT network constructions to discover and investigate potential inter-microbe HGT genes linked to carbon metabolism throughout genomic sequences of all accessible strains isolated from soils, followed by network constructions. Our results suggested that the HGT episodes may significantly contribute to genetic imports that contribute to the soil-borne microbiota’s increased carbon metabolic versatility and adaptivity. Our knowledge of microbial interplay and the adaptable development of microbes to deal with various situations have therefore been enhanced as a result of this work. However, further experimental work is still necessary to evaluate the occurrence of HGT in the vast and unexplored environment of the terrestrial subsurface.",
"discussion": "3. Discussion An editorial in the journal Science has advocated that more thorough investigations be conducted into the appearance and mechanics of HGT episodes (“Why does lateral transfer occur in so many species and how?”) ( American Association for the Advancement of Science, 2005 ). We thoughtfully evaluated that the current study could have provided several suggestions for this subject. A variety of soil-borne carbon-cycle HGT events of cross-class and even cross-phylum levels were discovered with considerably high transfer frequency ( Figure 1 ). The constructed HGT networks demonstrate that genetic exchange across microbial genera is a significant contributor to microorganisms’ biodiversity and adaptivity. The characteristic of a “small-world system” found in previous HGT networks was also observed in our study, which refers to a network that has a small diameter in terms of the number of nodes and a handful of strongly linked nodes that allow for flux to flow across the system ( Proulx et al., 2005 ). A small-world architecture in the HGT network indicates that significantly advantageous genes that arise in any microbe could transcend taxonomic boundaries and stretch another microbe through a limited amount of HGT episodes, where genomes with high betweenness can act as a link between otherwise unconnected parts of the network and transfer genes to numerous other genomes in the ecosystem with a small number of genetic transactions. Similarly, a recent study also found that the HGT rate was increased in organisms with similar ecological distributions ( Zhou et al., 2021 ). Besides, the observed variations in network properties ( Table 1 ) may have multiple implications for HGT events, highlighting the importance of understanding network dynamics in studying HGT events ( Kunin et al., 2005 ; Kloesges et al., 2011 ; Popa et al., 2011 ; Li L. et al., 2020 ; Li X. et al., 2020 ). For example, a higher number of nodes and edges in the “acceptor” network may indicate a greater potential for gene transfer due to more opportunities for contact between bacteria. Similarly, higher degrees or connector components may create more interconnectedness amongst bacterial populations and facilitate the spread of genetic material. On the other hand, a longer average path length between nodes in a network may reduce the likelihood of gene transfer, as the distance between two bacteria would be higher and the chance of interaction would be lower. Moreover, differences in eigenvector centralities among network groups may illustrate the presence of influential nodes--nodes with high centrality values that may play significant roles in mediating gene transfer. Modularity can also impact the strength and density of community structures within a network, potentially creating barriers or pathways for gene transfer. The phylum Proteobacteria was the most abundant taxon in soil samples, accounting for an average of 30% of metagenomic sequences ( Guo et al., 2018 ). In keeping with this, our study found that nodes of classes, including Actinobacteria as well as Proteobacteria classes, such as Alphaproteobacteria and Gammaproteobacteria are extensively present in our overall HGT network of carbon metabolism ( Figure 1 ), implying that HGT is widespread during the evolution of species in these classes. However, the sampling density of sequenced microbial genomes might be biased toward Proteobacteria, since the predominance of Proteobacteria in the genome dataset might be responsible for the enormous rate of HGTs within this group. Previous studies have confirmed that Proteobacterial MGEs constitute the major connected component in the virulence network, and extensive gene sharing exists among Actinobacteria and Gammaproteobacteria ( Tamminen et al., 2012 ; Jiang et al., 2017 ). Previous studies have also validated experimentally the possibility of MGE-mediated HGT in soil samples, in which plasmids from the donor strains Psudomonas putida KT2440, Escherichia coli MG1655, and Kluyvera sp. can be transferred to a wide range of bacterial phyla from agricultural soils ( Klümper et al., 2015 ), including (α–ε) Proteobacteria, Acidobacteria. Actinobacteria, Bacteroidetes, Firmicutes, Fusobacter, Gemmatimonadetes, Planctomyces, spirochaetes, Candidate division TM7, Verrucomicrobia as well as Eukaryote ( Zhang et al., 2015 ). Externalized genes carried by MGEs could act as containers for the shared genetic pool ( Norman et al., 2009 ). In our study, uneven functional distributions of the externalized gene are also discovered, in which HTGs are mostly incorporated into the peripheral functions of the carbon metabolic pathway (e.g., nutrient transport and dispensable reaction). In contrast, the core metabolic sections (e.g., intermediate reactions and biomass production) of putative competent significance are mostly evolutionarily unvaried (see Supplementary Figures S1–S5 at https://doi.org/10.6084/m9.figshare.22154828.v1 ). Similar findings were also reported previously. For example, according to research on the horizontally gained genes within the E. coli metabolic pathways ( Pál et al., 2005 ), HGT is more common among proteins engaged in the absorption and consumption of resources than those responsible for the generation of biomass, which suggested that their function influences the HGT probability of metabolic genes in the internal metabolic pathway. This uneven functional distribution could also be explained by the “complexity hypothesis” ( Jain et al., 1999 ; Muller et al., 2018 ), which proposed that proteins in a complicated system, like ribosomal systems or core metabolic cycles, are specialized to work. Decreased adaptation of the microbial recipients will arise from an HGT incidence that leads to substituting such a gene with a less-suited counterpart. The proportional influence of functionality type and the number of interactive participants on HGT incidence was also investigated, demonstrating that the “complexity theory” still holds up in the genome-scale analyses ( Cohen et al., 2011 ). On the other hand, soil multi-functionality is affected by the environment and by microbial community composition and diversity ( Zheng et al., 2019 ). Environmental factors in the soil also impact the HGT processes and lead to uneven distribution. The microbial HGT rate in soil relies on environmental stress variables like surface temperatures, pH ( Rochelle et al., 1989 ), soil composition ( Richaume et al., 1992 ), and wetness ( Richaume et al., 1989 ). The gene-sharing network revealed strong correlations between gene connectivity and the trailed soil variables ( Zhou et al., 2010 ). Besides, the persistence and transportation of genetically modified bacteria were impacted by fluidity in subsoil ( Trevors et al., 1990 ). In terms of biological variables, the existence of fungi ( Sengeløv et al., 2000 ), protozoa ( Johannes Sørensen et al., 1999 ), and roundworms ( Daane et al., 1996 , 1997 ) could also impact sexual plasmid transmission in soil. Moreover, the generation of leachate and root elongation looked to be the main contributors to the incidence of HGT inside the rhizosphere ( Mølbak et al., 2007 ). Transmission rates were roughly 10 times lower in the bean and cereal rhizospheres than in the control soils ( Musovic et al., 2006 ). Lastly, the other key factor contributing to the rise in local MGE quantities and HGT incidence in this environment is the implementation of manure, pesticides, and antimicrobials to the land ( Gotz and Smalla, 1997 ). Correspondently, various organic degradative genes were found mobilized among soil-borne microbiota via HGT in our study ( Figure 4D ), whose catabolic activities might be further applied for bio-remediation of polluted environments ( Top et al., 2002 ; Nojiri et al., 2004 ). It is recently confirmed that genes of microalgal origin have conferred Caenorhabditis elegans the ability to degrade cyanogenic toxins ( Wang et al., 2022 ). Previous studies also found that functional categories “biosynthesis and degradation of surface polysaccharides and lipopolysaccharides” and “DNA regulation and modification” tend to be enriched in the HGT entries. In consistence, we found glycosyltransferase in our study as the most abundant protein encoded by HTG related to carbon metabolism. The glycosyltransferase family was reported to be involved in the glycosylation and modifications of biomolecules, including the bases in DNA, which might alter the host’s gene expression pattern ( Iyer et al., 2013 ). Also, it was reported that glycosyltransferases were significantly enriched in horizontally transferred genes in the human gut, while soil microbiota has similar expansions of the glycosyltransferase repertoires as the gut ( Lozupone et al., 2008 ). Another generally enriched functional category in the HGT profile is metabolite transporter ( Cordero and Hogeweg, 2009 ; Paquola et al., 2018 ), which is also reflected by our results ( Figure 4F ). In conclusion, the inter-microbe HGT genetic traits in soil-borne microbiota genetic sequences that we recognized through our assessments, as well as their involvement in carbon metabolism and resilience to various environmental stressors typically found in territory ecosystems, suggest a pervasive and substantial effect of HGT on the evolvement of microbes. Nevertheless, the information we have provided here is not thorough. The examples of inter-microbe HGT documented so far represent just the tip of a giant biological iceberg. Upcoming studies are expected to offer a more intriguing view of both the scope and the biological importance of HGT in microbes as increasing numbers of genomic data of higher quality are becoming accessible for yet more branches in the tree of life for microbes. Further work will be necessary to evaluate the occurrence of HGT in the vast, heterogeneous, and isolated environment of the terrestrial subsurface and to assess the full impact of gene transfer on terrestrial subsurface microbial evolution."
} | 5,994 |
23028797 | PMC3445580 | pmc | 8,277 | {
"abstract": "In the recent discussion how biotic systems may react to ocean acidification caused by the rapid rise in carbon dioxide partial pressure ( p CO 2 ) in the marine realm, substantial research is devoted to calcifiers such as stony corals. The antagonistic process – biologically induced carbonate dissolution via bioerosion – has largely been neglected. Unlike skeletal growth, we expect bioerosion by chemical means to be facilitated in a high-CO 2 world. This study focuses on one of the most detrimental bioeroders, the sponge Cliona orientalis , which attacks and kills live corals on Australia’s Great Barrier Reef. Experimental exposure to lowered and elevated levels of p CO 2 confirms a significant enforcement of the sponges’ bioerosion capacity with increasing p CO 2 under more acidic conditions. Considering the substantial contribution of sponges to carbonate bioerosion, this finding implies that tropical reef ecosystems are facing the combined effects of weakened coral calcification and accelerated bioerosion, resulting in critical pressure on the dynamic balance between biogenic carbonate build-up and degradation.",
"introduction": "Introduction Since the turn of the millennium, ocean acidification (OA) has been recognized as a key factor in marine ecology, attracting a growing pool of research which identified OA to have a multitude of mainly negative effects on reproduction, growth, survival, and diversity of marine biota [1] – [3] . Among the best studied victims in this respect are organisms that produce carbonate skeletons, and particularly scleractinian corals that show significantly reduced skeletal growth rates with declining pH and lowered seawater carbonate saturation state [4] – [8] . In contrast, bioeroding organisms have largely been ignored, although they play a key role in carbonate cycling by abrading or dissolving materials such as coral skeletons, and thus need to be included in any equation concerning reef health or growth. This omission needs to be addressed, because chemically achieved bioerosion is expected to be facilitated with progressing OA [9] , [10] , potentially placing many bioeroders into the circle of “winners” of global climate change [11] . Marine bioerosion acts at different scales and is performed by a multitude of organisms employing different chemical and mechanical means in the process of attachment, grazing, or carbonate penetration [12] . On coral reefs, the largest proportion of internal bioerosion is often contributed by demosponges, which do not add to calcification as they have siliceous spicules, but frequently represent 60 to over 90% of total macroborer activity [13] , [14] . Single sponge species commonly remove around 10 and in extreme cases more than 20 kg m −2 yr −1 \n [15] , thereby balancing or even surpassing reef calcification rates at some sites [14] , [16] – [17] . In warm waters worldwide, the photosymbiotic clionaids of the so-called ‘ Cliona viridis species complex’ lead this process in terms of abundance, colony size, growth, and erosion rates [15] , [18] . Their symbiosis with dinoflagellate zooxanthellae appears to increase their competitive powers, and it is comparatively stress resistant [19] , [20] . ‘ C. viridis species’ routinely invade and kill live corals and have been reported to survive and increase in abundance where environmental conditions deteriorate [21] . Our model organism Cliona orientalis Thiele, 1900 belongs to this species complex and is one of the most competitive and abundant representatives of these bioeroders. It is widely distributed on Australia’s Great Barrier Reef (GBR), Indonesia and Japan [18] , [22] ( Fig. 1 ). 10.1371/journal.pone.0045124.g001 Figure 1 The zooxanthellate sponge Cliona orientalis at Orpheus Island, Great Barrier Reef, Australia. (A) Location of Orpheus Island (Palm Island Group) on the central GBR. (B) Medium-sized colony infesting the massive coral Porites sp. at the reef crest in Little Pioneer Bay, Orpheus Island. (C) Detail illustrating the oscula (exhalant pores; inhalant pores are microscopically small) and the Porites skeletal structure visible beneath the sponge tissue. (D) One of the eight replicate sets per treatment tank with 4 healed sponge-bearing coral cores. Sponges erode at cellular level by means of biochemical dissolution that leads to the formation of minute cup-shaped grooves and the mechanical extraction of so-called sponge chips of a diameter between 10 and 100 µm [22] . In order to dissolve carbonate, the sponge lowers the pH at the tissue-substrate interface where the specialised etching cells act [22] (exact etching agent unknown to date). Sponge bioerosion is conducted extracellulary potentially making the process sensitive to environmental conditions and change. The lower the environmental pH is to begin with, the less pronounced is the gradient between ambient seawater and the site of dissolution, and the lower will be the metabolic cost required for bioerosion. Hence we hypothesise that the pH lowering inherent to OA will increase the efficiency of the bioerosion process, leading to a significant increase of sponge bioerosion rates with increasing p CO 2 .",
"discussion": "Results and Discussion Before testing a possible pH dependency of sponge bioerosion, we assessed biologically-driven daily pH fluctuations in the treatment tanks as evidenced during a 24 h series of measurements logged both with and without sponge replicates in place ( Fig. 4 ). Despite the flow rate of ∼30 l/h, a pH oscillation of 0.07 points was determined at present-day p CO 2 when sponges were in the tank. The rise in pH coincided with the beginning of the 12 h irradiance period (simulated daylight), and values declined again after lights were turned off. This signal reflects the uptake of CO 2 (and linked rise of pH) during active photosynthesis of the symbionts in the sponge tissue. This flux was higher than the simultaneous generation of CO 2 from the sponge respiration, resulting in net photosynthesis during daytime. In contrast, during the following dark phase only respiration was taking place, both by the sponge and its photosymbionts, and led to a decrease in pH. In comparison, the photosynthetic activity of phytoplankton and some early algal turfs in the treatment tanks amounted to a change of only 0.01 pH points. The temperature in all treatment tanks also followed a synchronous light-dependent rhythm due to the warming by the lamps, thereby simulating daily temperature fluctuation. The carbonate saturation states for aragonite and calcite never became undersaturated (Ω <1), neither in the highest experimental p CO 2 nor when considering the diurnal pH and temperature fluctuations. Nevertheless, a relevant abiotic dissolution or microbial bioerosion of the coral substrate was ruled out by including clean dead coral cores of similar size and from the same source as controls – none of these lost weight, despite the larger exposed surface in the sponge-free cores ( Table 1 ). Estimates of proportional biomass [25] additionally indicate that microbial bioerosion by phototrophic or chemotrophic euendoliths in the sponge cores is negligible. We furthermore checked mean penetration depth and sponge tissue weight per sponge-bearing core after the experiment ( Table 1 ), which did not vary significantly between treatments and confirmed that our data were not biased by sponge biomass or tissue shrinkage, so that the change in buoyant weight recorded in our experiment can be addressed with confidence to the chemical and mechanical bioerosion activity of Cliona orientalis . 10.1371/journal.pone.0045124.g004 Figure 4 Diurnal pH and temperature oscillations. Biologically induced pH fluctuation (increase during photosynthesis; decrease as result of respiration) in the present-day p CO 2 treatment tank (393 µatm) with (diamonds) and without sponges (circles), showing the causal relationship with the illumination phase (top); temperature fluctuation in the same tank affected by heat radiated off the metal halide lamps (bottom, triangles). Sponge bioerosion rates reached a mean of 2.23±0.15 kg m −2 yr −1 in the present-day treatment ( Table 1 ). Bioerosion rates significantly increased with rising p CO 2 ( Fig. 5A , Table 2 ). At moderately elevated p CO 2 the mean bioerosion rate was 2.60±0.25 kg m −2 yr −1 , which corresponds to a 17% increase relative to the present-day rate. At strongly elevated p CO 2 , bioerosion was further enhanced, attaining a mean rate of 3.59±0.40 kg m −2 yr −1 and representing a 61% change compared to the present-day value. This increase in bioerosion rate reflects the enhanced efficiency of the sponges’ bioerosion process as a result of the lowered environmental pH, causing a shallower gradient between the environment and the etching site. The sponge apparently ‘takes advantage’ of the facilitated dissolution in the more acidic environment, as opposed to keeping bioerosion rates constant and only lowering the metabolic cost. In contrast to this distinct trend, a decrease in bioerosion rate of 2.22±0.45 kg m −2 yr −1 in the slightly lowered p CO 2 level was less than 1% lower and thus not significantly different compared to the present-day treatment ( Table 2 ). The physiological interpretation for our findings in the lowered p CO 2 is that the sponge is partly able to compensate for the less favourable conditions (hindered dissolution in more alkaline conditions), possibly at the cost of increasing the metabolic rate. The overall linear regression of bioerosion rate versus p CO 2 is highly significant (r 2 = 0.76; p <0.0001) and clearly supports the initial hypothesis that sponge bioerosion can be expected to accelerate with progressing OA. Based on the linear regression, the relationship between p CO 2 [µatm] and C. orientalis bioerosion rates [kg m −2 yr −1 ] can be formulated as in Eq. 1, and the respective relationship converted to changes in pH in Eq. 2. (1) \n (2) \n Keeping the limitation in ecological relevance inherent to short-term lab experiments in mind, this relationship translates to a predicted 25.4% increase in sponge bioerosion by the end of this century, following the BERN-CC reference model based on the SRES A2 emission scenario that corresponds to a predicted p CO 2 level of 836 µatm by the year 2100 [26] . The most optimistic SRES B1 model with a predicted 2100 p CO 2 of only 540 µatm would result in an 8.6% increase and the intermediate SRES A1B model with a 2100 p CO 2 of 703 µatm equates to a potential 17.7% increase in sponge bioerosion ( Fig. 5B ). A similar range of predictions can be made when applying the new Representative Concentration Pathways (RCPs) [27] with an 8.6%, 15.8%, and 30.9% increase for the RCP 4.5, 6, and 8.5 scenarios, respectively. 10.1371/journal.pone.0045124.g005 Figure 5 Increasing sponge bioerosion as a function of increasing p CO 2 . (A) Weight loss per replicate set translated to bioerosion rates for the four p CO 2 treatments. The linear regression of the 32 replicates (8 per treatment) is highly significant (r 2 = 0.76; p <0.0001). (B) Projected percent increase in sponge bioerosion relative to the present-day level, calculated for the BERN-CC model based on the SRES A2 (red), A1B (blue), and B1 (green) emission scenarios. 10.1371/journal.pone.0045124.t002 Table 2 Results ( p values) from the pairwise comparison of bioerosion rates in the four p CO 2 treatments (Mann-Whitney test; Bonferroni corrected p values in lower left triangle of matrix) performed after Kruskal-Wallis analysis (H = 21.25; Hc = 21.25; p <0.0001; n = 8 per treatment) and rejection of normal distribution for the present-day (393 µatm) and the elevated treatment (571 µatm) via Shapiro-Wilk test. below present p CO 2 339 µatm present-day p CO 2 393 µatm elevated p CO 2 571 µatm strongly elevated p CO 2 1410 µatm below present p CO 2 339 µatm – 0.7132 0.0831 0.0009 * \n present-day p CO 2 393 µatm 1.0000 – 0.0063 * \n 0.0009 * \n elevated p CO 2 571 µatm 0.4987 0.0379 * \n – 0.0014 * \n strongly elevated p CO 2 1410 µatm 0.0056 * \n 0.0056 * \n 0.0082 * \n – * significant difference. Due to the important role of bioeroding sponges, and of ‘ C. viridis complex’ species in particular, this finding suggests severe consequences for coral reef health. Coral reef calcification and bioerosion are antagonistic processes in a dynamic balance [28] , [29] . This balance will become seriously strained when bioerosion is accelerated by OA, while at the same time, coral net calcification rates are declining [4] – [8] . This situation will push the carbonate budget towards negative values, and on some reefs negative carbonate budgets have already been recognised as result of intensive sponge bioerosion [16] , [17] . Pioneer experimental evidence for an increase of bioerosion rates generated by specific bioeroders due to seawater acidification was provided for euendolithic microborers. Biosphere 2 experiments showed that particularly the dominant microboring chlorophyte Ostreobium quekettii grows faster under elevated p CO 2 (750 µatm) [10] . However, in contrast to endolithic algae which occasionally even support stressed calcifiers [30] , bioeroding sponges are always in antagonism to calcifiers, and the species we worked with is known to often overwhelm and kill live corals [18] , [21] . Several mesocosm experiments and field studies demonstrated an increase of total dissolution – including bioerosion, but rarely addressed as such – partly leading to a net loss of carbonate [31] . Coral reefs in the eastern tropical Pacific, where cool, CO 2 -rich upwelling water masses lead to naturally low pH and saturation states, are poorly cemented and prone to intense bioerosion [17] , serving as a model for coral reef development in a high-CO 2 world [9] . Lowest mean pH (seawater scale) values of 7.88 and a corresponding Ω Ar of 2.49 were reported from Galápagos [9] . Hence at least part of pH conditions predicted by OA scenarios is already experienced by tropical reef corals and bioeroding sponges at present time. This may apply not only for the eastern Pacific but for many shallow coral reefs with relative long water residence times, as a result of carbon fluxes related to calcification and the remineralisation of organic matter [32] . At the GBR for instance, spatial and temporal pH fluctuations are in the magnitude of 0.4 pH units [33] . Hence, the three lower p CO 2 treatments of the present experiment were within the range of natural fluctuations currently experienced on some coral reefs, whereas the strongly elevated treatment looks far into the future and may never be reached. Intriguingly, another important factor in climate change – global warming – may partly counteract the development caused by OA. Rising temperatures reduce the physicochemical dissolution capacity of calcium carbonate in seawater [34] and could also slow down chemical bioerosion. However, within tolerance limits of physiological processes, i.e. chemical reactions can be accelerated by elevated temperature, and interaction of p CO 2 and temperature may have complex effects as has been demonstrated with respect to coral calcification rates [35] , [36] . This observation calls for multifactorial experiments that consider both, the isolated as well as concerted effects of p CO 2 and temperature on sponge bioerosion and other bioerosion processes. And, as an indispensable step, the impact of climate change on bioerosion needs to be addressed in long-term in-situ experiments. Ultimately, these data will convey critical insights into global trends of biologically caused decalcification and the possible threat of increasing bioerosion on the balance between skeletal growth and bioerosion on tropical coral reefs."
} | 3,977 |
39307304 | PMC11530827 | pmc | 8,282 | {
"abstract": "Cytochromes P450 (P450s) are a superfamily of heme-containing enzymes possessing a broad range of monooxygenase activities. One such activity is O -demethylation, an essential and rate-determining step in emerging strategies to valorize lignin that employ carbon–carbon bond cleavage. We recently identified PbdA, a P450 from Rhodococcus jostii RHA1, and PbdB, its cognate reductase, which catalyze the O -demethylation of para -methoxylated benzoates ( p -MBAs) to initiate growth of RHA1 on these compounds. PbdA had the highest affinity ( K d = 3.8 ± 0.6 μM) and apparent specificity ( k cat / K M = 20,000 ± 3000 M -1 s -1 ) for p- MBA. The enzyme also O -demethylated two related lignin-derived aromatic compounds with remarkable efficiency: veratrate and isovanillate. PbdA also catalyzed the hydroxylation and dehydrogenation of p- ethylbenzoate even though RHA1 did not grow on this compound. Atomic-resolution structures of PbdA in complex with p -MBA, p- ethylbenzoate, and veratrate revealed a cluster of three residues that form hydrogen bonds with the substrates’ carboxylate: Ser87, Ser237, and Arg84. Substitution of these residues resulted in lower affinity and O -demethylation activity on p -MBA as well as increased affinity for the acetyl analog, p- methoxyacetophenone. The S87A and S237A variants of PbdA also catalyzed the O -demethylation of an aldehyde analog of p -MBA, p -methoxy-benzaldehyde, while the R84M variant did not, despite binding this compound with high affinity. These results suggest that Ser87, Ser237, and Arg84 are not only important determinants of specificity but also help to orientate that substrate correctly in the active site. This study facilitates the design of biocatalysts for lignin valorization.",
"discussion": "Discussion In this study, we demonstrated that the P450-reductase pair, PbdAB, catalyzes the efficient O -demethylation of three related LDACs: p -MBA, veratrate, and isovanillate. This is consistent with our previous findings that PbdAB is essential in the catabolism of these three compounds by RHA1 ( 16 ). The high affinity and specificity of PbdA for p -MBA is consistent with what has been reported for two members of CYP199A subfamily, A2, and A4, ( 18 ). However, PbdA differs from CYP199A2 and CYP199A4 with respect to its relative affinity and activity for veratrate. Thus, CYP199A2 and CYP199A4 bound p -MBA with three orders of magnitude higher affinity than veratrate. In contrast, PbdA bound p -MBA with approximately only five times higher affinity than veratrate. Moreover, CYP199A2 did not detectibly turn over veratrate, while PbdA did so at half the rate of p -MBA. This was somewhat unexpected given that the active site residues of CYP199A2/4 and PbdA are remarkably conserved. However, Ala174 (PbdA numbering) is a notable exception, corresponding to valine in CYP199A2 and CYP199A4 ( Fig. S7 ). The increased side chain size could cause steric clash between the 3-methoxy group and the side chain ( Fig. S8 ). Mutagenesis experiments could provide valuable insight into the effect of this residue. As veratrate is a major monoaromatic component of methylated lignin streams ( 10 ), high activity on veratrate would be advantageous for lignin-valorizing biocatalysts. PbdAB also efficiently catalyzed the hydroxylation and desaturation of p -EB, as reported in CYP199A4 ( 21 ). Nevertheless, RHA1 did not grow on p -EB ( Fig. S6 ). Indeed, a naturally occurring catabolic pathway for p -EB has yet to be reported. However, derivatives of Pseudomonas sp. B13 were evolved to utilize p -EB as a sole substrate via meta -cleavage, overcoming both regulation and enzyme inhibition, both of which are bottlenecks in its degradation by the WT strain ( 26 ). It is therefore likely that the high affinity and specificity of PbdA for p -EB reflect its chemical similarity to p -MBA and this is not a physiologically relevant reaction. Structural studies of PbdA, coupled with substitution of active site residues, identified Arg84, Ser87, and Ser237 as specificity determinants through their respective interactions with the substrate carboxylate. This binding motif is conserved in the CYP199A family. Substitution of either Arg84 or Ser87 disrupted binding and activity significantly more than substitution of Ser237 ( Table 3 ). This may reflect the fact that the former are coplanar while Ser237 is on a distal helix. Substitution of both serines in the double variant was most deleterious for binding and turnover of p- MBA. Indeed, S87A/S237A had the least negative binding energy, ΔG°, of any tested mutant ( Table S6 ). Moreover, the ΔΔG° of the double variant versus WT PbdA was partially additive with respect to the two single variants, indicating additional stabilization mechanisms which compensate for the loss of hydrogen bonding to the carboxylate. Substitution of the Arg84 had a lower effect on binding energy than the substitution of both serines, possibly due to compensation by hydrogen bonding from the ordered water coordinated to Arg 236 ( Fig. 2 A ). The interaction of PbdA with substrate analogs further reveals the importance of interactions with the carboxylate in productively orientating the substrate for efficient O -demethylation. Thus, PbdA did not detectably O -demethylate p -MAP or p -MBAL despite binding both compounds with high affinity and consuming NADH in their presence. This is consistent with neither compound effectively displacing the heme iron-coordinated water molecule, as indicated by the spin state transition ( Table 3 ). Although CYP199A4 catalyzed the O -demethylation of p -MAP and p -MBAL, it was at 0.25 and 0.1% the rate, respectively, of p -MBA ( 27 ). Interestingly, the S87A and S237A variants O -demethylated p -MBAL, although the arginine variant did not, despite having a relatively high affinity for this compound and p -MBAL inducing a relatively high degree of spin state transition. In the double variant, the O -demethylation of p -MBAL was remarkably well-coupled to NADH consumption ( Table S5 ). None of the variants detectably turned over p -MAP, although this entire compound induced NADH consumption in all of them. This lack of turnover suggests that proper substrate orientation mediated by hydrogen bonding to the carboxylate is necessary for catalysis. Structural studies of PbdA and select variants in complex with substrate analogs might elucidate the underlying causes for the lack of catalysis. This comprehensive structural and functional characterization of the P450 O -demethylase PbdA highlights its potential as a biocatalyst for the transformation of lignin streams. More specifically, advances in lignin processing are generating mixtures of novel methylated LDACs. Understanding the binding and turnover of methylated LDACs by PbdA facilitates the latter’s engineering toward biocatalytically upgrading such lignin streams."
} | 1,733 |
39412998 | PMC11530905 | pmc | 8,283 | {
"abstract": "Summary Here, we present a protocol for identifying the components of a quorum sensing signaling system in bacteria. We describe the steps for characterizing the novel response regulator and receptor of the cis -2-dodecenoic acid (BDSF) quorum sensing signaling system in Burkholderia cenocepacia . The technical assays of this protocol include generating a random mutant library, chromatin immunoprecipitation sequencing (ChIP-seq), electrophoretic mobility transfer assay (EMSA), microscale thermophoresis (MST), and molecular simulation docking. For complete details on the use and execution of this protocol, please refer to Li et al. 1 and Yang et al. 2"
} | 165 |
32376967 | PMC7203126 | pmc | 8,284 | {
"abstract": "We present a Penicillium rubens strain with an industrial background in which the four highly expressed biosynthetic gene clusters (BGC) required to produce penicillin, roquefortine, chrysogine and fungisporin were removed. This resulted in a minimal secondary metabolite background. Amino acid pools under steady-state growth conditions showed reduced levels of methionine and increased intracellular aromatic amino acids. Expression profiling of remaining BGC core genes and untargeted mass spectrometry did not identify products from uncharacterized BGCs. This platform strain was repurposed for expression of the recently identified polyketide calbistrin gene cluster and achieved high yields of decumbenone A, B and C. The penicillin BGC could be restored through in vivo assembly with eight DNA segments with short overlaps. Our study paves the way for fast combinatorial assembly and expression of biosynthetic pathways in a fungal strain with low endogenous secondary metabolite burden.",
"conclusion": "Conclusions In this study, the consecutive deletion of well-expressed BGCs led to a secondary metabolite deficient strain of P. rubens that is suitable for integration and in vivo assembly of heterologous BGCs. By the use of in vivo homologous recombination employing multiple DNA fragments, a complete BGC can be reassembled while introducing at the same time promotors to enhance the expression. This methodology speeds up fungal synthetic biology leading to more freedom in the design-build-analyze-cycle. A major advantage of the platform strain is that novel heterologous compounds can be purified with reduced interference from endogenous secondary metabolites. We demonstrated this approach by heterologous expression of the calbistrin BGC from Penicillium decumbens , obtaining the melanization-inhibiting decumbenones as final products. During the construction of the platform strain, intermediate strains were obtained with a different set of highly expressed NRPS genes. Metabolic profiling revealed an interesting interplay between the various NRPS enzymes. Because the substrate requirements of the enzymes PcbAB, ChyA, RoqA and HcpA show a certain degree of overlap, they likely compete for substrates and thus deletion of one BGC can result in higher levels of metabolites produced by the other BGCs. For instance, chrysogine biosynthesis seems to prevent accumulation of both roquefortine and fungisporine-related molecules by acting as a sink for anthranillic acid, the precursor for tryptophan biosynthesis. Since the chrysogine BGC is expressed under conditions required for penicillin biosynthesis, a deletion of the chrysogine BGC in a Pen-BGC strain might lead to a decreased penicillin production and shift the metabolic fingerprint to roquefortine-related molecules. The low expression of the remaining SM core genes in the 4xKO strain created here contributes to the clean secondary metabolite profile as indicated by the LC-MS data and adds to the applicability of this strain as a platform for secondary metabolism. Our study also suggests that the demand for cysteine due to penicillin production naturally increases sulphate uptake via SutA/SutB and thus ensures higher methionine levels that decrease the autophagy response. The majority of amino acids did not display severe intracellular changes under the conditions utilized here, except methionine and nicotinic acid, the degradation product of tryptophan, suggesting that CSI of P. rubens towards an increased yield of ß-lactams did not result in a major impact on the cells ability to regulate amino acid metabolism by either reducing synthesis (methionine) or increasing degradation (tryptophan) of excess amino acids. Except for an extracellular increase of valine, other amino acid levels did not change drastically, hence the metabolism remains sufficiently flexible after CSI to respond to a decreased demand of certain amino acids. This will make the strain characterized here suitable for expression of both PKS- and NRPS-containing BGCs.",
"introduction": "Introduction The fungal kingdom contains a massive reservoir of biosynthetic gene clusters (BGCs) encoding secondary metabolites, offering the discovery of potential new natural lead compounds for numerous applications like pharmaceutical drugs or food ingredients. However, successful follow-up of fungal genome mining studies 1 , 2 for novel secondary metabolites is often hampered by difficulties in handling promising fungi and identifying the specific conditions required for BGC activation 3 – 5 . To overcome these difficulties inherent to natural producers, a common approach is the transfer of promising BGCs into a more tractable organism. Although reported to be successful in prokaryotes like actinomycetes 6 , 7 and eukaryotes such as baker’s yeast 8 – 10 and filamentous fungi 11 , 12 , yields of heterologous secondary metabolites can vary substantially from the natural producer. Moreover, the native BGC-related product fingerprint of the host organism is mainly left unchanged, which drains cellular resources into unwanted BGC products, complicates target compound purification and can even lead to unspecific enzymatic conversion of the target compound if the specificity of native BGC enzymes is low. Thus, we envisioned that a tractable fungal host with a low background of endogenous secondary metabolites simplifies detection of novel molecules in broth samples as well as downstream purification. Penicillium rubens (former name: Penicillium chrysogenum ) is an industrially relevant fungal cell factory primarily used for production of ß-lactam-derived antibiotics. With the growing interest in fungal natural product discovery, precise genetic engineering of filamentous fungi gaining momentum 13 and synthetic biology tools being frequently utilized to recode BGCs for heterologous hosts 14 , 15 , we reasoned that the specialization of P. rubens into a penicillin cell factory might be favorable for further advances towards a platform strain for expression of any novel BGC 16 . Similar approaches have been undertaken in Aspergillus nidulans 17 where a strain with low background expression of endogenous BGCs was used for heterologous expression of BGCs randomly present on a large plasmid 11 . Several decades of classical strain improvement (CSI) have led to accumulation of point mutations 18 that resulted in strains optimized for high ß-lactam yield in large scale fermenters 19 and low unwanted secondary metabolite production. The superior fermentation characteristics of such strains were successfully employed for the production of cephalosporins 20 and, after deletion of the penicillin BGC, also for the heterologous polyketide pravastatin 21 . P. rubens research has led to a full genome sequence 22 and a metabolic model 23 which makes it attractive for future rational strain improvements. In addition, the efficiency of integrating multiple DNA fragments into P. rubens has been increased by utilizing split-marker approaches 24 and the targetable nuclease Cas9 25 . However, direct in vivo recombination for fast construction of different BGC pathway combinations has not yet been demonstrated. Moreover, since the precursors for the biosynthesis of penicillins, α-aminoadipic acid, L-cysteine and L-valine originate from diverse anabolic routes, a careful elucidation of intracellular amino acid pools would be required to assess the impact of CSI on the flexibility of the metabolism to respond to high and low amino acid demands. Here, we report on the construction of a P. rubens strain lacking four highly expressed secondary metabolite BGCs resulting in a near to complete secondary metabolite deficient metabolome under the cultivation conditions tested here. We performed genomic and transcriptome analysis, characterized its amino acid profile and demonstrated its suitability for efficient BGC recombination by reconstructing the Penicillin BGC (Pen-BGC) of 17 kb by in vivo recombination with 8 DNA fragment with short (110 bp) overlapping flanks. Finally, we utilized this new platform strain for expression of the heterologous calbistrin BGC from P. decumbens , yielding decumbenone A, B and C in the culture supernantant. This study paves the way to utilize P. rubens for exploration and production of novel BGCs.",
"discussion": "Results and Discussion 4 NRPSs display robust expression during low growth rates To prioritize secondary metabolite BGCs from the industrial strain P. rubens DS68530 (derived from Penicillium rubens Wisconsin 54–1255 via CSI and targeted gene deletion, see Fig. 2b ), for deletion, expression levels of 49 annotated 22 BGC core enzymes (nonribosomal peptide synthetases (NRPS), polyketide synthases (PKS) and hybrid enzymes thereof) were extracted from 22 publicly and 4 in-house available transcriptome data sets (Supplementary Information SI1 ) and grouped into high and low penicillin production conditions (Supplementary Information SI2 ). Penicillin yield was primarily affected by strain lineage and supplementation with side-chain precursors for production of Penicillin V and G. The growth rate was controlled by carbon limitation to alleviate the glucose repression of the Pen-BGC -mediated by CreA 26 . 17 out of 49 (35%) core BGC enzymes were not expressed in at least 22 of the 26 conditions covered by the transcriptome data and were therefore considered silent. However, besides the well-expressed ACV-tripeptide forming NRPS pcbAB ( Pc21g21390 ), genes coding for three further NRPSs showed expression across multiple strain backgrounds under glucose-limited growth rates at 0.05 h −1 and below. As displayed in (Fig. 1 ), these are Pc21g12630 - chyA , located in the chrysogine BGC 27 , 28 , Pc21g15480 - roqA in the roquefortine cluster 29 and the NRPS producing fungisporin 30 ( Pc16g04690 – hcpA ). Four other BGC core genes (Supplementary Information SI1 ) showed fluctuating and lower expression under some conditions analyzed here, including Pc21g05070 ( sorA ) and Pc21g05080 ( sorB ), coding for 2 PKSs required for sorbicillin biosynthesis 31 . However, the mutation L146F in the ketosynthase domain of SorA of P. rubens DS68530 was shown to result in a non-functional enzyme 32 . Similar to this observation, increased expression of Pc16g11480 (encoding a PKS termed PKS7) was observed when strains were grown in shake flasks. PKS7 carries a mutation (A952D) in a putative linker region between domains, which could potentially affect enzyme functionality. Pc18g00380, coding for a NRPS-like gene (NRPS-like7) was constantly expressed at 10 to 15% of the actin expression level without having acquired any mutations during CSI, and no known metabolite has been associated to this core enzyme of a BGC. The remaining BGC core genes showed only very little expression and might thus not contribute significantly to the metabolite fingerprint of P. rubens under the tested cultivation conditions. Figure 1 Genomic structure of four BGCs displaying strong expression and relative expression of the core gene under three different conditions. ( A) Schematic organization of the BGCs identified as being strongly expressed. BGC core genes are shown in red, closest genes not part of the cluster are colored black. All loci are drawn to scale and arrow directions denote orientation of transcription. ( B) Relative expression of identified BGC core genes from 26 transcriptome analyses with strong expression. Experiments were grouped according to carbon limitation and penicillin yield. n/a not applicable (because no quantities reported). Overall, these data suggest that the identified BGCs have a stronger impact on the hosts secondary metabolite fingerprint due to increased product formation when the penicillin BGC is lower expressed or completely deleted. Indeed, the encoded NRPSs belong to BGCs which were characterized previously as abundant products in the culture broth 29 , 30 , 33 of strains lacking the penicillin BGC. Therefore, the chrysogine and roquefortine BGCs along with the fungisporin-producing NRPS hcpA were prioritized for deletion. Construction and genomic analysis of a P. rubens strain devoid of four BGCs A recently developed methodology of Cas9-aided transformation 25 was used for sequential and complete deletion of the prioritized BGCs and the intermediate strains 2xKO ( ∆hdfA, ∆pen-BGC, ∆chy-BGC ), 3xKO-A ( ∆hdfA, ∆pen-BGC, ∆chy-BGC, ∆roq-BGC::amdS ), 3xKO-B ( ∆hdfA, ∆pen-BGC, ∆chy-BGC, ∆hcpA::amdS ) and 4xKO ( ∆hdfA, ∆pen-BGC, ∆chy-BGC, ∆roq-BGC::ergA, ∆hcpA::ble ) were obtained (Fig. 2 and Table 1 ). Due to repeated transformations and treatment with Cas9 RNPs, the genome of P. rubens 4xKO was analyzed for mutations not previously reported in this strain lineage 18 , using DS68530 as the reference strain. We observed 46 mutations in 4xKO, of which 18 mutations were located in 12 genes and the remaining mutations were intergenic. Remarkably, 13 of the 46 identified mutations were also present in DS68530 at frequencies below 50% in the population of spores used to generate material for gDNA and sequencing, suggesting that these mutations were then further enriched during clone selection and strain construction steps we conducted. The 18 genic mutations were classified as frameshift (1), intronic (1), non-synonymous SNP (12) or synonymous SNPs (4). Mutated genes encode for low-expressed hypothetical proteins where no clear biological role was immediately conclusive (Supplementary Information SI3 ) and did not contain a sgRNA off-target-site. Additionally, we also Sanger-verified the putative off-target sites with less than 4 mismatches and no bulges on the RNA or DNA of the five sgRNAs used during the transformation by PCR amplification and amplicon sequencing. None of these off-target sites did contain a mutation (Supplementary Information SI4 ), hence the applied Cas9-RNP method did not lead to off-target mutations in our hands, but the process of repeated transformations and selection caused a few novel mutations. This is not surprising as removal of the 8 pen-BGC copies from DS17690 also resulted in an accumulation of 18 previously not observed genic SNPs in the derived strain DS68530 34 . Moreover, it was also reported for Aspergillus fumigatus 35 that spontaneous mutations can occur in the absence of Cas9. Also, no larger deletions or insertions were identified among the SNPs which were observed in Aspergillus niger capable of performing NHEJ 36 . The P. rubens strains utilized here are devoid of hdfA / ku70 and thus, homology-directed repair (HDR) will be the dominant mechanism for DNA damage, therefore additionally limiting the possibility of mutations. Taken together, the obtained 4xKO strain does carry very few new mutations compared to the parental strain DS54468 which are not caused by repeated treatment with Cas9 RNPs, stressing that our applied methodology is efficient and reliable. Figure 2 Effect of BGC deletion on secondary metabolites and amino acid levels. ( A ) Overview of core BGC genes, first products in pathways and number of reactions steps leading to the final products 6–6-aminopenicillanic acid (6-APA), chrysogine, meleagrin and fungisporine. ( B) Strain lineage of P. rubens , including strains utilized in this study. Penicillin yields are denoted by superscript (+)-symbols ranging from ( + ) – low, ( ++ ) – intermediate to ( +++ )- high as far as reported. Figure adapted with modifications from 32 . ( C) Total-ion-chromatograms of DS68530 and 4xKO, taken after five days of growth in SMP. ( D) Changes in peak area of selected secondary metabolites associated with removed BGCs after five days of growth in SMP medium (n = 3). A time-course series of all secondary metabolites can be found in supplementary Information SI4 . ( E) Summary of changes in intra and extracellular amino acids and metabolites observed in the 4xKO strain compared to the penicillin-producing strain DS54468 cultivated at a growth rate of 0.05 in a glucose-limited chemostat. A schematic view of amino acid metabolism is shown based on the KEGG 42 pathways of P. rubrum . Values next to amino acids indicate log2 fold changes if significant. Decreases are indicated by blue, increases by red and unchanged amino acids by grey background. If the change is statistically not significant but p < 0.10, the increase or decline in concentration is denoted by (+) or (−), respectively. Abbreviations: NAC = nicotinic acid; HCIT = homocitrate; aAA = α-aminoadipic acid. Table 1 Strains created in this study and transformations performed. Strain genotype parental strain donor DNA strategy clones total µg marker cassette used tested/positive clones (colony PCR) 2xKO (∆hdfA, ∆Pen-BGC, ∆Chy-BGC) DS68530 1 part, 1500 bp homology, marker free 67 10 24/2 3xKO-A (∆hdfA, ∆Pen-BGC, ∆Chy-BGC, ∆Roq-BGC::amdS) 2xKO 1 part, 100 bp homology, acetamide selection 106 5 6/6 3xKO-B (∆hdfA, ∆Pen-BGC, ∆Chy-BGC, ∆hcpA::amdS) 2xKO 1 part, 100 bp homology, acetamide selection 213 5 6/6 4xKO (∆hdfA, ∆Pen-BGC, ∆Chy-BGC, ∆Roq-BGC::amdS, ∆hcpA::ble) 3xKO-A 1 part, 100 bp homology, phleomycin selection 79 5 6/4 4xKO-B (∆hdfA, ∆Pen-BGC, ∆Chy-BGC, ∆Roq-BGC::ergA, ∆hcpA:: ble) 4xKO 1 part, 100 bp homology, terbinafine selection 238 5 6/5 4xKO-B-PenBGC pen-BGC in IGR 4xKO-B 1 part, >1000 bp homology, acetamide selection 11 3 6/6 DS68530–PenBGC pen-BGC in IGR DS68530 4 parts, >1000 bp homology, acetamide selection 41 3 6/6 4xKO-B-PenBGC-p40s pen-BGC in IGR, p40s for all genes 4xKO-B 8 parts, 100 bp homology, acetamide selection 26 2 17/4 DS68530-PKS17-OE Integrating pIPNS in front of Pc21g16000 DS68530 1 part, 100 bp homoloy, acetamide selection 146 4 6/6 4xKO-B-PKS17-OE Integrating pIPNS in front of Pc21g16000 4xKO-B 100 bp homology, acetamide selection 183 4 6/6 4xKO-B-Cal-BGC Integrating cal-BGC in IGR 4xKO-B 6 parts, 100 to >1000 bp homology, acetamide selection 97 3 16/16 P. rubens devoid of four BGCs shows low secondary metabolite levels The 4xKO strain as well as the intermediate 2xKO, 3xKO-A and 3xKO-B strains, and the parental strain DS47274 (1x pen-BGC ) (see Fig. 2b ) were subjected to shake-flask cultivations using secondary metabolite production (SMP) medium 37 . The culture supernatant was subjected to LC-MS to identify changes in secondary metabolite profiles from day 2 to 7 of cultivation (Fig. 2 , Supplementary Information SI5 , Supplementary Information SI6 ). Importantly and as expected, the total ion chromatograms of untargeted LC-MS runs (Supplementary Information SI7A ) of the 4xKO strain revealed that all penicillin, chrysogine, roquefortine and fungisporin-related metabolites were absent. In the intermediate strains, differential responses in secondary metabolite profiles were apparent. As shown in Fig. 2 , the level of chrysogine increased moderately (1.5 fold on day 5) upon deletion of the Pen-BGC while the level of modified chrysogines (chrysogine 6, 9 and 10) increased up to 29-fold (Supplementary Information SI6 ). However, these compounds are only a minor fraction (with normalized peak areas between 10 3 to 10 5 ) of all produced secondary metabolites whereas chrysogine is present in high quantities (normalized peak areas at 10 7 ) already after 48 h but does not accumulate further throughout the cultivation (Supplementary Information SI6 ). Besides, the levels of histidyltryptophanyldiketopiperazine (HTD) and meleagrin decreased to 0.35- and 0.1-fold of wildtype levels, respectively, while fungisporin related compounds (e.g. YFVV, VFWV) remained almost unchanged. After deletion of the chrysogine BGC (Chy-BGC), an almost 4-fold increase of HTD levels (normalized peak areas at 10 6 ) and a 2.8-fold increase in meleagrin levels was observed together with up to 16-fold more of the linearized fungisporin 30 tetrapeptides consisting of tyrosine, phenylalanine and valine (YFVV, VYFV, FVVY) while tryptophan-containing tetrapeptides (YWVV, VYWV, WVVY) increased only by 1.7-fold (Fig. 2c ). Upon deletion of the roquefortine BGC (Roq-BGC), the product pattern of fungisporins was changed and a 45-fold increase of tryptophan-containing tetrapeptides was detected, whereas the increase in non-tryptophan-containing tetrapeptides was only 2-fold. Similarly, when the fungisporin NRPS gene hcpA 30 was deleted, a 5.8-fold increase of meleagrin compared to levels in DS27472 was observed. Additionally, expression of remaining BGC core genes in the 4xKO strain were quantified using qPCR after 5 days of growth in shake flasks using SMP (Supplementary Information SI8A ) . Of the 45 remaining SM core genes, 7 genes showed an increased expression with log2 FC > 2, however all genes were expressed with a relative expression level below 5% of the reference actin. The extracellular metabolome of the 4xKO strain was also analyzed for the appearance of novel peaks and filtered for changes in m/z abundance using XCMS 38 . Albeit we detected numerous m/z features with a log2 FC greater than 2, only 5 features had a relative peak area greater than 1×10E6. Also, multiple features with an m/z greater than 750 (Supplementary Information SI7 ) were observed. Since no proteins were precipitated prior to analysis of the broth, it might be possible that some of these higher molecular weight features can be related to protein fragments. In conclusion, we did not observe unexpected changes in the secondary metabolites excreted by the 4xKO strain, which represents astrain with a reduced extracellular secondary metabolite metabolome we characterized further on. Characterization of P. rubens devoid of four BGCs using chemostat cultivation LC-MS data indicated changed patterns in BGC related products of intermediate strains which are likely caused by altered amino acid availability due to deletion of BGCs that require specific amino acids and thus draining the available pools for these amino acids. Therefore, a quantification of intracellular amino acids levels in the 4xKO strain was performed under steady-state growth conditions, which can be controlled in glucose-limited chemostats. The maximum growth rate (µ max ) on glucose of two glucose-limited batch cultivations was found to be similar, i.e. 0.15 ± 0.001 h −1 and 0.14 ± 0.003 for DS54468 and 4xKO, respectively. A growth rate of 0.05 h −1 was selected for glucose-limited chemostat cultivations for both strains, as under these conditions there is acceptable production of penicillin 39 while this resembles the growth rate of P. rubens on lactose 40 , 41 , the carbon source used in SMP medium. A comparison of biomass concentration at several points during steady state between both strains revealed a slight increase of 6% in biomass concentration (6.59 ± 0.153 g/kg for DS54468 and 6.98 ± 0.054 g/kg for 4xKO; p = 0.0034, two-tailed students t-test, n = 5 for DS54468 and n = 7 for 4xKO) while dissolved oxygen tension (DOT), CO 2 production, O 2 consumption and base addition remained unchanged (Supplementary Information SI9 )). Morphology of both strains was regularly checked microscopically. Both strains appeared similar in length and aggregation of hyphae during exponential and steady state phase (Supplementary Information SI10 ). To examine possible changes in amino acid pool in the 4xKO strain, mycelium samples from chemostat cultivations of DS54468 and 4xKO were analyzed using LC-MS. The analysis of intra- and extracellular amino acids (Fig. 2d ) indicated an overall modest change in amino acid levels with 13 out of 19 quantified intracellular amino acids remaining unchanged. However, a significant change occurred for sulfur-containing and aromatic amino acids. While intracellular levels of cysteine remained unaltered (0.8 ± 0.4 µM/g CDW), the extracellular level decreased from 2.3 nM to 1.7 nM (−0.37 log2 FC) whereas the intracellular concentration of methionine was reduced from 0.09 ± 0.03 µM/g CDW in DS54468 to 0.02 ± 0.01 µM/g CDW in the 4xKO strain (log2 FC of −1.86) and was completely undetectable in the culture supernatant of 4xKO (DS54468: 0.2 nM, log2 FC of −5.0). All aromatic amino acids increased at a log2 FC of around 0.5 (Trp: 0.11 ± 0.05 µM/g CDW; Tyr: 0.42 ± 0.07 µM/g CDW; Phe: 0.55 ± 0.06 µM/g CDW) in the 4xKO strain. Additionally, an increase of intracellular (log2 FC of 1.74) and extracellular (log2 FC of 2.22) nicotinic acid (NAC, [m + H] + = 124.0393 m/z) was observed. This metabolite is produced from tryptophan and acts as a precursor of nicotinamide adenine dinucleotide (NAD). The intracellular level of valine remained unchanged (3.2 ± 0.8 µM/g CDW), however the valine level in the culture broth increased from 0.03 nM to 0.23 nM (2.73 log2 FC) and the intracellular concentration of α-amino adipic acid increased by a log2 FC of 0.47 in the absence of the penicillin BGC. Similarly, a moderate increase of histidine, the precursor for HTD and roquefortine was measured with concentrations increasing from 1.2 ± 0.2 µM/g CDW in DS54468 to 1.7 ± 0.3 µM/g CDW in the 4xKO strain (log2 FC of 0.49). While the intracellular level of alanine, a precursor for chrysogine, remained unchanged (32.5 ± 4.5 µM/g CDW) no extracellular alanine was detected in the culture supernatant of 4xKO (DS54468: 6.5 nM, log2 FC of −5.0). The intracellular and extracellular level of GlcNAc increased by log2 FC 2.02 and 1.62, respectively. Overall the changes we observe here are modest but indicate the increased levels of amino acids in the absence of utilizing BGCs. Transcriptome profile of P. rubens 4xKO displays distinct expression changes Samples for transcriptome analysis by RNA-sequencing (RNA-seq) were taken from steady state chemostat cultivations and analyzed for changes in transcripts, resulting in 4274 differently expressed genes (Supplementary Information SI11 and SI12 ). Out of the 45 remaining BGC core genes, three NRPS-like and a single PKS (Pc21g00960) showed a log2 FC > 2, however, none of these genes was expressed above a level of 2% relative to actin, showing that silent BGC clusters were not activated in our 4xKO strain consistent with the lack of new secondary metabolites in the extracellular metabolite profile. We next aimed to identify changes in transcript abundance that resulted from absence of the four BGCs and are not solely due to a reduction of the penicillin synthesis burden. We retrieved expression changes from available microarray datasets comparing high-versus-low penicillin production condition for the differently expressed genes identified by RNA-seq and calculated z-scores for each gene (Fig. 3a ). This was possible for 3834 genes (87.9% of the differently expressed genes) with trackable expression behavior in at least 50% of the considered microarray datasets. Among these, 1594 genes (41.6%) were expressed very similar (|z | <0.2) compared to strains where production pressure was reduced by omitting the addition of the penicillin sidechain precursors phenoxyacetic acid or phenylacetic acid to the culture medium. A subset of 2077 genes showed an altered expression with 0.2 < z < −0.2. Very different expression behavior was observed for 82 genes with z above 1.25 and 81 genes had a z-score below −1.25, mainly covering BGC-related genes that were deleted (Fig. 3b and Supplementary Information SI13 ). It was also possible to track genes with a higher log2-FC showing similar responses (Supplementary Information SI14 ). FunCat enrichment analysis (Fig. 3c and Supplementary Information SI14 ) of genes with |z | >0.2 identified 20 categories as enriched (p < 0.05), amongst them metabolism, biosynthesis and degradation of phenylalanine, supporting further analysis of the gene set by mapping it to KEGG 42 Pathways (Fig. 3d and Supplementary Information SI16 ). Figure 3 Analysis of transcriptome changes in 4xKO compared to DS54468 when growing in a glucose limited chemostat at a dilution rate of 0.05 h −1 . ( A) Scheme for identification of genes that are differently expressed due to the absence of four BGCs. If the log2 FC of a gene in 4xKO and DS54468 is more different than observed log2 FCs from microarray experiments of strains grown in absence or presence of penicillin sidechain precursor, the z-score will become negative or positive, depending on the direction of the change. ( B) Distribution of z-scores for sufficiently covered genes and visualization of z-score over contigs, sorted from 1 to 49. Orange and red dots represent genes with a significantly different z-score (rate of false positives <0.05, based on random sampling of normally distributed numbers). A clustering effect of negative z-score is seen for HcpA, Chy-, Roq- and Pen-BGC which are highlighted. C) Enriched FunCat categories (p < 0.05, FDR corrected) derived from 2440 genes where |z | > 0.2. D) Overview of KEGG 42 pathways of P. rubens Wisconsin 54–1255 with up- and downregulated expression identified in this study. Indeed, this gene set contained several genes that were involved in synthesis of the aromatic amino acids tryptophan, tyrosine and phenylalanine, which showed an increased abundance in the 4xKO strain. The entry reaction into the shikimate pathway is mediated by the 3-deoxy-7-phosphoheptulonate synthase (Pc18g02920, log2 FC 0.5) and subsequent enzymes showed also increased expression such as the anthranilate synthase (Pc13g12290, log2 FC 0.4). Further, the fate of aromatic amino acids differs, as the expression of the tryptophan synthase (Pc22g00910 log2 FC −1.4) is decreased in 4xKO, but further conversion of tryptophan to nicotinic acid is presumably enhanced because of increased expression of the kynureninases Pc22g20570 and Pc22g11870 (log2 FC 0.3 and 0.3, respectively) as well as the 3-hydroxyanthranilate 3,4-dioxygenase Pc20g09330 (log2 FC of 0.2). A decreased expression of tyrosinase (Pc22g18500, log2 FC −2.4) was noted which is presumably involved in formation of melanin. The second reaction in the catabolic route of tyrosine, the conversion of 4-hydroxyphenylpyruvate into homogentisate was upregulated with increased expression of the 4-hydroxyphenylpyruvate dioxygenase Pc22g07130 (log2 FC 0.4). The observed increase of phenylalanine levels seems to trigger an increased expression of Pc12g09020 (log2 FC 0.5), coding for a maleylacetoacetate isomerase, which is involved in phenylalanine degradation. Also, an increased degradation of purines by adenosine deaminase (Pc06g00210, log2 FC 2.7) and guanine deaminase Pc16g11230 (log2 FC 0.6) might occur, while the adenine deaminase (Pc16g10530) showed decreased expression (log2 FC −0.5). Further downstream reactions in this pathway were also found to be upregulated, such as expression of IMP dehydrogenase (Pc13g07630, log2 FC 1.8), IMP and pyridine-specific 5′-nucleotidase (Pc12g13510, log2 FC 0.5) and xanthine dehydrogenase (Pc22g06330, log2 FC 0.6), while expression of the putative urate oxidase encoded by Pc22g20960 was decreased (log2 FC of −1.3). The adenosyl homocysteinase encoded by Pc16g05080 was lower expressed (log2 FC −0.3), putatively providing less adenosine. Also, expression of genes involved in pyrimidine biosynthesis was changed, namely the lower expression of the 4-amino-5-hydroxymethyl-2-methylpyrimidine phosphate synthase THI5 homolog encoded by Pc21g15700 (log2 FC −0.7) and an increased expression of the cytosine deaminase Pc16g10090 (log2 FC 0.5). The biosynthesis of ribose 5-phosphate (R5P), the precursor required for purine biosynthesis and the pentose phosphate pathway was also found to be downregulated as the ribose 5-phosphate isomerase encoded by Pc22g21440 showed decreased expression (log2 FC −0.8). The link between the pentose phosphate pathway and glycolysis was also decreased as the Pc21g16950-encoded transaldolase was lower expressed (log2 FC −0.4), presumably resulting in an accumulation of D-arabitol, as the NADP+-dependent D-arabinitol dehydrogenase Pc16g08460 showed increased expression (log2 FC 1.2). Interestingly, it was found that the 6-phosphofructo-2-kinase Pc20g01550 (log2 FC 0.5) also showed increased expression. This is insofar interesting as the product, fructose-2,6-bisphosphate, strongly activates glycolysis through allosteric activation of phosphofructokinase 1. Further, we also found decreased expression of 2 pyruvate decarboxylase enzymes, encoded by Pc18g01490 (log2 FC – 0.7) and Pc13g09300 (log2 FC −0.6) which break down pyruvate into acetaldehyde and carbon dioxide, during anaerobic fermentation. Also, the carbonic anhydrase Pc22g06300 (log2 FC −0.6) was downregulated, which indicates that there might be less induction of these enzymes in the 4xKO strain. Regarding the decrease in methionine and cysteine, this can perhaps be explained by decreased sulfur supply, as the phosphoadenosine phosphosulfate reductase Pc20g03220 was lower expressed (log2 FC −0.5). This enzyme is required for fixation of inorganic sulfur via sulfite on 3′-phosphoadenosine-5′-phosphosulfate (PAPS). In addition, an increase of the cysteine dioxygenase expression (Pc21g04760, log2 FC 1.7), initiating the conversion of cysteine into taurine or degradation to sulfate was observed and might cause a decrease in methionine. Moreover, the cystathionine gamma-lyase (Pc21g05430, log2 Fc −0.8) breaking down cystathionine into cysteine, α-ketobutyrate and ammonia was downregulated. The lack of methionine might have been sensed by the cell as the L-methionine (R)-S-oxide reductase Pc20g05770 showed decreased expression (log2 FC −0.4). Since methionine and the related S-adenosylmethionine (SAM) act as cellular nutrient state sensors and have a significant impact upon regulation of autophagy 43 , 44 , the decrease in methionine levels also explains the upregulation of 11 autophagy-associated genes (log2 FC between 0.25 and 0.47, Supplementary Information SI16 ), amongst them the well-expressed (497.3 RPKM in DS54468) autophagy-related protein 8 (Atg8), encoded by Pc12g05370 (log2 FC 0.26). Atg8 is required for the formation of autophagosomal membranes during macroautophagy 45 and is coupled to phosphatidylethanolamine (PE) by the cysteine protease Atg4 46 (Pc20g08610, log2 FC 0.28). An increased expression of these genes is already observed when lowering the penicillin biosynthesis burden, as indicated by |z | <0.2 (Supplementary Information SI13 ) . One exception is Pc12g10930 (z = −0.31, log2 FC of 0.47), encoding for the autophagy-related protein 13 (Atg13) that interacts with Atg1 47 and whose phosphorylation status mediates interactions with other autophagy related proteins 45 . Here, an increased expression of atg13 might suggest a further increase in autophagy as compared to strains still capable of producing penicillin. This could also result in an increased degradation of peroxisomes, which has been shown to be regulated by atg1 in P. rubens 48 . Taken together, these changes in gene expression are consistent with a reduced cellular demand for amino acids in the absence of the four highly expressed BGCs. It appears that the 4xKO cells redirect gene expression to avoid accumulation of aromatic amino acids. Also, since tryptophan and adenine are normally required for synthesis of NAD and NADP, having an increased pool under high-penicillin-production conditions 49 is beneficial. However, these reservoirs trigger increased expression of degradation enzymes when demands are decreased as observed in the 4xKO strain. Overall, these changes suggest that the obtained strain will be suitable for the integration of heterologous gene clusters and expression thereof. To verify this, we next set out to overexpress endogenous and integrate heterologous secondary-metabolite related genes. P. rubens platform strain is suitable for multi-part donor DNA assembly and achieves increased yield of PKS-derived YWA1 To evaluate the secondary metabolite deficient strain as a platform, we first re-introduced the pen-BGC into the intergenic region (IGR) between Pc20g07090 and Pc20g08100 using in vivo homologous recombination using up to 8 DNA fragments with only 100 basepair overlap between fragments. We used two strategies for this approach, maintaining the native configuration of the cluster or replacing the promoters by the stronger p40s 50 promoter (Fig. 5a ). While the first strategy led to a higher transformant number and successful integration, we observed a higher frequency of multiple gene copies being integrated when p40s was used, suggesting that recombination of very similar parts is less successful and prone to errors (Fig. 5b and Table 1 ). The recombination was successful as evidenced by the concentration of penicillin V after five days of growth in SMP medium supplemented with phenoxyacetic acid (POA) (Fig. 5c ), even though the concentration of penicillin V was slightly lower for the recombined strains as compared to DS56830-penBGC and the reference strain for single-copy Pen-BGC production, DS54468. Investigation of the expression of all pen-BGC genes revealed high levels of expression in DS56830-penBGC and the strains expressing the pen-BGC genes from the p40s promoter (Fig. 5d ). The pen-BGC was similarly expressed in the 4xKO strain compared to the parental strain DS54468. Our results indicate that we can re-assemble a complete BGC from up to 8 DNA fragments with short homology successful, allowing efficient BGC pathway assembly with high functionality. To further explore the performance of the platform strain, the polyketide synthase PKS17 (Pc16g1700) which is not expressed in P. rubens under submerged cultivation conditions was overexpressed by placing it under control of the pIPNS promoter (Fig. 4a ). Pc16g1700 encodes an iterative, non-reducing type I PKS, termed PKS17 or pcAlb1 , producing the heptaketide naphthapyrone YWA1 by condensation of one acetyl-CoA and six malonyl-CoA moieties. YWA1 is the precursor for dihydroxynaphthalene (DHN)‐melanin in several fungi 51 and the BGC encoding the required enzymes is present in P. rubens 52 . Replacing the native promoter by pIPNS resulted in production of YWA1 in both DS68530-PKS17-OE and 4xKO-B-PKS17-OE, whereby the latter strain background increased the production of YWA1 by 25% on day 3 and by 600% on day 5 (Fig. 4b ). qPCR was used to measure expression of Pc16g1700 on day 5, and this showed no significantly different expression between DS68530-PKS17-OE and 4xKO-B-PKS17-OE (Fig. 4c ). Although YWA1 is further processed into insoluble pigments which were not quantified further, these results indicate that the supply of precursors for YWA1, malonyl-CoA and acetyl-CoA could be increased in the 4xKO background strain. Indeed, data from chemostat cultivations of 4xKO showed that Pc13g03920, encoding the P. rubens ortholog for the S. cerevisiae acetyl-CoA carboxylase (ACC) Sce ACC 53 (UniProt Q00955.2) displayed a moderate increased expression (log2 FC of 0.36). ACC catalyzes the rate-limiting step in fatty-acid biosynthesis, the carboxylation of acetyl-CoA to malonyl-CoA, which is the limiting substrate for the biosynthesis of fatty acids via fatty-acid synthase 54 . Since the biosynthesis of YWA1 presents a drain for malonyl-CoA, an increased supply of malonyl-CoA via ACC could explain the increased level of YWA1. Our experiments show that the obtained 4xKO strain is suitable for producing increased amounts of the polyketide YWA1. Figure 4 Overexpression of PKS17 in DS68530 and 4xKO-B. ( A) Schema showing the strategy used for integrating the IPNS promoter in front of Pc21g16000. ( B) The initial molecule produced by PKS17 is the naphthopyrone YWA1, which was quantified by LC-MS in fermentation broth of the indicated strains after 3 and 5 days. ( C) Expression of Pc21g16000 quantified by means of qPCR on day 3 of growth in SMP. The gene is not expressed in DS68530 and 4xKO-B and expression when replacing the promoter is unchanged between DS68530-PKS17-OE and 4KO-B-PKS17-OE, as seen by the difference in the similar Δct. Figure 5 Integration of the Penicillin cluster into DS68530 and 4xKO-B. ( A) Scheme for recombination of parts obtained by PCR into the intergenic region of Pc20g07090 and Pc20g08100 using either the native promoters or p40s for expressing all genes of the pen-BGC. ( B) Copy number of integrated pen-BGC genes in the obtained strains. ( C) Penicillin V concentration after five days of growth in SMP + POA medium ( D ) Changes in gene expression for pcbAB, pcBC, penDE and parA relative to the single-copy pen-BGC strain DS54468. Integration of heterologous Calbistrin gene cluster into the P. rubens platform strain results in production of decumbenones To demonstrate that the obtained 4xKO platform strain is suitable for integration of heterologous BGCs, we integrated the calbistrin-BGC (cal-BGC) recently identified in Penicillium decumbens 55 . The products of the Cal-BGC are calbistrins and decumbenones, with the latter known to impair melanization of Magnaporthe grisea 56 , the cause of rice blast. Integration of the Cal-BGC into the genome of P. rubens 4xKO-B (Table 1 ) was achieved via in vivo homologous recombination (Fig. 6a and Supplementary Information SI17 ). After liquid cultivation, we detected the linear moiety (dioic acid), decumbenone A, B and C but no Calbistrin A, C and versiol in both CYA and SMP medium samples of 4xKO-B-calBGC (Fig. 6b and Supplementary Information SI18 ). Since our previous study could not rule out the possibility of a second PKS producing the linear moiety required for synthesis of calbistrins, we examined P. rubens for a possible upregulation of closely related PKSs with a potential homolog present in the calbistrin producers A. versicolor , A. aculeatus , P. decumbens and C. tofieldiae . The qPCR measurements did reveal a moderate increase in gene expression for Pc16g04890 (log2 FC of 1.3, Supplementary Information SI19 ), containing a C-methylation domain and an a enoyl-reductase domain, structurally resembling a highly-reducing PKS 57 proposed to be necessary for synthesis of the linear moiety, however the confirmation of this hypothesis requires further experimental validation. Since we did not observe production of calbistrin A and C, this could suggest either a non-clustered broad specificity transesterase forms the ester bond of the calbistrins or hydrolysis of calbistrins is occurring very rapidly in P. rubens . Figure 6 Integration of the calbistrin cluster from P. decumbens into 4xKO-B and verification of production. ( A) Scheme for recombination of six parts obtained by PCR into the intergenic region of Pc20g07090 and Pc20g08100. Obtained clones were verified by colony PCR (Supplementary information 19 ). ( B) Total-ion-chromatograms of samples taken five days after inoculation of CYA medium. Shown are 4xKO-B, a representative clone (4xKO-B-CalBGC-C2) and P. decumbens , serving as a positive control. Arrows indicate the retention times of the depicted molecules. ( C ) Peak areas of calbistrin-related metabolites quantified in SMP medium and CYA medium taken five days after inoculation. Peak areas are depicted as mean of biological triplicates for 4xKO-B-CalBGC and biological duplicates for P. decumbens . No calbistrin A and C were detected in supernatant of 4xKO-B-CalBGC. An overview of retention times used m/z for quantification and obtained culture dry weight can be found in (Supplementary Information 20 and 22 ). ( D) Appearance of previously not observed peak in 4xKO-B CalBGC after 7 days of cultivation in CYA medium at a retention time of 12.1 min. Most abundant m/z in this peak were m/z 255.122, 273.132 and 290.159. Expression of the Cal-BGC reduced biomass productionof P. rubens cultured in SMP medium and also abolished spore pigmentation (Supplementary Information SI20 ). However, for 4xKO-B-calBGC, the level of all decumbenones were increased in SMP medium at least one-fold when compared to CYA (Fig. 6c and Supplementary Information SI18 ) with maximum levels observed on day 5 of growth. Besides reduced amounts of decumbenones, we also observed appearance of a significant peak on day 7 in 4xKO-B-calBGC, composed of 3 m/z: 255.122, 273.132 and 290.159 (Fig. 6d ), not observed in P. decumbens , suggesting degradation of decumbenones. Taken together, these observations indicate that the Cal-BGC contains all relevant genes for production of decumbenones and these can be successfully transferred to P. rubens for heterologous expression and high-level production."
} | 11,159 |
32163891 | PMC7133052 | pmc | 8,285 | {
"abstract": "Highlights • S-MFCs with different electrode heights were investigated at the cascade level. • S-MFCs can be scaled up to 4 cm electrode height without performance decrease. • S-MFCs cannot be scaled up to 12 cm electrode height without performance decrease. • The colonisation of the Cathodic biofilms do not exceed 6 cm depth. • The cathode coverage by the biofilms seems to reflect the S-MFC efficiency.",
"conclusion": "4 Conclusion The scalability of individual bioreactors is important aspect of the research in the microbial fuel cells field. It is necessary to define the size of the individual bioreactors to maintain maximum efficiency when assembling them into stacks that are aimed at implementing the technology under real conditions of use. In this perspective, the results presented here further define the scalability range of S-MFC treating urine. A previous study has shown that the lower scalability limit on the individual bioreactor was between 1 cm and 2 cm electrode height. Here, results suggest that single S-MFC bioreactors could be scaled up to 5–6 cm electrode height before having decreased electrochemical performances. This hypothesis is supported by the similar electrochemical performance of either the single bioreactors or the cascades of S-MFC under 2 cm and 4 cm conditions. Conversely, the S-MFCs under 12 cm conditions have shown decrease performance. This poor performance was shown to be linked to the cathodic biofilm development. In S-MFC treating neat urine, the cathodic biofilm only colonised the upper 5–6 cm of the cathode. This indicates that the cathode has a mixed redox potential between the top and bottom part of the cathode, with the upper part being more oxidised than the lower part which was more reduced. These results imply that if the size of the individual reactor should be increased, the design should be modified to match these redox differences and minimise their impact on performance.",
"introduction": "1 Introduction Microbial fuel cells (MFCs) are of increasing interest because they combine the production of low levels of electricity and the treatment of different types of wastewater: the reduced organic matter contained in wastewater is converted directly into electricity through the metabolic activity of anaerobic electro-active microorganisms [1] , [2] , [3] , [4] . Like in other anaerobic bioreactors, the enriched microbial community degrade the organic matter through a series of anaerobic metabolic reactions. The key characteristic of microbial fuel cells is that the enriched communities performing anaerobic respiration have the capacity to employ an electrode as the end-terminal electron acceptor. During this process, protons, smaller organic molecules and CO 2 are transferred into the electrolyte. The electrons collected by the anode flow towards the cathode through an external load, whilst protons diffuse towards the cathode half-cell. At the cathode, protons and electrons react through a reduction reaction with an oxidant of a higher redox potential (e.g. oxygen) [5] , resulting in the production of current (electron flow). The main advantage of this technology is that it can treat waste streams of various sources (e.g. activated sludge [6] , neat urine [7] and others [8] , [9] ). In developed urban settings, urine is the source of 75%, 50% and 10% of the nitrogen, phosphorous and COD present in domestic wastewater, respectively, whilst it only amounts to less than 5% of the total volume [10] , [11] . Treating this type of waste stream prior to reaching wastewater treatment plants is therefore an attractive solution [11] , [12] to increase the energy efficiency of wastewater treatment [7] , [13] , [14] , [15] , [16] . To the authors’ best knowledge, the MFC is the only biotechnology able to directly treat neat urine – with no dilution and without succumbing to inhibition due to high ammonium concentrations – and does not require any energy input [17] . A recent study has shown that the chemical oxygen demand (COD) and total nitrogen (TN) loadings can be reduced by 88% (from 5.586 to 0.672 g l −1 ) and 29% (from 4.525 to 3.233 g l −1 ), respectively, at a hydraulic retention time of 44 h [17] . Although these final concentrations would not allow direct discharge into the environment, the removal rates are close to the industrial sector (92% COD and 20% TN reduction) [18] . These reduction rates were obtained with self-stratifying membraneless MFCs (S-MFC) that were recently developed for the treatment of this particular fuel. With microbial fuel cells, usable power levels and enhanced treatment are obtained when a plurality of MFCs are assembled in stacks [19] , [20] . For this reason, single MFC units of a stack have to be simple in design and also cost effective. Such an equilibrium between size, design simplicity and cost has been reached with S-MFC [14] , [21] , [22] . S-MFCs exploit the capacity of microorganisms to structure, in any wet environment, horizontal microenvironments characterised by specific bio-chemical conditions (i.e. redox state of chemical elements, type of dominating metabolic activity) [23] . Here, the S-MFCs exploits this phenomenon in a urine column with the cathodes being placed in the upper oxidised layers and the anodes being placed in the lower reduced layers. The main advantage of this design is that it allows having a plurality of vertical electrodes. Hence, such design authorities the build modules that can be scaled-up in width and length (i.e. more vertical electrodes and longer electrodes, respectively) with no performance losses [14] , [17] . Due to the nature of how this particular fuel is collected, any implementation at ground level would benefit from having shallow units. Until now however, all the studies on S-MFCs have employed units with comparable heights (i.e. ≈10 cm urine column depth; 4.5 cm electrode height) [14] , [17] , [21] , [22] , [24] . Although S-MFCs can be scaled in width and length with negligible power density losses, the height scalability of this type of MFC is yet unknown. Hence, the present study reports on the height scalability of S-MFCs mounted with electrodes of 2 cm, 4 cm and 12 cm heights. In these bioreactors, the cathode and anode electrodes had the same height with 3 mm distance from each other. Here, the S-MFC comprised only a single cathode placed above a single anode. The experimental set-up comprised triplicate cascades of either 6 S-MFC, 3 S-MFC and 1 S-MFC, for the 2 cm, 4 cm and 12 cm conditions, respectively (see Fig. 1 ). Under these conditions, all cascades had the same hydraulic retention time (HRT), total electrode surface area and total displacement volume. The cathodes were based on a carbonaceous-based catalyst pressed over a stainless steel (grade 316) current collectors, and the anodes were made from carbon veil. Once inoculated, the S-MFCs were fed by a continuous flow of urine and maintained under potentiostatic conditions (400 mV). When steady-state was reached, the electrochemical properties were investigated. Fig. 1 Illustration of the S-MFCs set-ups employed in the present study. 3D CAD models (a) of the three tested conditions (2 cm, 4 cm, 12 cm); the red arrows shows the inlets, the outlets and the flow direction of the fuel. In (b) the picture shows an assembled example of each of the three tested MFC-size. In (c) and (d) the pictures show one of the triplicates assembled cascades: 6 reactors for the 2 cm condition (c) and 3 reactors for the 4 cm condition (d). The 12 cm condition consisted of a single S-MFC. Fresh fuel was pumped in the top MFCs and then cascading to the underneath downstream ones. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)",
"discussion": "3 Results and discussion 3.1 Cathode polarisation curves in untreated urine The cathodes were initially characterised in urine from the collecting tank as electrolyte (i.e. partially hydrolysed, pH = 8.4; EC = 29.46 mS cm −1 ). The polarisation experiments were run in a three-electrode configuration, as described in Section 2.2 , on the first S-MFC of each cascade. The hypothesis was that since the wet surface area of the electrodes were proportionally scaled between the three tested conditions, the displayed current density should be the same. Hence, the oxygen reduction reaction (ORR) rate should be proportional to the amount of wet surface area of cathode. As the OCP is a thermodynamic characteristic that is independent from the size of electrodes, the open circuit potential (OCP) should be the same for all tested configurations. In fact, Fig. 2 shows similar OCPs measuring 108 ± 24 mV (vs Ag/AgCl). Particularly, the 12 cm condition had the lower value of 83 ± 4 mV (vs Ag/AgCl) while the 2 cm and 4 cm conditions had slightly higher value of 110 ± 9 mV (vs Ag/AgCl) and 131 ± 8 mV (vs Ag/AgCl), respectively). The slightly lower OCPs of the 12 cm cathodes might be due the lower oxygen diffusion from the surface of the water column (air/liquid interface) to the bottom part of the cathode. Although the cathode surface area of the S-MFC mounted with 12 cm electrode is either three times larger (vs 4 cm conditions) or six times larger (vs 2 cm conditions), the current density at 0 mV vs Ag/AgCl (0.0217 ± 0.0004 mA cm −2 ) was four times lower than the 2 cm and 4 cm conditions, 0.0993 ± 0.0072 mA cm −2 and 0.0847 ± 0.0078 mA cm −2 , respectively. These results further support the hypothesis that in the S-MFCs mounted with 12 cm electrodes aerobic and anaerobic zones were forming over the depth of the electrode generating mixed-potentials and therefore limiting the ORR. Conversely, the 2 cm displayed slightly higher but comparable current densities at 0 mV (vs Ag/AgCl) compared to 4 cm conditions, with differences quantified in less than 10%. These results are in line with previous studies on the scalability of individual S-MFC [22] , [25] . Fig. 2 Polarisation curves of the cathodes in untreated and partially hydrolysed urine (pH = 8.06; EC = 14.86 mS cm −1 ) prior to inoculation (□ 2 cm; △ 4 cm; ○ 12 cm). (a) shows the absolute current, whilst (b) shows the current normalised by the total cathode surface area. Error bars stand for the standard deviation of the triplicates. 3.2 Temporal behaviour of the S-MFC cascades After the initial cathode polarisation, cascades were mounted in triplicate with 6 units per cascade for the 2 cm condition, 3 units per cascade for the 4 cm condition and a single unit for the 12 cm condition. All the S-MFCs composing each cascade were electrically connected in parallel. Also, all cascades had the same total volume (108 ± 2 ml), the same total surface area of cathode and anode (50.4 ± 1 cm 2 and 360 ± 1 cm 2 , respectively) and the same HRT of 606 min ( Table 1 ). Each cascade was then run under potentiostatic condition at 400 mV. Two different groups can be observed through the 60 days run ( Fig. 3 ). On one hand the cascades with 2 cm and 4 cm tall electrodes showed similar power outputs levels (3.61 ± 0.21 mW between day 15 and 40). On the other hand, the 12 cm conditions displayed much lower power output over the entire length of the temporal run (0.66 ± 0.12 mW between day 15 and 60). The power level of these S-MFC mounted with 12 cm tall electrodes only represent 18% of what the two other conditions produced. At the inoculation phase, up to day 3, all tested conditions behaved similarly. However, after day 3, the power production of the 12 cm conditions plateaued whilst the power produced by 2 cm and 4 cm cascade continue increasing similarly, up to day 15 when steady state was reached ( Fig. 3 ). These results indicate that a factor was limiting the performances of the 12 cm S-MFCs. The results from the initial cathode polarisation tend to indicate that the limited power production of the 12 cm conditions could be linked to the limited ORR in the cathodic layers. Moreover, the hypothesis that the ORR is limited in the 12 cm conditions is further supported by the observed development of the cathodic biofilms ( Fig. 4 ). Fig. 3 Temporal absolute overall power output of the 3 tested cascades (□ 2 cm; △ 4 cm; ○ 12 cm conditions). All S-MFC in each cascade were electrically connected in parallel. Error bars stand for the standard deviation between triplicates. Fig. 4 Pictures of the developed cathodic biofilm in the tested S-MFCs 30 days after inoculation. Picture of one of (a) the 2 cm conditions, (b) the 4 cm conditions, and (c) the 12 cm conditions. On the S-MFCs mounted with 12 cm electrodes, the cathodic biofilm does not cover the totality of the cathode. The biofilm only colonises the cathode up to a depth of 5.5–6.0 cm depending on the triplicate observed ( Fig. 4 c). Conversely, the 2 cm and 4 cm conditions displayed a complete colonisation of the cathode by the cathodic biofilm. The development of this cathodic biofilm could serve as a indicator of the efficiency of the electro-bioreactor’s design. Since the cathodic biofilm in the 12 cm conditions could not extend further than 5.5–6.0 cm in depth, this implies that the redoxcline cannot be established beyond that depth, in S-MFCs treating urine. Interestingly, as mentioned above, diverse environmental conditions such as aerobic and anaerobic might occur along the electrode depth creating mixed potentials and lowering the cathode and the overall performance. These results lead to the observation that S-MFCs cannot be scaled-up in size beyond a cathodic immersion depth of 5.5–6.0 cm. Nonetheless, the cascades of S-MFCs mounted with either 2 cm tall or 4 cm tall electrodes displayed similar power outputs ( Fig. 3 ), hence, confirming that S-MFC can be scaled between 2 cm and 4 cm tall electrodes without performance losses. This hypothesis further support a previous study on the miniaturisation of S-MFCs [25] . However, in the present study the scalability is observed at the cascade scale: A cascade of 6 S-MFCs with 2 cm tall electrodes produce as much overall power as a cascade of 3 S-MFCs with 4 cm tall electrodes. 3.3 Characterisation of the individual S-MFCs and the assembled cascades Once all cascades reached steady state, at day 40 ( Fig. 3 ), the electrodes of each first S-MFC of each cascade were characterised through a linear sweep voltammetry experiment ran in a three-electrodes configuration (see Section 2.2 ). The polarisation experiments were run after leaving the SSM-MFCs for 1 h under open circuit conditions. In order to minimise the ohmic losses caused by the electrolyte separating the two electrodes [28] , the anode and cathode potentials were recorded separately (2 different polarisation experiments) against a Ag/AgCl (3 M KCl) reference electrode placed either next to the middle of the cathode or next to the middle of the anode. During the cathode polarisation, the anode was used as the counter electrode and vice versa during the anode polarisation, the cathode was used as the counter electrode. The following day, the cascades that have all the S-MFCs electrically connected in parallel were characterised through a linear sweep voltammetry experiment ran in a two-electrodes configuration (see Section 2.2 ). As for the other results, the S-MFCs mounted with 12 cm tall electrodes had lower overall performance than the 2 cm and 4 cm conditions ( Fig. 5 ). The open circuit potential (OCP) of the cathode of the 12 cm conditions was half the OCP (67 ± 5 mV) of the 2 cm and 4 cm conditions (128 ± 5 mV; Fig. 5 a). Interestingly, the anode’s OCP were increasing with the size of the reactor (2 cm: −498 ± 2 mV; 4 cm: −523 ± 1 mV; 12 cm: −545 ± 6 mV). The anode OCP is mainly influenced by the presence of reduced elements (e.g. organics) and of oxidised elements (e.g. oxygen). The possible explanation for higher potential can be related to the reactor size: the taller is the reactor (12 cm > 4 cm > 2 cm) the further away from the anode is the surface of the electrolyte and the oxygenic atmosphere. Fig. 5 Electrochemical characterisation of S-MFCs. (a) individual electrode characterisation of the first S-MFC of each condition (□ 2 cm; △ 4 cm; ○ 12 cm). (b) current production normalised by the total surface of each cathodes. In (a) and (b), white symbols stand for the cathode curves and the black symbols stands for the anodes curves. (c) polarisation curves of the cascade of each condition. In (c) the white symbols stand for the power curves and the black symbols stands for the potential curves. The open circuit voltage (OCV) of all the S-MFCs was similar and measured 623 ± 3 mV, 643 ± 5 mV and 618 ± 10 mV for the 2 cm, 4 cm and 12 cm conditions, respectively. From the perspective of scalability, the results indicate that only the 12 cm conditions differed from the two other conditions. Even though the anode polarisations show that the 12 cm conditions produced more overall current ( Fig. 5 a), the normalised data for the surface area clearly shown that despite similar behaviour till 300 μA cm −2 , the current densities of the 12 cm conditions went in diffusion limitation much before compared to the two other conditions, which the limitation was reached around 500 μA cm −2 ( Fig. 5 b). As seen in Fig. 5 b, the anodes and cathodes current densities of the 2 cm and 4 cm conditions were comparable, further supporting the scalability of S-MFCs between 2 cm and 4 cm tall electrodes. The limited cathodic biofilm development ( Fig. 4 c) together with the lower OCP of the cathode ( Fig. 5 a and b) indicates that part of the 12 cm conditions cathodes is under anoxic condition. The fact of having part of the cathode exposed to both aerobic and anaerobic conditions correspond to a microbial snorkel as defined by Erable et al. [29] , thus explaining why at the scale of the 12 cm conditions S-MFCs have limited current output. The results of the linear sweep voltammetry experiment ran on the cascade indicate that the 2 cm and 4 cm conditions produced at 350 mV similar maximum power outputs of 5.68 ± 0.34 mW and 5.22 ± 0.29 mW, respectively. Conversely, the 12 cm S-MFCs/cascade produced a maximum power of 1.48 ± 0.15 mW at 350 mV ( Fig. 5 c). These results further support the fact that S-MFCs cannot be mounted with cathodes immersed below a 5.5–6.0 cm depth. 3.4 Behaviour of individual S-MFCs within the cascades The previous LSVs experiments, in a two-electrodes configuration, have focus on the whole cascade ( Fig. 5 c). However, beside the 12 cm conditions “cascades” that only comprised a single unit, each of the 2 cm and 4 cm conditions cascades comprised either 6 units or 3 units, respectively. Results have shown that at the cascade level both 2 cm and 4 cm conditions were similar. As shown in a previous study on urine-fuelled MFCs, the size of the reactor and the hydraulic retention time (HRT) affects the performance of both the single unit and the cascade [30] . In the present study and with the described reactors, although at the level of the cascade all parameters were identical, at the level of a single S-MFC the HRT of the 2 cm conditions was 50% shorter than the HRT of a 4 cm conditions ( Table 1 ). Hence, the performance of a single S-MFCs will differ between the two sizes. For this reason, the performance comparison will be made between S-MFCs of the same size, whilst the behaviour of each cell within the cascade will be compared between the two tested sizes. The behaviour of each S-MFC in function of their position within a cascade was thus characterised by a LSV experiment in a two-electrodes configuration. Nonetheless, results show the impact of the position of a S-MFC within a cascade on its electrochemical performance ( Fig. 6 ). Fig. 6 Potential and power curves of each of the S-MFC of the 2 cm conditions (a, b, respectively) and the 4 cm conditions (c, d, respectively), function of their position within the cascade: 1st MFC of the cascade ( □ ), 2nd of the cascade ( ), 3rd of the cascade ( ), 4th of the cascade ( ), 5th of the cascade ( ), 6th of the cascade ( ). The 2 cm condition stack had 6 MFCs cascading into each other, and the 4 cm cascade had 3 MFCs. Error bars stand for the measured variation between triplicates. Compared to the downstream S-MFC, the first S-MFCs (Level 1) of the 2 cm condition cascades displayed a lower maximum power transfer point (MPT) of 1134 ± 28 µW ( Fig. 6 b). The MPT of the downstream S-MFCs ranged from 1300 ± 78 µW (level 4) to 1473 ± 219 µW (level 6). Independently from their positions, all S-MFCs of the 2 cm conditions had the MPT at 350 ± 15 mV ( Fig. 6 a). Conversely, the S-MFCs with 4 cm tall electrodes displayed similar maximum power transfer point at 375 ± 15 mV ( Fig. 6 c) (1923 ± 140 µW), independently from their position within the cascade ( Fig. 6 d). It has to be noted that the average MPT of the first cascade of the 4 cm conditions was slightly lower than the two other cascades (1780 ± 62 µW vs. 1994 ± 62 µW), whereas all three cascades of the 2 cm conditions were more homogenous (1373 ± 32 µW). The impact of the HRT can be seen at two levels. Firstly, compared to the 4 cm conditions’ S-MFCs that had double the HRT, the individual S-MFCs of the 2 cm conditions cascades had higher power densities. Secondly, the level-1 S-MFCs of the 2 cm conditions had a lower MPT than the other downstream S-MFCs. The hypothesis of this lower performance is that either the electrolyte did not yet have the time to be sufficiently reduced and/or the organic content did not have sufficient time to be broken down and be accessible to the electroactive microorganisms. Another hypothesis is the fact that the electrolyte contains oxygen once entering the first S-MFC (2 cm), which is fully consumed and then the electrolyte is completely anaerobic in the 2nd level of the 2 cm S-MFC. The faster flow might affect the self-stratification in the level 1 S-MFC, Moreover, as the flow is slower in the 4 cm condition, it might be possible that the oxygen is consumed within the first S-MFC not affecting the electrochemical output. Another observation is the fact that the S-MFCs from the 2 cm conditions displayed greater variation than the 4 cm conditions, as reflected by the errors bars of the power curves in Fig. 6 . 3.5 Cathodes polarisation function of the reference depth During the polarisation experiment of a cathode, the reference electrode slipped from its original position and moved near the surface of the electrolyte. This change in position/depth resulted in an increase of the current production. This observation thus raised the question of the methodology to employ when investigating the electrochemical properties of self-stratifying MFCs. Up to this point, for LSV experiment run in a three-electrodes configuration, the reference electrode was placed next to the geometrical centre of the immersed part of the cathode ( Fig. 6 ). Since the height is a parameter that was investigated in the present study, the depth at which was placed the reference electrode differed from one size to another. Hence, an experiment was set to investigate the impact of the reference electrode depth position on the measured current outputs. LSV experiments in a three-electrodes configuration were performed on the cathodes of two S-MFCs mounted with 4 cm tall electrodes. This experiment was not performed on the 12 cm condition S-MFCs due to the time needed to recover the OCP (≈6 h; data not shown), nor on the 2 cm conditions since only 15 mm of their cathode was exposed to the electrolyte. Due to recovery time between experimental runs, only two S-MFC representatives of the 4 cm conditions were chosen, the S-MFC having the higher power output (the last S-MFC of the third cascade; C3) and the one with the lower power output (the first S-MFC of the first cascade; A1). Polarisation curves and power curves of the S-MFCs A1 and C3 are reported in Fig. 7 a. LSV experiment on the cathodes started from the top of the urine column (2 mm) and continued by 10 mm increments, until the bottom of the immersed part of the cathode (30 mm). To confirm that the observed variations were due to the position of the reference electrode and not to the succeeding polarisations, the last run was performed at 2 mm depth and compared to the initial cathode polarisation run at this depth. The results show, for both S-MFCs (A1 and C3), that the current produced by the initial and last 2 mm depth runs are similar ( Fig. 7 b and c). Hence, the successive polarisations have not impacted the results. Fig. 7 Polarisation of the best (C3, ○ )and less performing (A1, □ ) of the 4 cm conditions S-MFCs in a two-electrode configuration (a). In (a), the black symbols stand for the potential curves and the white symbols stand for the power curves. Cathode polarisation of S-MFC function of the depth of the reference electrode in either the 1st S-MFC of the 1st cascade (A1; b) or in the last S-MFC of the third cascade (C3; c): first with the reference electrode at 2 mm depth ( ○ ), then at 10 mm depth ( ), 20 mm depth ( ), 30 mm depth ( ) and finally again at 2 mm depth (●, control). These are single experimental runs. Following results from the LSV experiment run in two-electrodes configuration ( Fig. 7 a), the cathode of the C1 S-MFC produced more current than the A1 ( Fig. 7 b and c). Nonetheless, both S-MFCs showed the same response to the increase of the reference electrode depth: the more the reference electrode was immersed in the electrolyte, the less current was produced by the cathodes. This observation implies that the ORR rate was more sluggish in the lower part of the cathode, compared to the ORR rate of the upper part of the cathode. This difference in the ORR rates can be justified by the bio-physico-chemical stratification of the urine column occurring all along the cathode and therefore that the system is more complex than originally thought. This could explain why no oxygen has been measured in the lower part of the cathodic layer [14] . Interestingly, such phenomenon was not observed for the anodes, which is assumed to be completely anaerobic. These results imply that when reporting the performance of the cathode in S-MFCs, the positioning of the reference electrode should be carefully evaluated and therefore also the results interpretation because diverse environmental conditions (aerobic, anaerobic and anoxic) might occur within the same electrode. 3.6 Outlook and future work This work is based on scaling up self-stratifying MFCs and in order to understand the limit in height, 2 cm (6-unit cascade), 4 cm (3-unit cascade) and 12 cm (single unit) were tested. It was noticed that, during over 70 days of operations, the cathodic biofilms colonisation front stops roughly after 5 cm and roughly before 6 cm. It was therefore found that this was the critical condition and further investigation around that specific height should be carried out in further investigations. Experiments conducted on 12 cm height cathode showed mixed potentials occurring and potential readings and conditions different in function of the position of the reference electrode within the electrolyte. Further investigation should also been driven towards: (i) the identification of bacterial profile within the column of urine along the cathode electrode; (ii) measurement of the oxygen profile along the column for identifying the concentration of oxygen and correlate it with the bacterial species. As microbial fuel cell is a low current/power producing technology, the reduction and limitation of cost is imperative, therefore the choice of the electrode is somehow imposed. In our case, carbon veil was used as anode electrode material. This material is relatively cheap and guarantee high surface area for biofilm attachment and development [31] , it provides the possibility of perfusion limiting or neglecting diffusion problems [32] , it was shown to be durable for long terms operations [21] and has good electrical conductivity and resistance to corrosion and harsh environments [21] . The properties of this material as well as other carbonaceous-based electrodes can be improved by doping the surface with carbonaceous powder [33] , functional groups [34] , active polymers [35] or metals nanoparticles [36] . All these functionalisations produces changes on the surface such as: (i) enhance the electrical conductivity of the carbon substrate; (ii) modify surface roughness inserting anchoring point on the surface; (iii) increment the hydrophillicity of the surface helping the biofilm attachment and growth, (iv) act as mediator improving the electron transfer to the electrode surface [37] . In parallel, also the cathode used in this work is a traditional AC/PTFE mixture pressed over a stainless steel mesh. Commercially available AC possesses high surface area and they are available in large quantity at low cost. AC is a good catalyst mainly because of its high porosity being the neutral media limited significantly by the pH and the very low concentration of H + and OH − that are reagents during ORR [38] . Due to its high surface area, AC does not exhibit high electrical conductivity that can be enhanced by the addition of carbon black, graphene or other carbonaceous materials [39] . Another way of enhancing the ORR catalysis in neutral media is utilising platinum (Pt) or Pt-based catalysts but the excessive cost and the low durability decrease their extensive utilisation for this type of fuel cells. At last, platinum-group-metal-free (PGM-free) catalysts based on nitrogen coordinated with transition metals such as Mn, Fe, Co and Ni have captured the attention of scientists worldwide [39] . The addition of a small quantity of these catalysts can boost the performance importantly without affecting heavily the overall cost [40] . Focusing on the cathode utilised for this experimentation, cost might be reduced without influencing negatively the output by decreasing the AC/PTFE loading on the current collector. Generally, the advancement in electrode materials properties certainly would lead to an increase in performance output but it would certainly also increase the overall cost of the system. Therefore critical attention has to be devoted on this topic probably exploring a detailed cost-benefit analysis."
} | 7,603 |
39883687 | PMC11781620 | pmc | 8,288 | {
"abstract": "Poly-gamma-glutamic acid (γ-PGA) is mainly synthesized by glutamate-dependent strains in the manufacturing industry. Therefore, understanding glutamate-dependent mechanisms is imperative. In this study, we first systematically analyzed the response of Bacillus subtilis SCP017-03 to glutamate addition by comparing transcriptomics and proteomics. The introduction of glutamate substantially altered gene expression within the central metabolic pathway of cellular carbon. Most genes in the pentose phosphate pathway (PPP), tricarboxylic acid (TCA) cycle, and energy-consuming phase of the glycolysis pathway (EMP) were down-regulated, whereas those in the energy-producing phase of glycolysis and those responsible for γ-PGA synthesis were up-regulated. Based on these findings, the fermentation conditions were optimized, and γ-PGA production was improved by incorporating oxygen carriers. In a batch-fed fermentor with glucose, the γ-PGA production reached 95.2 g/L, demonstrating its industrial production potential. This study not only elucidated the glutamate dependence mechanism of Bacillus subtilis but also identified a promising metabolic target for further enhancing γ-PGA production.",
"introduction": "Introduction Poly-gamma-glutamic acid (γ-PGA) is a biopolymer naturally synthesized by microorganisms [ 1 ]. It possesses exceptional biochemical characteristics, including water absorption, moisture retention, and biocompatibility. As a novel, environmentally friendly biomaterial [ 2 ], it has extensive applications in various fields, including food, cosmetics, agriculture, medicine, and environmental fields [ 3 , 4 ]. Therefore, γ-PGA biosynthesis has garnered considerable attention. γ-PGA-producing strains can be categorized into two groups: glutamate-dependent and glutamate-independent [ 5 ]. The former requires an external supply of glutamate to generate γ-PGA, and includes strains such as Bacillus subtilis NX-2 [ 5 ], B . licheniformis WX-02 [ 6 ], B . subtilis chungkookjang [ 7 ], and B . subtilis GXA-28 [ 8 ]. In contrast, glutamate-independent strains can produce γ-PGA without the addition of exogenous glutamate, synthesizing it de novo from carbon sources. Examples include B . amyloliquefaciens LL3 [ 9 ], B . subtilis C10 [ 10 ], and B . licheniformis GXG-5 [ 11 ]. Notably, glutamate-dependent strains exhibit higher γ-PGA production than glutamate-independent strains, making them the primary choice for industrial γ-PGA production [ 5 ]. The microbial γ-PGA synthesis pathway consists of three primary stages: precursor synthesis, polymerization transfer, and catabolism. In B . subtilis , key metabolic pathways, such as glycolysis (EMP), pentose phosphate pathway (PPP), tricarboxylic acid (TCA) cycle, and amino acid metabolism, play essential roles in γ-PGA biosynthesis [ 12 ]. L- glutamate, as a precursor, can be converted into D-glutamate through multiple enzymatic routes [ 13 ], as identified in B . subtilis IFO 3336 [ 14 ]. These L-glutamate and D-glutamate molecules undergo polymerization facilitated by the polymerization, a process that requires ATP consumption. The polyglutamic acid synthase (Pgs) is encoded by the pgs operon and encompasses four genes: pgsB , pgsC , pgsA , and pgsE in B . licheniformis and B . amyloliquefaciens [ 13 , 15 ], with homologs in B . anthracis ( capB , C , A , and E ), and B . subtilis ( ywsC , ywtABC ) [ 16 – 18 ]. In addition, the gene encoding γ-PGA hydrolase (PgdS) is located downstream of the pgs operon in the B . subtilis (natto) genome [ 13 ]. Specifically, PgsB is an ATP-dependent amino ligase that catalyzes γ-PGA formation [ 18 , 19 ]. PgsC is a cell membrane component of the γ-PGA synthesis system, featuring a structural resemblance to the N-acetyltransferase domain of N-acetylglutamic acid synthase [ 20 ]. PgsA contains a cell membrane-anchoring region [ 2 ] that may have implications for γ-PGA synthesis or transport [ 21 , 22 ]. The regulatory mechanisms governing γ-PGA biosynthesis remain incompletely understood, with limited research in this area. Zeng et al. compared the genomic differences between glutamate-dependent and glutamate-independent strains, and identified 13 genes related to γ-PGA biosynthesis that were mutated in glutamate-dependent strains. However, the relationship between these mutations and glutamate dependence has not yet been conclusively established [ 11 ]. Sha et al. explored the dependence mechanism of B . subtilis NX-2 on glutamic acid during γ-PGA production through transcriptome analysis. Their findings revealed that the addition of glutamate significantly up-regulated genes associated with glycolysis, PPP, TCA cycle, glutamic acid synthesis, and γ-PGA synthesis. Overexpression of these genes causes the accumulation of γ-PGA, demonstrating the pivotal role of intracellular glutamic acid synthesis in regulating γ-PGA production in glutamate-dependent strains [ 23 ]. Li et al. preliminarily investigated the co-production mechanism of γ-PGA and nattokinase using transcriptomic technology. Their study revealed the up-regulation of genes related to carbohydrate metabolism. Furthermore, by analyzing major metabolic pathways, such as carbohydrate metabolism, potential target genes for enhancing γ-PGA production were identified [ 24 ]. Previous studies have shown that γ-PGA biosynthesis is regulated by two intracellular signal transduction mechanisms. ComA is a DNA-binding transcription factor and a corresponding regulator of the two-component ComP-ComA system. At high cell densities, the cell density signal of ComX is transferred from phosphorylated ComP to phosphorylated ComA. This event induces the expression of DegQ, which transfers the cell density signal to the DegS-DegU two-component system. This can promote DegU phosphorylation, and the combination of phosphorylated DegU with the pgsBCA gene promoter activates its expression [ 25 , 26 ]. Phosphorylated DegU directly engages the pgsBCA promoter, thereby activating gene expression. Consequently, knockout of DegU and SwrA results in the loss of γ-PGA synthesis capability [ 27 , 28 ]. SwrA and DegU facilitate the expression of the pgsBCA operon through protein-protein interactions, serving as key factors in activating its expression. However, the relationship between these regulatory factors and the addition of glutamate remains unclear [ 13 , 23 , 29 , 30 ]. To elucidate the regulatory mechanisms governing γ-PGA biosynthesis, this study focused on the glutamate-dependent strain B . subtilis SCP017-03. The effects of exogenous glutamate supplementation on γ-PGA production and cell growth were investigated. Moreover, a comparison and analysis of the differential expression of bacteria at the transcription and protein levels in the presence and absence of glutamate was conducted using transcriptomics and proteomics technologies. This analysis focused on the differential expression observed in glycolysis, tricarboxylic acid cycle, pentose phosphate pathway, glutamic acid metabolism, and γ-PGA synthesis pathway, resulting from the addition of glutamate. Based on these findings, fermentation conditions were optimized, with a significant improvement in γ-PGA production. This study provides a comprehensive and systematic understanding of the molecular mechanisms underlying glutamate dependence in various γ-PGA strains. Furthermore, this study offers a valuable theoretical foundation for future gene modification endeavors to achieve high-yield γ-PGA production.",
"discussion": "Discussion Currently, industrial production of γ-PGA primarily relies on glutamate-dependent strains, necessitating the addition of exogenous sodium glutamate for γ-PGA synthesis. Therefore, thorough investigation of the mechanisms underlying glutamate dependence is crucial. However, research in this area has been limited. In this study, the response of strain SCP017-03 to glutamate addition was investigated using transcriptomics and proteomics. In contrast to the findings of Sha et al. [ 23 ], which nearly all genes in the three central metabolic pathways were significantly up-regulated, our transcriptome results showed a significant down-regulation in the expression of coding genes in the PPP and TCA cycle pathways. Conversely, most coding genes in the EMP pathway were up-regulated. This disparity was attributed to decreased dissolved oxygen in the culture medium caused by the highly viscous γ-PGA product. This decreased the expression of genes encoding the TCA and PPP pathways. However, most coding genes in the glycolytic pathway were up-regulated, promoting pyruvate accumulation and directing metabolism towards the fermentation product, lactic acid. The significant up-regulation of dehydrogenase-encoding genes upon glutamic acid addition further validated this hypothesis. Building upon this analysis, subsequent experiments in this study enhanced γ-PGA production by increasing the dissolved oxygen levels in the culture medium. Interestingly, the pgsBCA gene cluster responsible for γ-PGA synthesis was present in the genome of the control group without exogenous glutamic acid, with FPKM values reaching 859, 744, and 1243, respectively, although γ-PGA synthesis was not observed. This suggested that the lack of sufficient glutamic acid as a precursor within the cells and protein-level regulation may be contributing factors. In contrast, the experimental group supplemented with exogenous glutamate exhibited significant up-regulation in the expression of the pgsBCA gene cluster, with log 2 FC values of 2.63, 2.83, and 2.78, respectively, which is consistent with the findings of Sha et al. [ 23 ]. In summary, the response mechanism to glutamate addition varied among different γ-PGA-producing strains, highlighting the complexity of this phenomenon. The gene encoding the γ-PGA degrading enzyme pgdS is located downstream of the pgsBACE operon. PgdS is secreted into the extracellular space, and facilitates γ-PGA degradation. Additionally, previous studies have shown that γ-glutamyltranspeptidase (GGT) and DL-endopeptidase (CwlO) are also related to the degradation of γ-PGA [ 36 – 38 ]. In this study, observations were made in both the control group (CK), lacking glutamate, and the experimental group (GA), with glutamate supplementation. pgdS exhibited minimal expression in both groups, whereas the expression of cwlO remained low, with FPKM values of 59 and 282, respectively. However, the expression level of ggt decreased, with its FPKM dropping from 1231 to 266. Feng et al. conducted individual knockout experiments of the pgdS , ggt , and cwlO genes in glutamate-independent strains. γ-PGA production improved only when the cwlO gene was singly knocked out, and further enhancements were observed through simultaneous knockout of the pgdS and ggt genes [ 39 ]. Similarly, Scoffone et al. reported doubling of γ-PGA production after double knockout of pgdS and ggt in glutamate-dependent strains [ 40 ]. Furthermore, overexpression of pgdS efficiently biosynthesizes low-molecular-weight γ-PGA [ 41 ]. Proteomic analysis in this study revealed that genes in the energy consumption stage of the glycolytic pathway exhibited a slight down-regulation, whereas genes associated with the energy production stage were up-regulated. Furthermore, the PPP and TCA cycle pathways displayed noticeable down-regulation, aligned perfectly with the metabolic pathway responses observed in transcriptome sequencing following glutamate supplementation. Additionally, dynamic changes in the expression of key genes involved in γ-PGA synthesis were analyzed using qPCR at various fermentation time points. The results indicated that during the initial stage of fermentation (10 h), genes from all central metabolic and γ-PGA synthesis pathways were up-regulated. As fermentation progressed (20 h and 48 h), the genes associated with γ-PGA synthesis continued to display up-regulation, while those related to central metabolic pathways exhibited a declining trend. In summary, for glutamic acid-dependent bacteria, the addition of glutamate had a dual purpose: activating γ-PGA synthetase expression and providing the ample supply of glutamate necessary for γ-PGA synthesis. The mechanism underlying the high yield of γ-PGA is complex and encompasses aspects such as glutamate supply, γ-PGA synthesis, transport, secretion, energy metabolism, and glutamate metabolism. In this study, qPCR analysis revealed that in the experimental group supplemented with glutamate, the expression of the pgsBCA gene cluster exhibited significant up-regulation throughout the 10th to 48th hours of fermentation, compared to the control group. Moreover, the genes degS and degU displayed substantial up-regulation at the 10th and 20th hours of fermentation, with a slight increase at the 48th hour, while degQ showed down-regulation. Transcriptome analysis indicated that the coding genes comX , comA , and swrA were expressed at low levels during the 20th hour of fermentation and were slightly down-regulated after glutamate addition. In parallel, proteomic analysis demonstrated a significant up-regulation of comP expression at the sixth hour of fermentation. Moreover, the quorum sensing system (QS system), which is known to influence γ-PGA synthesis by primarily affecting comA expression and indirectly regulating γ-PGA synthesis, has been considered [ 42 , 43 ]. The results indicated significant down-regulation of the coding genes for Spo0K permease ( oppABCDF ) and phosphatase RapC ( rapHCFK ), whereas the expression of comA was slightly down-regulated. This suggested that comA may be regulated by other regulatory factors or genes. In the 5-L fermentor, 50 g/L of sodium glutamate was initially added to the culture medium, and 31.6 g/L remained at the end of fermentation. Meanwhile, the γ-PGA output reached 92 g/L, indicating that the glutamate supplied for γ-PGA synthesis primarily originated from de novo synthesis through carbon catabolism, rather than from exogenous glutamate supplementation. This differs from the findings of Sha et al. [ 23 ], in which γ-PGA was derived mainly from exogenous glutamate. Therefore, the main role of exogenous glutamate appears be to stimulate the expression of genes related to γ-PGA synthesis in the strain SCP017-03. Additionally, several regulatory factors including DegU, DegQ, DegS, ComP-ComA, and SwrA exhibited differential expression. Based on a comprehensive analysis of strain SCP017-03 at the physiological, protein, and transcription levels, along with the optimization of fermentation medium composition and enhanced dissolved oxygen conditions, several key findings emerged. Glucose and yeast extract were identified as the optimal carbon and nitrogen sources, respectively, which increased the production of γ-PGA from 36.5 g/L to 40.12 g/L. However, the accumulation of high viscosity γ-PGA in fermentation broth may hinder oxygen transfer, thereby negatively affecting cell growth and γ-PGA production [ 5 ]. Various strategies have been proposed to address this issue. For instance, Su et al. integrated the vgb gene (encoding hemoglobin) into the genome of a γ-PGA-producing strain [ 44 ] to augment dissolved oxygen levels. Zhang et al. successfully increased γ-PGA production to 39.4 g/L by introducing four different organic oxygen carriers into the fermentation medium [ 5 ]. Feng et al. proposed an aerobic plant cellulose experimental bed to enhance γ-PGA production [ 45 ]. In this study, we focused on promptly mitigating dissolved oxygen deficiency in fermentation cultures and enhancing γ-PGA production by optimizing oxygen carriers. The results demonstrated a significant increase in γ-PGA production by introducing n-heptane during a 24-hour fermentation period, increasing the production from 40.12 g/L to 44.3 g/L. Furthermore, in a 5 L fermentor, the production reached an impressive 95.2 g/L after batch feeding with glucose. In summary, this study presented a pioneering integration of transcriptome and proteome analyses, revealing the glutamate-dependent mechanism underlying γ-PGA synthesis in B . subtilis . Through a comparative analysis of strains cultured with and without glutamate, key genes involved in glycolysis, PPP, TCA cycle, and γ-PGA synthesis pathway were identified. This exploration revealed the potential factors and metabolic pathways crucial for γ-PGA production, thereby providing an initial framework for understanding the production mechanism of γ-PGA. Subsequent research focused on fermentation optimization, resulting in a remarkable output of 95.2 g/L with the introduction of an oxygen carrier, demonstrating promising prospects for industrial-scale production. The findings of this study offer a valuable metabolic target for further enhancement of γ-PGA production, with substantial significance for large-scale production of targeted metabolites."
} | 4,251 |
35722267 | PMC9205306 | pmc | 8,289 | {
"abstract": "The study of above- and below-ground organ plant coordination is crucial for understanding the biophysical constraints and trade-offs involved in species’ performance under different environmental conditions. Environmental stress is expected to increase constraints on species trait combinations, resulting in stronger coordination among the organs involved in the acquisition and processing of the most limiting resource. To test this hypothesis, we compared the coordination of trait combinations in 94 tree seedling species from two tropical forest systems in Mexico: dry and moist. In general, we expected that the water limitation experienced by dry forest species would result in stronger leaf-stem-root coordination than light limitation experienced by moist forest species. Using multiple correlations analyses and tools derived from network theory, we found similar functional trait coordination between forests. However, the most important traits differed between the forest types. While in the dry forest the most central traits were all related to water storage (leaf and stem water content and root thickness), in the moist forest they were related to the capacity to store water in leaves (leaf water content), root efficiency to capture resources (specific root length), and stem toughness (wood density). Our findings indicate that there is a shift in the relative importance of mechanisms to face the most limiting resource in contrasting tropical forests.",
"conclusion": "Conclusions We found that in the two forests we studied, which differ in precipitation and seasonality, the level of coordination among leaves, shoots, and roots in seedlings was similar, but the most functionally connected traits were different. In the dry forest, the most central traits were all related to water storage (LWC, SWC, RTh), while in the moist forest they were related to the capacity to store water in leaves (LWC), root efficiency to capture resources (SRL), and stem toughness (WD). Our findings suggest that, along with precipitation, there is a shift in the relative importance of mechanisms to face the most limiting resource. In the dry forest, this is the water storage capacity, soil vertical foraging, and water exploitation-drought tolerance trade-offs. In the moist forest, the growth-survival trade-off is most important. However, further studies of leaf, stem, and root coordination including different ontogenetic stages and multiple sites over environmental gradients are needed to clarify whether plants respond to limiting resources under a “whole plant” strategy, or whether limiting resources or phylogenetic constraints act on different plant organs independently.",
"introduction": "Introduction The different features of an organism do not represent stochastically independent dimensions, but rather they are correlated with one another through the interplay of genetic constraints and selective pressures ( Pavlicev, Cheverud & Wagner, 2009 ). Moreover, the evolutionary maximization of one function is frequently attained by minimizing another, which is described as a functional trade-off. When plant traits are consistently correlated among species, they form axes or dimensions of trait variation ( Wright et al., 2006 ), in which traits that are functionally or developmentally related to each other evolve in a coordinated fashion in response to selective pressures ( Pavlicev, Cheverud & Wagner, 2009 ; Machado et al., 2019 ). For example, at a global scale, plants are differentiated by their traits along a trade-off between rapid acquisition vs conservation of resources ( Reich, 2014 ). Variation in functional traits reflects the adaptation of organisms to their abiotic environment. This idea has been tested by estimating the relationship between traits and some fitness components, such as survival or reproduction, which has shown a much stronger selective effect of the physical environment than biotic factors ( Caruso, Maherali & Martin, 2020 ). Consequently, the relationships between specific plant functional traits with environmental conditions have demonstrated that certain characteristics have potential adaptive value. These include, for example, increased photosynthetic capacity and leaf nitrogen content ( Vogan & Maherali, 2014 ), rooting depth ( Schenk & Jackson, 2002 ), deciduousness ( Palomo-Kumul et al., 2021 ) and cavitation resistance in dry environments ( Maherali, Pockman & Jackson, 2004 ), and specific leaf area, leaf dry matter content ( Poorter, 2009 ), leaf area, and plant height ( Guerin et al., 2022 ) in mesic environments. Although it is clear that plant strategies exist and respond to environmental selective pressures, it is currently debated whether this plant response occurs at an organismic scale, with their different organs tightly converging functionally ( Grime, 2006 ; Reich, 2014 ), or whether each organ responds individually, resulting in independent axes of functional variation ( Baraloto et al., 2010 ). Some studies have found coordination ( i.e. , trait correlation or covariation) among root, stem, and leaf traits, supporting the idea that species have diversified across ecological strategies in response to environmental gradients resulting in a whole plant strategy ( de la Riva et al., 2016 ; Freschet et al., 2010 ; Ávila Lovera et al., 2022 ). However, others have found that different tissues have diversified under independent selective pressures, or respond to independent phylogenetic constraints, resulting in decoupled organs ( Fortunel, Fine & Baraloto, 2012 ; Valverde-Barrantes, Smemo & Blackwood, 2015 ; Bowsher et al., 2016 ; Wang et al., 2017 ). Thus, there is still no consensus. One possibility is that such a discrepancy may result from limited empirical evidence, and the fact that leaf and stem traits have been more extensively studied than root traits, leaving the relationships between root traits and their functions poorly understood ( Freschet et al., 2021 ). Tropical forests have been used as models to test plant coordination hypotheses because they have high species and functional diversity ( Vleminckx et al., 2021 ). While strong coordination among leaf, stem and root traits has been found in dry forests ( Markesteijn et al., 2011a ; Méndez-Alonzo et al., 2012 ), but see ( Silva et al., 2018 )), studies in moist forests have shown leaves and roots to be independent axes of variation ( Fortunel, Fine & Baraloto, 2012 ; Vleminckx et al., 2021 ). Together, these findings suggest the theoretical expectation proposed by Dwyer & Laughlin (2017) that under less restricted conditions, many possible conformations of leaf, stem, or roots are biophysically possible and ecologically successful ( Dwyer & Laughlin, 2017 ; Santiago et al., 2018 ). In addition, if coordination among traits results from natural selection purging uncoordinated variants, correlations would be expected mainly among the traits with a higher selective value in a given environment (not among any possible trait combination). Therefore, assessing the coordination of traits related to potentially relevant functions under contrasting environmental conditions could improve our understanding of functional coordination and its role in maintaining biodiversity. In this study, we assessed whether coordination between above ground (leaf and stem) and below ground (roots) organs differ between seedlings of species from a dry versus a moist tropical forest in Mexico. The two forests differ in their most limiting resource: water in the dry forest, and light in the moist forest. We hypothesized that environmental restriction of a particular resource could result in stronger coordination among traits involved in the acquisition and processing of that resource ( i.e. , the traits under the strongest selective pressure). Specifically, we expected that in the dry forest, where drought pulses impose a strong selective pressure for fine-tuned synchronization among above ground and below ground organs to acquire, transport, and use water, there would be stronger above–below ground trait coordination ( Markesteijn et al., 2011a ; Méndez-Alonzo et al., 2013 ; Méndez-Alonzo et al., 2012 ; Paz, Pineda-García & Pinzón-Pérez, 2015 ; McCulloh et al., 2019 ). In the shaded, moist forest, water is available nearly year-round, so we expected weaker coordination between leaves and roots ( Fortunel, Fine & Baraloto, 2012 ). However, we expected stronger coordination among aerial traits related to light capture and use, which depend strongly on leaf biochemistry ( Wright et al., 2004 ), leaf angle, branching pattern, and other aerial traits ( Valladares, Skillman & Pearcy, 2002 ). To test these hypotheses, we took advantage of the extensive knowledge of the ecology and functional strategies of seedlings in both forests and analyzed patterns of leaf, stem and root traits coordination among 94 tree-seedling species by means of multivariate correlation and network analysis.",
"discussion": "Discussion Coordination is similar between forests, but trait relationships are not We hypothesized that water shortage in the dry forest selects for stronger coordination among leaf, stem, and root traits, while shade in the understory of the wet forests favors strong coordination only between leaf and stem traits. However, we found no support for such hypothesis; in both forests we detected modules that included traits from all three organs. This is similar to findings by Flores-Moreno et al. (2019) , who also found high connectivity across trait networks within and between tissue types in both tropical and temperate forests. A novel contribution from our study is the use of network analysis to study inter-trait coordination not only between stem and leaf traits, but also with root traits. Interestingly, the finding that the pattern of coordination of key traits was different between the dry and the moist forest supports the hypothesis that the selective value of the same traits differs under different conditions ( Dwyer & Laughlin, 2017 ; Flores-Moreno et al., 2019 ). For example, in the moist forest we found weak coordination between the morphological efficiency to capture resources above ground and below ground ( i.e. , direct correlation between SLA-SRL, Fig. 2 ), but these traits were not correlated in the dry forest. The existence of a growth-defense trade-off has been well described among moist forest species ( Poorter & Bongers, 2006 ); fast growth in gaps is promoted by cheap, short-lived, and physiologically active leaves (indicated by high SLA), while high survival in the forest understory is enhanced by the formation of long-lived, tough leaves that reduce herbivory, mechanical damage, or leaf turnover ( Poorter & Bongers, 2006 ). The coordination between SRL and SLA detected in our study ( Figs. 2 and 3 ), suggests that the growth-defense trade-off involves both above- and below-ground morphological traits. Species with high SLA that are adapted to rapidly acquire light, seem to also have roots that are capable of absorbing soil resources efficiently. Meanwhile, species with high investment in structural defense of leaves (low SLA), tend to also have structurally well-defended roots (low SRL). However, the weak SLA-SRL correlation ( r = 0.38, p = 0.01) and the inconsistency of the relationship between these traits in published literature ( Weigelt et al., 2021 ) may be related to the fact that SRL depends on root density (RD) and root thickness (RTh), and plants can construct roots with many combinations of these values. In addition, RTh is also related to symbiosis with mycorrhizal fungi. Consequently, roots vary not only along a conservation-acquisition trade-off, but also, and orthogonally, along a collaboration-do it yourself trade-off ( Bergmann et al., 2020 ). The role played by symbiosis in root traits (SRL and RTh) needs further investigation in this moist forest. The lack of SLA-SRL coordination in the dry forest may also reflect the fact that the development of dense tissues with high carbon investment is not necessarily the predominant strategy to deal with drought. Commonly, drought-deciduous species have short-lived, low-cost leaves with high SLA, coupled with low-density stems and thick root tissues containing high water and carbohydrate reserves ( Powers & Tiffin, 2010 ; Pineda-García, Paz & Tinoco-Ojanguren, 2011 ; Palomo-Kumul et al., 2021 ). Conversely, maximum root depth (MRD) was coordinated with leaf and stem traits in the dry forest, but not in the moist forest ( Fig. 3 ). The relevance of MRD in this and other dry forests is not surprising, given the importance of this trait for seasonal water uptake in seedlings in arid environments ( Ackerly, 2004 ; Maeght, Rewald & Pierret, 2013 ), especially during drought periods ( León et al., 2011 ; Padilla & Pugnaire, 2007 ). Furthermore, the coordination of dense stems with deep roots in the dry forest ( Figs. 2 and 3 ) highlights the importance of a more permanent source of water when stem storage capacity is low ( Schwinning & Ehleringer, 2001 ; Hasselquist, Allen & Santiago, 2010 ; Paz, Pineda-García & Pinzón-Pérez, 2015 ). Likewise, thick roots (RTh) were directly related with high SRL and leaf water content (LWC) in the dry forest (and inversely in the moist forest), an indication that RTh may be related to water economy in the dry forest. Commonly in dry forests, roots store important amounts of water in their tissues ( Paz, Pineda-García & Pinzón-Pérez, 2015 ), while in moist forests thick roots are commonly fibrous. Although our moist forest seedlings were obtained directly from the field while dry forest seedlings were grown in pots, we are confident that this did not affect root trait measurements, particularly MRD. First, in our study of dry forest plants, we rarely observed roots reaching the bottom or lateral edges of the pots, indicating the soil volume did not impede root growth in a specific direction. Second, in a previous study where MRD was measured in four neotropical forests with the same methodology, extracting seedlings from the ground, roots were deeper in the site with the longest dry season ( Paz, 2003 ), similar to our study. Trait centrality When considering the traits’ centrality values (the relative importance of each trait as a connecting node), we found that LWC was the most important trait in both forests ( Fig. 4 ), reflecting the key role of leaf hydration in plant growth and many other physiological processes, regardless of forest type. Previous studies have proposed that in addition to leaf turgor, maintenance of leaf water content may be critical for growth processes such as cell elongation and division, as well as for leaf gas exchange ( Bartlett, Scoffoni & Sack, 2012 ). In addition, leaf water content has been found to be associated with hydraulic traits at the stem level, such as stem water content and stem conductivity ( Pineda-García et al., 2015 ) and is a good predictor of growth and survival under dry conditions in other tropical forests ( Cifuentes et al., 2020 ). The frequent correlations of wood density with hydraulic and leaf traits in adult trees ( Santiago et al., 2004 ; McCulloh et al., 2011 ; Greenwood et al., 2017 , among others) have led to the idea that wood density could represent a central trait affecting the stem and leaf economy ( Chave et al., 2009 ; McCulloh et al., 2011 ; Méndez-Alonzo et al., 2013 ). However, strikingly, in our study we detected strong evidence suggesting that among tropical seedlings, LWC may be another key functional trait; this is an idea worth testing in other tropical forests. Interestingly, SWC was the second most important trait in the dry forest, but not in the moist forest. This may be explained by the close association of SWC with the water storage capacity versus soil vertical foraging and water exploitation versus drought tolerance trade-offs described in tropical seasonally dry forest ( Paz, Pineda-García & Pinzón-Pérez, 2015 ; Pineda-García et al., 2015 ). Drought avoiders, which have high photosynthetic rates, xylem hydraulic conductivity, and growth rate when water is available ( Pineda-García et al., 2015 ), maintain a narrow safety margin between plant water potential and P 50 (the potential that would induce 50% of hydraulic conductivity loss by the formation of emboli) ( Pineda-García, Paz & Meinzer, 2013 ; Markesteijn et al., 2011b ). Given that these species tend to have shallow roots, they have evolved a great capacity to store water in the stem and leaves, which allows them to survive as the soil desiccates during the dry season ( Paz, Pineda-García & Pinzón-Pérez, 2015 ). On the contrary, drought-tolerant species, which have lower photosynthetic rates, xylem hydraulic conductivity, and growth rate, have a denser stem with limited water storage capacity (low SWC) ( Pineda-García et al., 2015 ; McCulloh et al., 2019 ). Due to their lower capacity to decouple hydraulically from the soil, these dense-tissue species are associated with deep roots that penetrate deeper into the soil and rock interstices and rely on a more constant, although unsaturated, soil water content ( Schwinning & Ehleringer, 2001 ; Padilla & Pugnaire, 2007 ; Zhou et al., 2020 ). Thus, differential responses to drought explain the importance of SWC, a trait strongly involved in water economy in the dry forest. As expected, SLA had very low centrality in the dry forest ( Fig. 4 ), where competition for light is likely not as strong as competition for water, and where SLA is strongly related to non-morphological traits such as leaf phenology. However, the low centrality of SLA in the moist forest is intriguing. This lack of finely tuned connection between a central trait in the leaf economy spectrum and other leaf and stem traits suggests that different combinations of leaf traits and plant architectures can yield similar capacities for growth in shaded forest ( Valladares, Skillman & Pearcy, 2002 ). This hypothesis needs further investigation in our study system. Contrary to expectation, SRL, a trait typically claimed to be a key morphological determinant of the efficiency of carbon investment in water absorption, was poorly connected in the dry forest, but highly connected in the wet forest ( Fig. 4 ). Although it is possible that this was due to differences in soil nutrients between forests, this seems unlikely since nutrient levels overlap strongly between the study sites. Our field observations suggest that in the dry forest, high values of SRL may be indicative of different functions depending on the species: fine roots deploying large absorptive surfaces, or thick roots with high water and low carbon contents acting as water storage more than for water absorption. Together, the lack of SRL centrality and the clear correlations of MRD with leaf and stem traits in the dry forest suggest that in habitats with a high risk of drought at the seedling stage, developing deep roots has a higher selective value than developing thin, efficiently absorptive roots ( Padilla & Pugnaire, 2007 ; León et al., 2011 ; Paz, Pineda-García & Pinzón-Pérez, 2015 ). Conversely, in the moist forest, in the context of strong competition for light, SRL and WD are important under a growth-survival trade-off, where species that acquire soil resources efficiently grow fast (high SRL) and have low-density stems (low WD), and vice versa ( Poorter, 2009 ; Pineda-García et al., 2015 )."
} | 4,877 |
18183306 | PMC2173943 | pmc | 8,291 | {
"abstract": "Background Diatoms are unicellular algae responsible for approximately 20% of global carbon fixation. Their evolution by secondary endocytobiosis resulted in a complex cellular structure and metabolism compared to algae with primary plastids. Methodology/Principal Findings The whole genome sequence of the diatom Phaeodactylum tricornutum has recently been completed. We identified and annotated genes for enzymes involved in carbohydrate pathways based on extensive EST support and comparison to the whole genome sequence of a second diatom, Thalassiosira pseudonana . Protein localization to mitochondria was predicted based on identified similarities to mitochondrial localization motifs in other eukaryotes, whereas protein localization to plastids was based on the presence of signal peptide motifs in combination with plastid localization motifs previously shown to be required in diatoms. We identified genes potentially involved in a C4-like photosynthesis in P. tricornutum and, on the basis of sequence-based putative localization of relevant proteins, discuss possible differences in carbon concentrating mechanisms and CO 2 fixation between the two diatoms. We also identified genes encoding enzymes involved in photorespiration with one interesting exception: glycerate kinase was not found in either P. tricornutum or T. pseudonana . Various Calvin cycle enzymes were found in up to five different isoforms, distributed between plastids, mitochondria and the cytosol. Diatoms store energy either as lipids or as chrysolaminaran (a β-1,3-glucan) outside of the plastids. We identified various β-glucanases and large membrane-bound glucan synthases. Interestingly most of the glucanases appear to contain C-terminal anchor domains that may attach the enzymes to membranes. Conclusions/Significance Here we present a detailed synthesis of carbohydrate metabolism in diatoms based on the genome sequences of Thalassiosira pseudonana and Phaeodactylum tricornutum. This model provides novel insights into acquisition of dissolved inorganic carbon and primary metabolic pathways of carbon in two different diatoms, which is of significance for an improved understanding of global carbon cycles.",
"introduction": "Introduction Diatoms are abundant unicellular algae in aquatic habitats. They can produce enormous amounts of biomass and are thought to be responsible for about 20% of global carbon fixation. As much as 16 gigatons of the organic carbon produced by marine phytoplankton per year, or about one third of total ocean production is thought to sink into the ocean interior preventing re-entrance of this carbon into the atmosphere for centuries [1] . Recent assessments suggest that diatom-mediated export production can influence climate change through uptake and sequestration of atmospheric CO 2 \n [2] , [3] . The role diatoms play in mitigating atmospheric CO 2 concentrations is of special interest now with the rising levels of this “greenhouse gas” and consequent global warming. A significant fraction of the organic carbon generated by diatoms remains in the upper ocean and supports production by higher trophic levels and bacteria. Despite the important role of diatoms in aquatic ecosystems and the global carbon cycle, relatively little is known about carbon fixation and carbohydrate pathways in these algae [4] . For example the exact mode of CO 2 fixation is largely unsolved. Ribulose-1,5-bisphosphate carboxylase/oxygenases (Rubisco) from diatoms have half-saturation constants for CO2 of 30–60 µM [5] despite the fact that typical sea water contains about 10 µM CO 2 \n [6] . To prevent potential CO 2 limitation, most diatoms have developed mechanisms to concentrate dissolved inorganic carbon (DIC) via a CO2 concentrating mechanism (CCM) [7] . Although most of the Calvin cycle enzymes in diatoms are very similar to those in land plants, there are indications that they may be differently regulated by light [8] . Furthermore, some metabolic pathways appear to be missing altogether from diatoms [9] . Finally, there is only scarce information available on the localization, synthesis and storage of chrysolaminaran, the principle storage carbohydrate in diatoms. Diatoms may produce and secrete vast amounts of carbohydrates that play important roles in phototrophic biofilms, yet very little is known about synthesis and secretion of these carbohydrates. Research on diatoms advanced significantly with publication of the whole genome sequences of the centric diatom Thalassiosira pseudonana \n [10] and of expressed sequence tags (ESTs) from the pennate diatom Phaeodactylum tricornutum \n [11] . Recent availability of whole genome sequence and about 100,000 ESTs for Phaeodactylum tricornutum provide additional opportunities to understand unique physiological characteristics of diatoms. Together with new experimental resources such as genetic transformation, now feasible for several diatom species [12] – [14] and various laboratory-based studies of their physiology [8] , diatoms have become model photosynthetic representatives for non-green algae. Diatoms have an evolutionary history distinct from higher plants. Diatoms are eukaryotic chimeras derived from a non-photosynthetic eukaryote that domesticated a photoautotrophic eukaryotic cell phylogenetically close to a red alga [15] . After incorporation, the endosymbiont was successfully transformed into a plastid that retained a small plastid genome, but lost the nuclear and the mitochondrial genomes as distinct entities. In addition to the genetic consequences that resulted from extensive gene transfer events and genomic reorganization, secondary endocytobiosis also increased the complexity of diatom cell structure, with implications on physiology and biochemistry. A significant difference between diatom plastids and those of higher plants is that diatom plastids are surrounded by four rather than two membranes, the outermost of which is contiguous with the endoplasmic reticulum. This means that import of all nuclear-encoded plastid proteins and the exchange of metabolites like carbohydrates between the plastids and the cytoplasm must take place across four membranes. To accomplish this task, nuclear-encoded proteins imported into diatom plastids possess an N-terminal signal peptide that targets the protein first to the endoplasmic reticulum and a plastid localization peptide that targets the protein to plastid stroma [16] , [17] . Another striking difference between diatoms and green algae/land plants is their different nuclear and mitochondrial backgrounds because they arose from different host cells. We annotated genes involved in carbon acquisition and metabolism in the genome of the diatom P. tricornutum and compared these gene models to the only other diatom whole genome sequence of Thalassiosira pseudonana . The 5′-most ends of a majority of critical genes were identified based on EST support. This meant that N-terminal leader sequences could be predicted for most proteins and thus their targeting to different compartments within the cell. Here, we present a comprehensive model of the localization of enzymes and pathways involved in carbon assimilation and carbohydrate production and catabolism.",
"discussion": "Results and Discussion Structure of the genome and gene annotation Following publication of the draft Thalassiosira pseudonana Hasle & Heimdal (CCMP 1335) genome [10] , a majority of sequence gaps were closed at the Stanford Human Genome Center (SHGC; Stanford, CA, USA) and a new version of the genome sequence is now publicly available at http://genome.jgi-psf.org/Thaps3/Thaps3.home.html . A second diatom genome, from Phaeodactylum tricornutum Bohlin (CCAP1055/1), was subsequently sequenced and completed at the U.S. Department of Energy Joint Genome Institute (JGI, http://www.jgi.doe.gov/ , Walnut Creek, CA, USA) and SHGC, and is available publicly at http://genome.jgi-psf.org/Phatr2/Phatr2.home.html . In addition, 100,000 ESTs generated from P. tricornutum cells grown in 14 different conditions have been generated by Genoscope (Evry, France) and are available at http://www.biologie.ens.fr/diatomics/EST . Both genomes are approximately 30 Mb and contain between 10,000 and 11,500 genes. Assembly and annotation of the whole genome of P. tricornutum will be published separately (manuscript in preparation). Here we focus solely on those pathways involved in carbohydrate metabolism. In the following sections, we include protein IDs (Prot-ID) from version 2.0 ( P. tricornutum ) of the JGI sequence database in parentheses. See Table S1 for a list of annotated genes together with the Prot-IDs in T. pseudonana . Prediction of intracellular targeting Nuclear encoded proteins are translated in the cytosol and subsequently transported to their respective target locations. In most known cases an N-terminal targeting domain can send the proteins into the ER, mitochondria, plastids, the extracellular space or to other compartments. In land plants relatively similar transit peptides are used to target into plastids or mitochondria, making it sometimes difficult to predict the correct compartment. Mitochondrial import sequences in diatoms are similar to those in other eukaryotes. Diatom plastid presequences, however, differ significantly from those of land plants or green algae [16] . Diatom plastids are surrounded by four membranes, the outermost being studded with ribosomes and continuous with the endoplasmic reticulum (ER) [18] . Nuclear encoded plastid proteins of diatoms contain N-terminal bipartite presequences consisting of a signal peptide followed by a transit peptide-like domain. Such presequences are easily recognized due to an essential targeting motif with a characteristic signature at the signal peptide cleavage site [16] , [17] . CO 2 fixation: a biochemical (C4) or a biophysical CCM-like metabolism? The apparent photosynthetic affinity of diatoms for inorganic carbon (C i ) is considerably higher than expected based on the affinity of their Rubisco for CO 2 \n [5] . Extensive diatom blooms that occur during large iron fertilization experiments in high nutrient low chlorophyll regions of the oceans [19] , [20] suggest that diatoms are not CO 2 limited under natural oceanic conditions. Both results imply that diatoms possess efficient CO 2 concentrating mechanisms (CCM), although underlying mechanisms (either a biochemical C4 or a biophysical CCM, or both) are still controversial (see [3] , [4] ). Studies on the biochemistry of photosynthesis in the well-characterized marine diatom Thalassiosira weissflogii suggested that a C4-like pathway could exist whereby a C4 compound such as malate or OAA is decarboxylated, typically within the chloroplast, to deliver CO 2 to Rubisco [21] , [22] . The possibility of a C4-like pathway in the related species T. pseudonana was examined based on an in silico analysis of gene content [10] . The T. pseudonana genome appears to encode the enzymes phosphoenolpyruvate carboxylase (PEPC), phospoenolpyruvate carboxykinase (PEPCK) and pyruvate orthophosphate dikinase (PPDK). Each of these enzymes is required for C4-metabolism, although they also play a role in C3-metabolism. Subsequent analysis of transcript abundances for the putative C4-related genes in T. pseudonana indicated that the gene encoding PEPCK was up-regulated about 1.5 fold under reduced CO 2 concentrations, whereas expression of genes encoding PEPC and PPDK were unaffected [3] . Despite the presence of typical C4 enzymes in both Thalassiosira species, short 14 CO 2 labelling experiments showed marked differences between them [4] . In T. weissflogii , about 30% of the 14 C label (in 5 sec. experiments) was observed in malate and about 40% in triose phosphates. In contrast, in T. pseudonana production of 14 C-labeled C4 products was negligible. Roberts et al. [4] concluded that a typical C3 metabolism occurs in T. pseudonana , despite the presence of C4 enzymes, whereas an intermediate C3-C4 may function in T. weissflogii . Genes essential for C4 metabolism were identified in P. tricornutum . A PPDK (21988), which catalyzes the formation of PEP, was identified and includes both a signal peptide and a putative plastid targeting sequence suggesting that PEP is generated in the plastid ( Fig. 1 ). Two genes encoding PEPC have been identified ( Fig. 1 ). The predicted protein sequence for one of them (PEPC1, 56026) has a high degree of identity (ca. 40% amino acid identity) with the PEPCs from green algae and higher plants. It possesses a signal peptide, but a plastidic transit peptide was not detected suggesting that this protein is targeted either to the ER or to the periplastidic space of the plastids [17] . A second PEPC (PEPC2, 20853) has high similarity (ca. 40% amino acid identity) to PEPCs from bacteria and contains a predicted mitochondrial targeting presequence. Decarboxylation of OAA appears to occur via a mitochondrial-localized PEPCK (23074). This enzyme has the greatest similarity to PEPCK from the proteobacterium Campylobacter jejuni (58% amino acid identity). Two additional decarboxylating enzymes belonging to the malic enzyme family were identified (27477, 56501) and apparently both possess a mitochondrial presequence. One of these enzymes (ME1) (56501) is characterized by a dinucleotide binding site that binds NAD rather than NADP. Thus, P. tricornutum and T. pseudonana appear to have two mitochondrial malic enzymes that are either NAD- or NADP-dependent. Genes encoding a mitochondrial-localized malate dehydrogenase (MDH) (51297), a pyruvate-kinase (PK6) (56172) and a pyruvate-carboxylase (PYC1) (30519) were also identified ( Fig. 1 ). The respective substrates for this pathway may be transported into the mitochondria by a putative mitochondrial oxoglutarate/malate transporter (8990). 10.1371/journal.pone.0001426.g001 Figure 1 Model of carbon concentrating mechanisms (CCM) in diatoms based on annotations of the Phaeodactylum tricornutum and Thalassiosira pseudonana genomes. For discussion of the pathways see text. Enzyme abbreviations: CA: carbonic anhydrase; MDH: malate dehydrogenase; ME1: NAD malic enzyme, mitochondrial; PEPC: phosphoenolpyruvate carboxylase; PEPCK: phosphenolpyruvate carboxykinase; PK: pyruvate kinase; PPDK: pyruvate-phosphate dikinase; PYC: pyruvate carboxylase; RUBISCO: ribulose-1,5-bisphosphate carboxylase. Models were developed to evaluate how such a C4-like carbon fixation pathway could operate in P. tricornutum (see our working scheme, Fig. 1 ). The first hypothesized step in carbon fixation is delivery of HCO 3 \n − into cells either via specific transporters or by diffusion of CO 2 and its subsequent conversion to HCO 3 \n − through CA activity (see below). The hypothesized localization of PEPC1 (56026) to the ER or to that part of the ER that is connected to the plastid (CER) or to the periplastidic space (PPS) suggests that subsequent fixation of HCO 3 \n − into a C4 compound likely occurs within either the ER or the periplastidic space. The localization of the C4 decarboxylation that delivers CO 2 to Rubisco for fixation is not clear. Immuno-localization based studies provided early evidence that the decarboxylating enzyme PEPCK is located in the plastids of the centric diatom Skeletonema costatum \n [23] . Later, Reinfelder et al. [21] found that PEPCK activity co-localized with Rubisco activity in isolated plastid-enriched fractions from T. weissflogii and concluded that decarboxylation occurred within the plastids. Subsequent in silico analysis of both P. tricornutum and T. pseudonana indicated that the decarboxylating enzymes PEPCK and malic enzyme do not possess plastid targeting sequences. Moreover, there is no evidence that malate and/or oxaloacetate transporters in these organisms are localized to plastid membranes. Finally, addition of oxaloacetate to intact plastids isolated from the diatom Odontella sinensis \n [24] does not result in net O 2 evolution as would be expected from the malic dehydrogenase reaction due to turnover of NADPH. Combined, these results suggest that subsequent decarboxylation steps required to generate CO 2 for Rubisco delivery, at least in P. tricornutum and T. pseudonana , do not occur in the plastid. Recent evidence with single chlorenchyma cells of the higher plants Bienertia cycloptera and Borszczowia aralocaspica provides support for a compartmentalized separation of CO 2 generation via decarboxylation of C4 compounds and subsequent CO 2 fixation by Rubisco [25] , [26] . In these plant cells, PPDK is located in chloroplasts where it converts pyruvate to PEP. The PEP is then transported to the cytosol where it is carboxylated (using HCO 3 \n − ) via PEPC. The C4 acids produced diffuse to the proximal part of the cell where they are decarboxylated in the mitochondria by NAD-malic enzyme. The resulting CO 2 may enter the chloroplasts where it is captured by Rubisco. Both sequenced diatoms possess two malic enzymes that decarboxylate malate to pyruvate. An NADP-malic enzyme has been proposed for diatoms by Granum et al. [3] . A potential NAD-dependent malic enzyme (56501) was also identified that is predicted to be localized to mitochondria and displays sequence similarity to malic enzymes from the C4 plants Amaranthus hypochondriacus \n [25] , B. cycloptera and B. aralocaspica \n [25] . The dinucleotide binding site of the malic enzyme from A . hypochondriacus , P. tricornutum and T. pseudonana possesses a similar amino acid composition suggesting that NAD is the preferred co-factor. These data suggest that in diatoms, decarboxylation of malate to generate CO 2 may occur within mitochondria, which are often closely associated with plastids. It is important to note however, that any CO 2 molecules released from the mitochondria must cross six membranes to enter the plastid stroma. Moreover, it is likely that CO 2 would be converted to HCO 3 \n − by CA activity during movement between the mitochondria and plastids, thereby reducing at least part of the elevated CO 2 concentration. In this case, the C4 pathway would become a futile cycle whereby HCO 3 \n − is first fixed and then formed again, thereby dissipating ATP for PEP formation. Co-occurrence of PEPCK- and malic enzyme-based decarboxylation pathways in the same organism was also observed in the C4-plant Urochloa panicoides \n [27] , [28] . Apparently, in diatoms both enzymes may contribute to the decarboxylation of the C4-acid. In some of the higher plants which perform C4 metabolism, the pyruvate formed by decarboxylation of malate, using the NAD-malic enzyme, can be used for amino acid synthesis ( Fig. 1 ) [28] or oxaloacetate formation thereby replenishing mitochondrial pools of C4 acids. Oxaloacetate can also be oxidized in the TCA cycle. Decarboxylation of OAA by PEPCK generates PEP, which can be used for gluconeogenesis or be transformed into pyruvate by a pyruvate kinase ( Fig. 1 ). The presence of a plastid-targeted putative PEP-transporter TPT1 (24610) and the plastid targeted PPDK (21988) suggests that pyruvate is phosphorylated inside the plastid to produce PEP. PEP can be used to produce aromatic amino acids (Shikimate pathway) and lipids or can be exported to provide the acceptor molecule for HCO 3 \n − fixation by PEPC ( Fig. 1 ). This pathway is also known from C4 plants that decarboxylate PEP inside the mitochondria (see [25] , [28] – [29] ). However, most C4-plants carry out the first carboxylation step in the cytosol [29] . In P. tricornutum and T. pseudonana PEP might be exported from the plastid and carboxylated at the ER/CER/PPC by PEPC1 leading to a micro-compartmentalization of the enzyme to the outer chloroplast membranes. Lee and Kugrens [30] have speculated that the evolutionary success of heterokonts might be partly attributed to the use of the periplastidic space as an acidic generator of CO 2 from bicarbonate. In summary, the accumulated evidence indicates that a functioning C4 pathway in diatoms ( Fig. 1 ) requires a spatial separation between CO 2 production via decarboxylation of OAA and malate in mitochondria, and CO 2 utilization by Rubisco in the plastid. Furthermore, localization of the first carboxylation step by PEPC is still unclear. The energetic cost of a futile cycle raises the possibility that the C4 metabolism may help cells dissipate excess light energy, a pathway that would presumably require down-regulation under energy-limiting conditions. This suggests that the flow of metabolites in this pathway would be affected by light intensity. Localization of key enzymes and determination of expression of C4 related genes in cells exposed to low levels of CO 2 could shed light on this issue. Two lines of evidence support the hypothesis that a biophysical CCM operates in diatoms, as in many other aquatic photosynthetic organisms [7] , [31] – [33] . First, we identified in the genome of P. tricornutum three genes that encode different systems for bicarbonate uptake. One protein (45656) shows homology to sodium/bicarbonate transporters in various organisms and appears to be localized to the plastid. A second protein (32359) is homologous to sodium-dependent anion exchangers and also possesses the bipartite presequence for plastid targeting. The third protein (54405) shows similarity to Cl − /HCO 3 \n − exchangers abundant in red blood cells. T. pseudonana appears to possess a single sodium/bicarbonate transporter (24021) that is targeted to the plastid. The hypothesis resulting from these observations is that inhibitors shown to prevent HCO 3 \n − uptake as shown in Ulva sp. [34] , should inhibit uptake of HCO 3 \n − and thereby the rate of photosynthesis in both diatoms. The second form of support for a biophysical CCM is identification of numerous genes encoding carbonic anhydrases (CA) in the diatom genomes. The overall sequence similarities among CAs are rather low and they are commonly identified by the presence of conserved domains and by their biochemical properties [35] . Seven CAs are predicted for P. tricornutum . Two CAs (51305, 45443) are related to the beta type and show similarity to CAs found in both plants and prokaryotes. Based on the presence of a plastid targeting presequence and physiological experiments [36] , one of these proteins (51305) is located in the plastid as has been demonstrated by GFP fusion proteins [37] . The other (45443) has a signal peptide. The five other identified CAs (35370, 44526, 55029, 54251, 42574) apparently belong to the alpha family and all possess signal peptides. A recent study by Szabo and Colman [38] provided experimental evidence for the presence of CA in the periplasmic space of P. tricornutum suggesting that at least a subset of the signal peptide-possessing CAs are likely targeted to the periplasmic space. Surprisingly, similarity of the CAs between the two diatoms T. pseudonana and P. tricornutum is rather low. T. pseudonana appears to possess more intracellular CAs without signal and transit peptides than P. tricornutum. One exception is the carbonic anhydrase 22391 from T. pseudonana which also possesses a signal peptide. Whether this CA is secreted or targeted to ER or periplastidic space remains to be investigated, while it seems clear that it is not plastid localized. The difference between the two diatoms may indicate specialization of the enzyme depending on different ecological niches. This is supported by recent findings that expression of beta carbonic anhydrases may be regulated by several factors including CO 2 and light [39] . Despite the extensive in silico analyses described here, the potential mechanism by which CO 2 is delivered to Rubisco remains elusive. In the well-studied green alga, Chlamydomonas , a thylakoid-located alpha CA facilitates conversion of HCO 3 \n − to CO 2 , thereby raising its concentration in close proximity to Rubisco [40] . Mutants impaired in this CA demand high CO 2 concentrations for growth (see [7] , [31] ). In P. tricornutum , the plastid-localized CA (51305) is a beta type CA rather than an alpha type. However, it too localizes to the thylakoids where it forms particles, most probably close to the girdle lamellae [37] , and its expression is strongly enhanced under low CO 2 conditions by a mechanism involving cAMP [41] . The other beta type CA (45443) is probably located in the ER or the periplasmic space and is constitutively expressed even under high CO 2 concentrations [42] . The presence of bicarbonate transporters in the chloroplast envelope ( Fig. 1 ) is consistent with the operation of a biophysical CCM but is not essential for a C4-like CCM, where the initial HCO 3 \n − fixation occurs outside the plastid. Finally, enhanced uptake of inorganic carbon, both as CO 2 or HCO 3 \n − , would be consistent with both types of CCM, whereas raising the concentration of C i within the cells [43] is more consistent with the biophysical CCM. The high affinity of PEPC for HCO 3 \n − would be expected to alleviate the need to accumulate high C i concentrations internally. Induction of an extracellular CA at low CO 2 concentrations was also observed in P. tricornutum . Such a CA would be expected to facilitate the rate of CO 2 formation in the unstirred layer surrounding the cells and thereby to supply CO 2 for photosynthesis by either CCM type. In summary, clear evidence that supports either CCM mode as the only way to raise the CO 2 concentration in close proximity of Rubisco is presently missing due to lack of sufficient biochemical evidence. Photorespiration and glyoxylate metabolism Photorespiration is the inevitable consequence of the ability of either CO 2 or O 2 to cleave the double bond obtained in RuBP after enolization by Rubisco. In higher plants, photorespiration is thought to provide the photosynthetic machinery with some protection against photoinhibition [44] – [47] . Mutants of tobacco ( Nicotiana tabacum L.) defective in enzymes of the photorespiratory pathway demonstrated enhanced photoinhibition under high light conditions [48] , [49] . The specificity factor (τ) of Rubisco, a measure of its ability to discriminate CO 2 from O 2 , is considerably higher in diatoms than in cyanobacteria and green algae (reviewed in [50] ) suggesting a lower rate of O 2 fixation in diatoms than observed in members of the green lineage. This is supported by studies showing photorespiratory activity in diatoms at a reduced rate than expected from studies with higher plants [51] – [54] . When O 2 out-competes CO 2 for RuBP, one molecule of 2-P-glycolate and one molecule of 3-P-glycerate are produced. The latter may enter the Calvin cycle, whereas 2-P-glycolate, a metabolite known to inhibit the Calvin cycle enzyme triosephosphate isomerase [55] must be degraded via the photorespiratory pathway (see Fig. 2 ). In higher plants, metabolism of 2-P-glycolate takes place via the C2 cycle [56] . Following cleavage of the phosphate group, glycolate is exported out of the chloroplast and enters the peroxisome where it is oxidized to glyoxylate, followed by transamination to form glycine which enters the mitochondrion. The glycine decarboxylase complex together with serine hydroxymethyltransferase, catalyzes the condensation of two glycines to one serine with the consequent release of ammonium ion and CO 2 . The serine is further metabolized back to P-glycerate which may then enter the Calvin cycle in the chloroplast. Thus, out of four carbons entering the C2 cycle, three are converted back to PGA and one is released in the form of CO 2 . 10.1371/journal.pone.0001426.g002 Figure 2 Model for photorespiration and associated pathways in diatoms based on the annotations of the Phaeodactylum tricornutum and Thalassiosira pseudonana genomes. For simplicity, the number of oragenelle membranes has been reduced in this figure. A gene model for glycerate kinase (GK) could not be found in either genome. The bacterial-type glyoxylate to glycerate metabolism is not shown due to uncertainty in the localization of the enzymes. Enzyme Abbreviations: ACS: acetyl CoA synthetase; CTS: citrate synthase; GDC: glycine decarboxylase; GOX: glycolate oxidase; GK: glycerate kinase; HPR: hydroxypyruvate reductase /glycerate dehydrogenase; ICL: isocitrate lyase; ME1: NAD malic enzyme; MLS: malate synthase; PDH: pyruvate dehydrogenase; PGP: 2-phosphoglycolate phosphatase; RUBISCO: ribulose-1,5-bisphosphate carboxylase; SHMT: serine hydroxymethyltransferase; SPT/AGT: serine-pyruvate/alanine-glyoxylate aminotransferase. In microalgae, the cyclic process of PGA recovery is not well-studied. More often, glycolate metabolism is studied in the context of its excretion as a waste-product to circumvent unfavourable growth conditions [57] – [60] . A large fraction of glycolate produced by fixation of O 2 is released from the cell [61] – [63] and may serve as an important source of organic carbon in the water body. Leboulanger et al. [64] found high concentrations of glycolate in seawater at both oligotrophic and eutrophic sites, suggesting that photorespiration may be ubiquitous in the marine environment. Photorespiration may therefore represent an important loss of fixed carbon, either via released glycolate or CO 2 . Recent studies on photorespiration in the cyanobacterium Synechocystis sp. PCC 6803 showed considerable rates of glycolate formation even when the cells were exposed to high levels of CO 2 such as 5% CO 2 in air [65] . Interestingly in diatoms, when T. weissflogii and T. pseudonana were exposed to 14 CO 2 for 5 sec, a considerable label (15%) was detected in glycolate in T. pseudonana but only 5% in T. weissflogii \n [5] . This may indicate a higher level of CO 2 in the vicinity of Rubisco in the case of T. weissflogii and consequently a reduced oxygenase activity. Recent studies on the expression of key genes in the C2 cycle in Thalassiosira sp. suggest the photorespiratory pathway is active in diatoms and plays a critical role in carbon and nitrogen metabolism in the cell [63] , [66] . We have identified most of the enzymes likely involved in a C2-type glycolate pathway in the genomes of P. tricornutum and T. pseudonana ( Fig. 2 ). The annotation of enzymes in the C2 pathway confirms several differences between the photorespiratory cycle in diatoms and in higher plants, and corroborates the scheme proposed by [67] . In algae, it has been suggested that two types of glycolate-oxidizing enzymes exist: a glycolate oxidase in Chrysophyceae, Eustigmophyceae, Raphidophyceae, Xanthophyceae and Rhodophyceae, and a glycolate dehydrogenase in Chlorophyceae, Prasinophyceae, Cryptophyceae and Bacillariophyceae [68] . Winkler and Stabenau [67] further suggest that in diatoms glyoxylate is synthesized via a glycolate dehydrogenase in both peroxisomes and mitochondria. In both P. tricornutum and T. pseudonana , two proteins similar to glycolate oxidases (GOX) or possibly glycolate dehydrogenases (GDH) were identified: one of the two proteins (22568) contains the peroxisomal targeting motif PTS1, which is common in many eukaryotes and characterized by the consensus sequence (S/C/A)(K/R/H)(L/M) located at the extreme carboxy-terminus [69] . The other protein (50804) appears to be targeted to the mitochondria. This suggests that at least one glycolate oxidizing enzyme in each diatom is localized in the peroxisome, although based on the previous biochemical studies of Suzuki et al. [68] and Winkler and Stabenau [67] , it is unclear whether it catalyzes production of hydrogen peroxide. Moreover, the potential for peroxisomal activity is corroborated by identification of a malate synthase (54478). In both diatoms, the enzymes for serine synthesis and metabolism have been found targeted to the mitochondria: serine-pyruvate/alanine-glyoxylate aminotransferase (SPT/AGT, 49601), glycine decarboxylase and serine hydroxymethyltransferase GDC/SHMT (56477, 22187, 32847, 18665, 17456), hydroxypyruvate reductase (56499) ( Fig. 2 ). Interestingly, we were not able to identify a gene for glycerate kinase in P. tricornutum or in T. pseudonana . This enzyme catalyzes the last reaction of the C2 cycle and appears to be present in cyanobacteria, the green algal lineage, the red algal lineage, but only sporadically in alveolates and heterokonts. The absence of this enzyme poses the question of how, or whether, glycerate can be transformed into 3-P-glycerate to be reintegrated into the Calvin cycle. An alternative to glycerate, and thereby 3-P-glycerate, as the endpoint of photorespiration is the possibility that all the glycine and serine produced from the fixation of oxygen are instead shunted to other pathways. For example, the formation of the antioxidant glutathione from photorespiratory glycine has been previously demonstrated (reviewed in [70] ). Another pathway for glyoxylate metabolism, the tartronate semialdehyde pathway, is known in cyanobacteria [70] . Synechocystis mutants were used to illustrate that a C2 pathway and glyoxylate/glycerate pathway (via glyoxylate carboligase and tartronic semialdehyde reductase) cooperate in the metabolism of 2-phosphoglycolate [65] . Genes encoding a putative tartronate semialdehyde reductase (45141) and a putative glyoxylate carboligase, also called tartronate semialdehyde synthase (56476), have been found in both diatoms. The closest BLAST matches to the models for tartronate semialdehyde reductase are genes encoding 3-hydroxyisobutyrate dehydrogenases. The two enzymes are part of the same enzyme family, making a definitive assignment difficult. The putative P. tricornumtum tartronate semialdehyde reductase (45141) has a mitochondrial targeting peptide while the targeting for the T. pseudonana model (2669) is unclear. The closest BLAST matches to the predicted glyoxylate carboligase were acetolactate synthase, however, these two enzymes are also closely related and difficult to distinguish. Both predicted carboligases appear to have chloroplast transit peptides, but the evidence is weak and therefore targeting of glyoxylate carboligase remains uncertain in both diatoms. The presence of glyoxylate metabolism is supported by an early study of Paul and Volcani [51] showing that the activity of glyoxylate carboligase in the diatom Cylindrotheca fusiformis is affected by light intensity. These data suggest that similar to cyanobacteria, diatoms combine C2 and glyoxylate/glycerate pathways to metabolize 2-phosphoglycolate back to the Calvin cycle. Reductive/oxidative pentose phosphate pathway Photosynthetic carbon fixation in plants and algae is performed by the Calvin cycle. Some Calvin cycle enzymes in land plants are of cyanobacterial origin, while others have been replaced by protobacterial or eubacterial enzymes [71] . Carbon fixation in land plant plastids is highly regulated, either by substrates and ions like Mg 2+ or by light-dependent redox regulation either at the transcriptional [72] , [73] or the enzymatic level via the ferredoxin/thioredoxin-system [74] , [75] . In addition to the Calvin cycle (reductive pentose phosphate pathway), plastids from land plants and green algae possess an oxidative pentose phosphate pathway (OPP). This ubiquitous process produces NADPH and pentose-phosphates for biosynthesis of nucleotides, amino acids and fatty acids in the dark by decarboxylation of glucose-6-phosphate. As both pathways in plastids are interconnected, operating them simultaneously would result in a futile cycle, using up energy in the form of ATP without net CO 2 fixation. Thus in plastids of land plants and green algae some of the enzymes of the Calvin cycle like the phosphoribulokinase (PRK), glyceraldehyde-3-phosphate dehydrogenase (GAP-DH), fructose-1,6-bisphosphatase (FBP), and seduheptulose-1,7-bisphosphatase (SBP) are activated in the light via reduction by thioredoxin (and become inactive in the dark), while the key enzyme of the OPP, the glucose-6-phosphate dehydrogenase (G6PDH) is active in the dark, but inhibited after reduction in the light. In contrast to higher plants, there is apparently no complete oxidative pentose phosphate pathway (OPP) in the plastids of several diatoms [9] , [10] as well as in P. tricornutum , suggesting diatom plastids in general lack this pathway. Two putative 6-phosphoglucono-lactonases might be targeted to the cytosol (31882) and to the plastid (38631), however, both genes are not yet supported by ESTs. The other two required enzymes glucose-6-phosphate dehydrogenase (G6PDH, 30040, and the G6PDH component of a G6PDH/6PGDH fusion protein, 54663) and 6-phosphogluconate dehydrogenase (6PGDH, 26934, and the 6PGDH component of the G6PDH/6PGDH fusion protein, 54663) were found to be cytosolic enzymes, indicating that the complete OPP is only functional in the cytosol ( Fig. 3 ). 10.1371/journal.pone.0001426.g003 Figure 3 Model of the oxidative and reductive pentose phosphate pathways and related reactions in P. tricornutum . For simplicity, the number of organelle membranes has been reduced in this figure. The superscript numbers attached to the enzyme names indicate the number of isoenzymes within the respective compartment. Enzyme abbreviations: AL: aldolase; FBA: fructose-1,6-bisphosphate aldolase; FBP: fructose-1,6-bisphosphatase; GAPDH: glyceraldehyde-3-phosphate dehydrogenase; GPI: Glucose-6-phosphate isomerase; GPDH: glucose-6-phosphate dehydrogenase; PGL: phosphor-gluconate lactonase; PRK: Phosphoribulokinase; RUBISCO: ribulose-1,5-bisphosphate carboxylase; PGDH: 6-phospho-gluconolactone dehydrogenase, PGK: phospho-glycerate kinase; RPI: ribose-5-phosphate isomerase; RPE: ribulose-phosphate epimerase; RPI: ribose-5-phosphate isomerase; TKL: transketolase; TAL: transaldolase; TPI: triose-phosphate isomerase; SBP: seduheptulose-1,7-bisphosphatase. The only Calvin cycle enzymes encoded on the plastid genome are the small and the large subunit of the Rubisco (GenBank AY819643) [76] . All other genes encoding primary enzymes of the Calvin cycle have been identified in the nuclear genomes of P. tricornutum and T. pseudonana (see Fig. 3 ). The only exception is the gene for the sedoheptulose bisphosphatase (SBP). The only SBP gene we identified (56467) encodes a protein that lacks a plastid targeting sequence and thus appears to be localized within the cytosol. This SBP, however, is not contained in the large set of ESTs from P. tricornutum , indicating that it is not actively transcribed under the applied conditions. SBP catalyses the reaction from sedoheptulose-1,7-bisphosphate to sedoheptulose-7-phosphate in the Calvin cycle; it is unclear yet whether the SBP reaction does not occur in diatom plastids or - more likely - whether this reaction is performed by one of the plastidic FBPs as shown for FBP I in cyanobacteria [77] . Interestingly there is a gene encoding a plastidic FBP (FBPC2, 42456) with a bipartite plastid targeting presequence that is located about 500 bases upstream of the SBP gene in the same orientation (similar as in T. pseudonana ). There is a theoretical possibility that both genes might be transcribed together and - after excision of a putative intron - might be translated as a fusion protein, thus the SBPase could be imported in a piggy-back manner, although this hypothesis has not yet been supported by transcript analyses (Weber and Kroth, unpublished). Genes encoding the Calvin cycle enzymes fructose-1,6-bisphosphate aldolase (FBA) and FBP are present in several copies. There are two class II aldolases (22993, bd825) and one class I (24113) aldolase in the plastids, while a class I (42447) and a class II (29014) aldolase are found in the cytosol. Four plastidic FBPases have been identified (FBPC1: 42886; FBPC2: 42456; FBPC3: 31451; FBPC4: 54279) and one cytosolic enzyme (23247). The redundancy of isoenzymes may partially reflect the evolution of diatoms by secondary endocytobiosis [15] , [75] . Some of the isogenes may have either a cyanobacterial or a rhodophytic origin or are related to respective enzymes from oomycetes. Other genes may also have been transferred by lateral gene transfer from bacteria or have been duplicated within the heterokonts [78] . Redox-regulation of enzymatic activity is critical for plastid functions. Thioredoxin is a small protein that is reversibly reduced in the light by ferredoxin/thioredoxin reductase (FTR) and is able to reduce target enzymes resulting in altered enzymatic activities [75] . Several genes encoding thioredoxins (Trx) were identified in P. tricornutum , including the genes for Trxs f (46280) and m (51357) both possessing typical plastid targeting signals. Three genes encoding Trx h proteins (48539, 56471, 48141/56521) were identified, one of which (48539 plus possibly 48141/56521) contains a presequence for targeting into ER/periplastidic space (respective homologues are also found in T. pseudonana ). This is surprising because Trx h is located in the cytosol in all other organisms examined so far. Genes encoding two plastidic Trxs y (33356, 43384), a mitochondrial Trx o (31720) and a ferredoxin-thioredoxin oxidoreductase (50907, needed for Trx reduction) were also identified. These results imply that thioredoxin based light-regulation is functional in diatom plastids, although far fewer plastid enzymes in diatoms than in plants may be actual Trx targets (see [8] ). Another group of proteins involved in redox-regulation in land plant plastids are glutaredoxins (Glrx), which are involved in fine-tuning of the thioredoxin system [79] . In P. tricornutum we predict two glutaredoxins to be targeted into the plastids (43497, 39133), one to the cytosol (16854), and one to the mitochondria (37615). Similar to the unusual periplastidal/ER associated Trxs h (48539/48141), one glutaredoxin (56497) also contains a presequence for targeting into ER/periplastidic space. Taken together, thioredoxins and glutaredoxins are present in the mitochondria, plastids and cytosol of P. tricornutum and T. pseudonana , although their functionality and specificity is unclear. The plastidic fructose-bisphosphatase (FBP) is the only enzyme in diatoms for which there is direct evidence of redox-regulation by thioredoxin [9] . The PRK also possesses the conserved cysteines for redox regulation, although due to a shift of the redox midpoint potential of this enzyme, it does not get oxidized in vivo and thus is permanently active [9] . Diatom plastids also possess a different GADPH enzyme compared to green algae and land plants, termed GapC1 (25308), which does not contain the respective cysteines [80] and which is not affected by oxidation or reduction (Michels, A. Wedel, N., and Kroth, P.G., unpublished). The chloroplast ATPase in land plants is modulated by thioredoxin by lowering the energy threshold of the membrane potential necessary to activate the enzyme. The sequence cassette on the γ subunit containing the necessary cysteines (AtpC, 20657) in land plants is missing in diatoms as well as in red algae. Other plastidic enzymes which are affected by thioredoxin in land plants are not found in P. tricornutum or T. pseudonana . (i) In land plants and in green algae there are two malate dehydrogenases, one of which is NAD-dependent and one of which is NADP-dependent. The NADP-dependent enzyme is redox-regulated via thioredoxin and serves as a valve for excess NADPH [81] . Based on enzymatic and in silico analyses, the redox-regulated isoenzyme appears to be missing from diatom plastids (Mertens and Kroth, unpublished). (ii) ADP-glucose pyrophosphorylase (AGPase) in land plant plastids produces ADP-glucose, the substrate for starch synthesis [82] . Diatoms do not possess a plastidic AGPase, which is consistent with the fact that they export all carbohydrates immediately from the plastids and store them as chrysolaminaran in cytosolic vacuoles. (iii) The Rubisco activase responsible for activation of Rubisco [83] , is apparently also missing from diatom plastids as no gene for this protein has been found in the genomes of P. tricornutum or T. pseudonana . (iv) Another system regulating the Calvin cycle in land plants is the formation of enzyme complexes of GAPDH and PRK by the small protein CP12 via disulfide bridges [84] . In land plants and in green algae these complexes form in the dark, and in the light they are reduced by thioredoxin in the presence of NADPH, dissociate and release GAPDH and PRK activity [85] . A comparison of native GAPDH and PRK enzymes from stromal extracts of diatoms and land plants by gel filtration revealed that diatoms do not form GAPDH/PRK/CP12 complexes (Michels, Wedel and Kroth, in preparation), accordingly we were not able to identify genes for putative CP12 proteins in diatom genomes. Interestingly, during our genome analysis we identified a few cases of unusual gene fusions. When transcribed as a single mRNA they may form fusion proteins consisting of two metabolic enzymes that are connected by spacers of 8 to 25 amino acids. We found three metabolic enzyme pairs that apparently are fused to each other because they are transcribed by a single mRNA: the mitochondrial triosephosphate-isomerase/glyceraldehyde-3-phosphate dehydrogenase (TIM-GAPC3, 25308) [80] , a cytosolic UDP-glucose-pyrophosporylase/phosphoglucomutase (UGP/PGM, 50444), and a cytosolic glucose-6-phosphate-dehydrogenase/6-phosphogluconate-dehydrogenase (G6PDH/6PGDH, 54663) fusion protein. The fact that each pair of enzymes catalyzes two subsequent metabolic reactions indicates that fusing these genes may result either in a better regulation or a faster conversion of substrates. However, there is evidence that at least some of these fusion proteins may be cleaved post-translationally [80] (Majeed and Kroth, unpublished). Interestingly, the TIM-GAPDH (present in T.pseudonana and P. tricornutum ) and the UGP/PGM are found in the genomes of the stramenopiles Phytophthora ramorum and Phytophthora sojae , while the G6PDH/6PGDH is not. Glycolysis Glycolysis is a universal cytosolic pathway for degradation of hexoses and results in pyruvate, which may be targeted to the mitochondria in eukaryotic organisms performing aerobic degradation or may be utilized in various other ways in organisms capable of living in anaerobic conditions. Several enzymes involved in glycolysis occur as a number of isoenzymes in P. tricornutum and T. pseudonana. For instance there are five genes for phosphoglucomutases (PGM) present in the P. tricornutum genome: two of the gene products (48819, 50718) are likely to be targeted to the plastid while the other isoenzymes apparently are located in the cytosol (51225, 50118 and the PGM component of a UDP-Glucose-Pyrophosphorylase/Phosphoglucomutase fusion protein 50444). Similarly there are three phosphoglycerate kinases predicted to be targeted either to the cytosol (51125), the mitochondria (48983) or the plastid (29157). Recent analyses using GAPDH genes from diatoms and other organisms indicate a common origin of all chromalveolates (86). Of the six identified GAPDH enzymes in P. tricornutum , two are targeted to the mitochondria (32747 and the GapC3 component of a TPI/GapC3 fusion protein 25308) and one is targeted to the plastids (22122) [80] . GapC2, assigned to be cytosolic [80] is present in two copies encoded in the same orientation on chromosome 16, with a distance of approx. 24 kilo base pairs (51128, 51129). A third cytosolic GAPDH enzyme was additionally identified (23598). Three genes for glucose-6-phosphate isomerases (GPI) were found, encoding a plastidic GPI (56512) and two cytosolic enzymes [87] with genes located next to each other in opposite direction (23924, 53878). We found only genes encoding a plastidic (56468) and two mitochondrial enolases (bd1572, and the apparently unfunctional bd1874) in P. tricornutum. However, in T. pseudonana a cytosolic and a mitochondrial enolase have been found (40771, 40391). This indicates that all reactions of the glycolysis may potentially occur within the plastid ( Fig. 4 ), where some of them simply represent essential enzymes of the Calvin cycle. Also surprising is the fact that there are isoenzymes of the complete second half of the glycolysis possessing mitochondrial presequences (see Fig. 4 ). In some cases the respective enzymes have been shown to be targeted into the mitochondria by fusing the presequences to GFP (C. Rio Bartulos, personal communication). Similarly the translocation of glycolytic reactions to other organelles has been described in unicellular green algae [88] . 10.1371/journal.pone.0001426.g004 Figure 4 Model of the glycolytic reactions in the cytosol and related pathways within mitochondria and plastids of P. tricornutum . Enzyme abbreviations: PGM: phosphoglucomutase; GPI: Glucose-6-phosphate isomerase; PFK: Phosphofructokinase; FBA: fructose-1,6-bisphosphate aldolase; GAPDH: glyceraldehyde-phosphate dehydrogenase; PGK phospho-glycerate kinase; PGAM: phosphor-glycerate mutase; PK: pyruvate kinase; PPDK: pyruvate-phosphate dikinase. No gene encoding a hexokinase for phosphorylation of glucose was detectable in either P. tricornutum or T. pseudonana . Instead, genes for glucokinases were detected in both species. This observation conforms to the trend that sugar-specific kinases are typical in prokaryotes and unicellular eukaryotes, whereas hexokinases with broader substrate specificities are typical in multicellular eukaryotes [89] . The P. tricornutum cytosolic glucokinase (48774) might additionally be involved in the chrysolaminaran pathways (see below). Storage products–synthesis and degradation Chrysolaminaran is the principal energy storage polysaccharide of diatoms. The relatively high contribution of chrysolaminaran to marine particulate matter underscores this molecule's significant role in the oceanic cycling of carbon [90] – [92] . It generally comprises between 10 and 20% of the total cellular carbon in exponentially growing diatoms but can accumulate to up to 80% of the total cellular carbon in cells whose growth is limited by nitrogen [93] . Chrysolaminaran concentrations undergo a diel rhythm characteristic of an assimilatory and respiratory product, accumulating during the daylight and becoming depleted in the dark [90] , [94] , [95] . The structure of chrysolaminaran is fundamentally based on a β-1,3-linked glucan backbone, which is infrequently branched with mainly β-1,6-linkages [96] – [102] . Vacuolar localization of chrysolaminaran in several diatom species, including P. tricornutum and T. pseudonana , was demonstrated by staining with aniline blue [103] and by immunolabeling with a monoclonal anti-1,3-β-D-glucan antibody [99] . The biochemical pathways leading to chrysolaminaran synthesis and degradation have not been elucidated. However, enzyme assays of cell-free extracts from the diatom Cyclotella cryptica demonstrated that the formation rate of UDP-glucose was ≥20-fold greater than for any other nucleoside-diphosphate-glucose and that UDP-glucose served as a substrate for chrysolaminaran synthesis [104] . Furthermore, exo-1,3-β-glucanase activity was detected in several planktonic diatoms and upregulation of this activity coincided with chrysolaminaran degradation in the diatom Skeletonema costatum \n [94] . We focused on exo- and endo-1,3-β-glucanases and β-glucosidases as the primary enzymes involved in digesting chrysolaminaran. We found four putative exo-1,3-β-glucanases in P. tricornutum , all belonging to the glycosyl hydrolase family 16 (49294, 56510, 56506, 49610). All orthologues possess an N-terminal signal peptide, except (49610), and all contain a C-terminal transmembrane helix. In addition, one (56510) possesses a putative C-terminal ER-retention signal (REEL). Of the four exo-1,3-β-glucanases, only one (49294) was represented in T. pseudonana (13556). Three putative endo-1,3-β-glucanases were identified in P. tricornutum , two belonging to glycosyl hydrolase family 16 (54681, 54973) and one to family 81 (46976). One of these (54681) consists of 1028 amino acid residues and has both an N-terminal signal peptide and a C-terminal transmembrane domain. The second (54973) has a signal peptide but no transmembrane helix and it is only about half the length. Curiously, the third enzyme (46976) apparently lacks a signal peptide but has an N-terminal transmembrane helix, making it a candidate for a type II transmembrane protein. Among the sequences, 54681 from P. tricornutum and 35711 from T. pseudonana are most similar to each other. In a ClustalW tree of endo-β-glucanases, these sequences clustered with bacterial endoglucanases ( Rhodothermos marinus , Bacillus circulans , and Sinorhizobium meliloti ), and (54973) only very weakly associated with these. The family-81 endoglucanases from the two diatom species grouped together but were apparently still relatively divergent, the next closest sequences being a pair of family-81 endoglucanases from Arabidopsis thaliana . Three putative β-glucosidases were identified in P. tricornutum , one belonging to glycosyl hydrolase family 1 (50351) and the other two (45128, 49793) to family 3. In T. pseudonana , only a single β-glucosidase was identified (28413), and this belonged to glycosyl hydrolase family 1. In addition to a signal anchor (50351), only one of the P. tricornutum family-3 β-glucosidases (45128) appears to have a C-terminal transmembrane helix. All three P. tricornutum orthologues were represented by ESTs, but to varying degrees. Overall, at least 10 enzymes predicted to digest 1,3-β-glucans were identified in P. tricornutum . Presumably, at least one of the exo-1,3-β-glucanases and one of the endo-1,3-β-glucanases act complementarily to digest the principle β-1,3-linkages of chrysolaminaran. The products of efficient digestion by this suite of enzymes would be primarily free glucose, with relatively small amounts of glucosyl oligosaccharides dominated by β-1,6-linkages (e.g. gentiobiose) derived from surviving chrysolaminaran branch points. A β-glucosidase could hydrolyze such oligosaccharides to free glucose. The free glucose generated from complete chrysolaminaran degradation would subsequently be phosphorylated by glucokinase. The vacuolar localization of chrysolaminaran implies that the degradative enzymes are also localized there. However, the exo-1,3-β-glucanase (56510) possesses a C-terminal ER-retention signal in addition to the signal peptide and C-terminal transmembrane helix. In yeast, transmembrane domains can serve as localization signals for sorting proteins from the ER, with the destination (plasma membrane or vacuole) dependent upon transmembrane helix length and composition rather than on a specified sequence [105] . We identified one gene for a glucokinase in P. tricornutum . As described above we conclude the enzyme to be involved in the cytosolic glycolysis (48774). Although there is no EST support, it is possible that by intron splicing the glucokinase may possess a signal peptide, which might allow targeting to the vacuole (compare to 56514). Interestingly, similar to the glucanases the enzyme possesses a C-terminal transmembrane helix, indicating that it might be integrated into membranes as shown for various hexokinases from plants [106] . In addition to a number of bacterial sequences, the most similar sequence to the diatom glucokinases is the glucokinase of Cyanidioschyzon merolae . This enzyme apparently also lacks a hexokinase and its glucokinase also contains a C-terminal transmembrane helix [107] . The simplest model is that the diatom glucan-digesting enzymes and the glucokinase are anchored at their C-termini to cytosolic membranes like the vacuolar membrane either being oriented towards the cytosol or to the vacuole. The localizations of the β-glucosidases are heterogeneous, however, and for any one of them to serve as a vacuolar gentiobiase would require localization by mechanisms other than those that localize the β-glucanases or the glucokinase. The hypothesis proposed here for degradation of chrysolaminaran has implications for the generation of glucosyl phosphate intermediates from an energy storage glucan. First, enzymes responsible for chrysolaminaran degradation apparently were recruited during evolution from enzymes normally associated with extracellular polysaccharides. Second, and as a consequence, degradative and phosphorylating steps are decoupled in diatoms. In organisms that metabolize starch or glycogen, the degradative and phosphorylating steps are achieved either concomitantly by an ATP-independent pathway or separately by an ATP-dependent pathway in which phosphorylation is catalyzed by hexokinase (for reviews, see [108] , [109] . The apparent occurrence of only glucokinase in both P. tricornutum and T. pseudonana may, apart from reflecting their evolutionary heritage, be an adaptation to a dedicated ATP-dependent pathway for chrysolaminaran digestion. Bacterial glucokinases, such as those of Escherichia coli , Zymomonas mobilis , Bacillus stearothermophilus , and Streptococcus mutans , have a high specificity and moderately high but relatively narrow K M range for glucose (K M = 0.22–0.61 mM; [99] – [102] ) compared with broad-specificity eukaryotic hexokinases (K M = 0.007–2.5 mM; [89] , [110] . In diatoms, such a glucokinase could cope with substantial fluxes in glucose concentrations and ensure that the phosphorylation is efficient as high concentrations of free glucose are liberated during chrysolaminaran degradation. The affinities and kinetics of the diatom glucokinases will need to be characterized, to assess the validity of this hypothesis. The synthetic pathway of chrysolaminaran is essentially unknown. Based on enzyme activity assays of C. cryptica , UDP-glucose likely serves as the substrate for chrysolaminaran synthesis [104] . The apparent absence of genes encoding ADP-glucose pyrophosphorylase in either diatom species provides further support that UDP-glucose serves as the substrate for chrysolaminaran synthesis. Interestingly, the UDP-glucosyl pyrophosphorylase (UGP) from C. cryptica was not inhibited by 3-P-glycerate or inorganic phosphate, suggesting that the assimilatory glucan is synthesized outside the plastid [104] . Surprisingly, a UDP-sugar pyrophosphorylase of the UGP family was encoded in the genome (23639) and this enzyme is predicted to be targeted to the chloroplast. The plastid localization of a potential UGP in both diatoms suggests that UDP-glucose is used for synthesis of chrysolaminaran within the CER en route to the vacuole. The origin of glucose-6-phosphate as substrate for a plastidal UGP remains unclear but is presumably derived from other sugar phosphates circulating in the plastid. A second candidate for UGP is one derived from a UGP/PGM fusion protein apparently localized in the cytosol (50444). This enzyme could supply UDP-glucose to a membrane-bound glucan synthase (see discussion below). One or probably more glycosyl transferases likely synthesize the chrysolaminaran polymer, and these could be either orthologous to 1,3-β-glucan synthases in other organisms or perhaps more likely, novel enzymes due to the unique structure and function of chrysolaminaran. A single gene encoding 1,3-β-glucan synthase was identified in P. tricornutum . The deduced protein consists of over 2,100 amino acids and possesses 20 transmembrane domains and a signal peptide. A single 1,3-β-glucan synthase was also identified in T. pseudonana , although no signal peptide was detected for the predicted protein likely because the N-terminus was incompletely defined. The two diatom sequences were most similar to each other and showed similarity to callose synthase sequences from dicots such as A. thaliana and Oryza sativa . This membrane-bound enzyme likely catalyzes the addition of glucosyl residues from cytosolic UDP-glucose on one side of the membrane to the growing, non-reducing terminus of the polysaccharide chain, which protrudes from the enzyme on the opposite side of the membrane. This presumed mode of action is akin to that of plasma membrane-bound polysaccharide synthases such as the Thalassiosira chitin synthases [10] , [111] and the cellulose synthases of terrestrial plants and multicellular algae [112] , [113] . Extracellular callose has been reported in diatoms and was suggested to serve as a permeable seal in the girdle regions during cell division [103] , so it is feasible that the identified 1,3-β-glucan synthase is a plasma-membrane-bound callose synthase. The relatively high EST support for this gene under a variety of growth limiting conditions, however, argues for a more active role not limited to cell division. Localization of the 1,3-β-glucan synthase to either the vacuole or the CER would help to determine where UDP-glucose is accessed from–either the cytosol or the CER. If UDP-glucose is accessed from the vacuole, this would support the hypothesis that chrysolaminaran metabolism evolved by relocating to the vacuole enzymes involved in the synthesis and processing of extracellular polysaccharides. Three additional gene models were identified in P. tricornutum (48300, 56509, 50238) and in T. pseudonana (3105, 4956, 9237) that encode proteins with moderate similarity (up to 28%) to fungal Skn1 and Kre6, enzymes required for synthesis of fungal wall 1,6-β-glucans [114] . The diatom proteins all contain N-terminal signal peptides and single C-terminal transmembrane helices, suggesting they are also associated with the suite of enzymes that process β-glucans. Although the precise function of the fungal enzymes is unclear, they resemble family-16 glycosyl hydrolases and have been interpreted as potential glycosylases and/or transglycosylases [115] . Branching in terrestrial plant starches and mammalian and fungal glycogen is achieved by specific enzymes that hydrolyze internal α-1,4-glycosidic bonds and transfer the released reducing ends to C-6 hydroxyls of the acceptor polysaccharide chain [108] , [109] . If analogous processes occur in diatoms, the putative diatom glucosylase/transglucosylases could act as chrysolaminaran branching/debranching enzymes. Inositol and Propanoate pathways Many different cyclitols occur in plants with the most widespread and extensively studied being myo-inositol [116] , [117] . Myo-inositol becomes incorporated into several crucial cellular compounds including those involved in signal transduction (phosphatidylinositol [PI], phosphatidylinositol-4,5-bisphosphate [PIPs]), hormone regulation (indole acetic acid [IAA] conjugates), membrane tethering (glycerophosphoinositide [GPI] anchors), stress tolerance (ononitol, pinitol), oligosaccharide synthesis (galactinol), and phosphorus storage (inositol hexakisphosphate [IP6]). Its primary breakdown product, D-glucuronic acid, is utilized for synthesis of various cell wall pectic non-cellulosic compounds, expanding the list of processes impacted by inositol synthesis and metabolism. In both P. tricornutum and T. pseudonana genes encoding enzymes predicted to be involved in inositol metabolism are well-represented, in particular the methylmalonate-semialdehyde dehydrogenase (acylating) (MMSDH), myo-inositol 2 dehydrogenase (InDH), and triosephosphate isomerase (TIM). \n De novo synthesis of inositol has been studied mainly in yeast and proceeds from glucose 6-phosphate through inositol 1-phosphate in two steps catalyzed by inositol phosphate synthase (INPS) and inositol monophosphatase (IMP). Genes encoding these two enzymes are present in the genomes of both P. tricornutum and T. pseudonana . Myo-inositol can be interconverted to scyllo-inosose by the enzyme myo-inositol dehydrogenase (InDH; EC 1.1.1.18). Stein et al. [118] confirmed the presence of InDH in the red alga Galdieria sulphuraria and a further study by Gross and Meyer [119] examined the distribution of InDH in algae by confirming its presence through enzyme assays. On the basis of InDH activity they assigned the different algae tested into two distinct groups: one composed of red algae and Glaucocystophyta and the other composed of heterokontophytes and haptophytes. They also proposed an inositol/inosose shuttle across the mitochondrial membrane as an alternative to the mitochondrial NADH dehydrogenase present in higher plants and green algae. The mitochondria of red algae seem to be capable of using myo-inositol for respiration, which was hypothesized to exemplify the divergence of basic metabolism during algal evolution. In plants, neither synthesis nor degradation involves InDH and there was no InDH activity present in the green algae tested by Gross and Meyer [119] . Interestingly, neither synthesis nor catabolism involves scyllo-inosose as a reaction product in the algae studied, whereas in mammals [120] it seems to be involved in the synthesis of scyllo-inositol. The sequences predicted to encode InDH from both P. tricornutum and T. pseudonana were compared on the basis of their predicted amino acid sequences to other InDH sequences including those from the red algae C. merolae and G. sulphuraria . The predicted protein sequences from P. tricornutum (51869) and T. pseudonana (8703) formed their own clade separate from the other sequences analysed. This would seem to provide further evidence for the theory of Gross and Meyer [119] for a divergence in algal metabolism based on InDH, but the other P. tricornutum sequence (34720) was found within the clade formed by the red algal sequences. MMSDH is an enzyme involved in valine catabolism rather than inositol metabolism. Here it catalyzes the irreversible NAD+- and CoA-dependent oxidative decarboxylation of methylmalonate semialdehyde to propionyl-CoA. It has been suggested that a Bacillus version of the protein is located in an operon and/or involved in myo-inositol catabolism, converting malonic semialdehyde to acetyl CoA and CO 2 \n [121] . Without further investigation its role in inositol metabolism in both P. tricornutum and T. pseudonana is unclear. There are elementary differences between green algae and diatoms Due to their evolutionary history diatoms naturally display a completely different host cell/mitochondria/plastid relationship compared to green algae and land plants. There are clear differences between diatoms and the green algae and higher plants in the structure of thylakoids, plastid envelope membranes, the mode of carbohydrate storage and the photosynthetic properties including photoprotection (for a detailed comparison see [8] , [122] ). The genome of the unicellular green alga Chlamydomonas reinhardtii , also representing a single-celled alga but originating from a primary endocytobiosis event, has recently been sequenced [123] and perhaps not unexpectedly, most Chlamydomonas proteins with a plastidic function display similarity to diatom sequences. However, among the 153 diatom sequences we have analysed ( table S1 ) only 3 of them showed the highest similarity to a Chlamydomonas protein whereas 23 displayed the greatest similarity to a higher plant sequence. Although the general pathways of carbohydrates are similar between diatoms and Chlamydomonas , several peculiar differences were identified that may have resulted from intracellular translocation of enzymes and/or pathways. Two distinctive features stand out: The mode of CO 2 concentration in diatoms is still largely unclear, as well as the post-translational regulation of photosynthetic products."
} | 17,012 |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.